Dreamforce Tableau Keynote detailed breakdown: Dreamforce 23

0:00Hey, it's Tim here. In this video, we get

0:02to do one of my favorite things breaking

0:04down keynotes.

0:04This time is the Dreamforce Tableau keynote

0:07. I'm going to be breaking it down. And this

0:09one's a

0:09bit different because I actually ended up

0:11being part of this keynote. Stick around to

0:13the end of

0:14the video to find out more. Let's get

0:16started. Okay, so you can go ahead and grab

0:18the video

0:19for the keynote, you can go ahead and watch

0:21it even at this URL, I'll put it up on

0:22screen,

0:23and I'll put it in the description as well.

0:25It's on Salesforce Plus, just sign up, get

0:27an account,

0:27it's entirely free, I use my Trailblazer

0:30account to do this. And you can go ahead

0:32and watch this

0:33on demand. It kind of gives you a summary

0:35of the keynote. But I think it's really

0:37nice that these

0:37are available. I wish more of the

0:39conference itself was available on Sales

0:41force Plus more

0:42generally, but we can talk about that some

0:44other day. And the context here is that

0:46Tableau's keynote

0:48is not obviously the main keynote, the

0:49Dreamforce keynote is the main keynote. But

0:52then each of the

0:52products have their own sort of time in the

0:55limelight, as it were. And so Tableau's

0:58keynote

0:58is especially focused on just Tableau. The

1:01other piece of context for this is to bear

1:03in mind where

1:04it is, it's at Dreamforce. So this is going

1:06to be Salesforce centric, a lot of the

1:08features here

1:09should lean on the Salesforce platform. And

1:11a lot of the things they demo and show will

1:12naturally

1:13sort of show how Tableau sits in the Sales

1:16force ecosystem. So if you're not a heavily

1:19sort of

1:20utilized Salesforce organization, or you

1:22don't use Salesforce with Tableau that much

1:25, some of this

1:26might sort of just feel a bit foreign to

1:27you. But I think it's still worth paying

1:29attention to,

1:30because as Tableau evolves, clearly a lot

1:32of its innovation is going to lean on the

1:34Salesforce

1:34platform. And there will be some benefits

1:36that I think will start to play out. Some

1:39of that got

1:39covered in this keynote. But anyway, that's

1:42enough pre talk, that's enough sort of pre

1:44amble, we're

1:45going to go ahead and watch this, I'm going

1:46to go into full screen, and we're going to

1:47just get

1:48started. And the point of this video, by

1:50the way, is to do a minute by minute

1:51breakdown. So I'll

1:52watch the whole thing, I talk through it,

1:54if that's not your kind of thing, I will do

1:55a shorter,

1:56much briefer 10 minute roundup of this just

1:59tells you what happened. But I like to

2:01analyze what's

2:02going on behind the scenes and some of the

2:03messaging. So that's what it will be in

2:05this

2:05video. And that's what you should expect.

2:07As ever, it's all timestamp below. So let's

2:09hop in.

2:10All right. Hello. Welcome, everyone. And

2:14come on in. If you're still coming in, in

2:16the back,

2:16I think we have roughly 2000 people in the

2:18room here. It is going to be an amazing day

2:21,

2:21I promise you that. And it's going to be

2:23amazing conference for you. Hopefully you

2:25had a chance

2:26to see the keynote right before this. I

2:28also thank you for coming over here because

2:30I know it's

2:30probably a bit busy to cross the street and

2:32go across what street is that fourth street

2:35. And so

2:36thank you for getting here. It's going to

2:38be a great day. But what I like to do next,

2:41and I

2:41realize I don't have my clicker, but that's

2:44a great move, is to thank you. And I want

2:46to say

2:47that it is very important to us not just to

2:49thank you for being here and for spending

2:52time with us,

2:52but to thank you for being with us on this

2:54journey. At Tableau, we've been very

2:57focused,

2:58and we'll talk about a lot of that today on

3:00customer success and innovation. And you

3:02are

3:03all part of that. If you are a customer or

3:06a partner or a community member, I know we

3:09have

3:09our community members here. I've seen a few

3:10of them in the front, which is awesome. Let

3:12's give

3:12them a round of applause. And of course,

3:15our amazing employees who've put so much

3:20work into

3:20getting to this point as well. So thank you

3:23. And we will move on. I want to talk a

3:25little bit about

3:26as well what we've been doing over the last

3:2812 months, really since we got here.

3:31So two things. When Tableau talk about 12

3:34months or the context for this,

3:38they're ultimately talking about the last

3:39Dreamforce. So that's what they mean by 12

3:41months, not conference. And then secondly,

3:44it's a common theme, actually, at least for

3:46the last

3:46three years, this sort of constant reminder

3:48, we see this slide every single year. And

3:50it's kind of

3:51an interesting one, because it's almost a

3:54bit of a self accountability spot. There's

3:56obviously a lot

3:56here that sort of reminds you of Tableau's

3:58heritage. If you didn't know Tableau is 20

4:00years old. That's sort of what this says

4:032003 all the way to 2023. And interestingly

4:06, on the

4:07on the Salesforce Tableau website, now,

4:10there is a little bit of context that shows

4:13you the pre

4:14Salesforce post Salesforce world, that's

4:17really 2021 onwards. And you can kind of

4:20see, well,

4:20at least their interpolation of this, they

4:22're showing sort of all the milestone points

4:25, some of

4:26those are features, some of those are

4:28milestones, some of those are just, you

4:30know, hard numbers

4:31that should relate to the Tableau product

4:33in itself. So it's always nice to see this.

4:35And then

4:35I like this sort of touch point, since 2022

4:38. It's an interesting one, because 2022 was

4:42was basically

4:42December. So what they're really saying in

4:45the last year, 160 plus features. Now, the

4:49breakdown

4:49of this feature is actually what matters

4:50the detail behind this is actually super

4:52important. I've done

4:53this at a user group before. And we've

4:54actually highlighted that some of these

4:56have just been

4:57enhancements. Some of these have just been,

4:59you know, quality of life improvements.

5:02And some of them have been net new features

5:04. So again, at some point, I just keep

5:06reeling off

5:07videos, I'll do at some point in the future

5:09, I just need more help. 160 features, I'd

5:12love to

5:12sort of break that down into the different

5:14groups and say, look, these were new

5:16features, these were

5:17enhancements, these were changes, and

5:19actually show that story a little bit more,

5:21there is a great

5:22vis that Jock McKinley has made that

5:24actually shows every single feature

5:26released in Tableau.

5:27And he himself has like a standardization.

5:30So they call it oil, water, and some other

5:32sort of

5:33categorization. And I think it's really,

5:35really good. I'll put a link to it in the

5:36description as

5:37well. And so you can check out that link,

5:39it will say Jock McKinley innovation charts

5:41or something

5:42like that. But yeah, I think this is

5:43interesting context. It's always

5:45interesting how to have it

5:46talk about their own successes as well.

5:48They kind of have to be proud about them

5:49and show them off.

5:51But at the same time as what's not on there

5:52, that's also what's interesting.

5:54You know, between now and 12 months ago, as

5:57I said, we've been very busy. We've been

5:59laser

6:00focused on delivering customer success and

6:03innovation. We've shipped over 160 new

6:06capabilities, things like embedding

6:08playground, things like data cloud for

6:10Tableau. You heard

6:12a little bit about Pulse today in the main

6:13keynote, you're going to see more today.

6:15Things

6:16like HIPAA compliance, important needs for

6:19our healthcare and life sciences customers,

6:21and on and

6:22on. And all of these things you'll be able

6:24to see more of here at the conference and

6:27of course,

6:27by speaking to some of our folks. So that's

6:30really exciting. Now, what's also exciting

6:33is to say that we just celebrated our 20th

6:35anniversary.

6:36It's a pretty, pretty incredible

6:41achievement. If I just pause there and

6:44think about it, like,

6:45what products weren't didn't exist 20 years

6:47ago, the iPhone didn't exist 20 years ago,

6:49and Twitter didn't exist 20 years ago, no

6:51longer exists. There's so many things that

6:55you will

6:56forget didn't exist 20 years ago, and Table

6:58au has been around that whole time. Now,

7:00obviously,

7:01you know, Tableau has not been this sort of

7:03blockbuster analytics tool for 20 years,

7:05it's probably been, let's say, a main

7:08player for half of that, maybe even a third

7:10of that,

7:11really strictly speaking. But it is now

7:13definitely an incumbent tool, it's very

7:16much got to that sort

7:17of same level of, you know, embeddedness or

7:21adoption as something like Power BI, or

7:24back

7:24in the day used to be micro strategy, it's

7:27now in that role of incumbent, and it has

7:29to sort of

7:29think about its own future. And I think

7:31that's where this point is going. Tableau

7:33and Salesforce,

7:36we have spent the last 20 years really

7:38revolutionizing analytics, and business

7:41intelligence. And we owe that all to you,

7:44everyone here in the room. Now, of course,

7:47we also became part of Salesforce roughly

7:50four years ago. And that has also been a

7:52journey that

7:53we are on. Lots of amazing things have

7:55happened since then, if you look in the

7:57keynote, some of

7:58the announcements that were made today with

8:00Data Cloud. And also, if you think about,

8:02you know,

8:03one of the things we're seeing in this new

8:05AI revolution is really this pull, the pull

8:08to do

8:08more than just focus on the data

8:11professional, the IT or analyst, right, the

8:14person who's really doing

8:16all that hard work. Customers are telling

8:18us to do more to go beyond. And so really,

8:21that is sort of

8:22defining our next chapter. Our next chapter

8:25is really all about everyone, all of you,

8:28and all

8:28of the people that you work with. Because

8:31it's not just having data with your

8:33analysts, it's also

8:35having data and being able to be data

8:37driven for everyone inside your company.

8:39And we'll talk about

8:40that. So again, at the Tableau keynote,

8:44this pitch was more subtle, it was a little

8:48softer.

8:49Here, I think it's just flat out more

8:51blatant. What Tableau is essentially saying

8:54here is look,

8:55for the last 20 years, we've been focused

8:57around the data analysts and enabling them.

8:59That's where our growth has come from. That

9:02's what has got us to where we are today.

9:04But that final mile, that final, you know,

9:06group of users in your business who aren't

9:08data analysts,

9:09the people who consume data information,

9:12for them, we need to push even harder. And

9:15so what Ryan

9:15literally says here is look, our past has

9:18been on data analyst, the next horizon,

9:20where we're heading,

9:21where we're going to the future is going to

9:24be for everyone. So in the previous keynote

9:27roundup,

9:27I literally said that look, Tableau's

9:29future is going to be focused around

9:31everyone else,

9:32consumers of data rather than just being

9:34about data analysts, which means you have

9:36to understand

9:37that the feature and the focus, it's not

9:39happening yet. There's still a big push on

9:42the core

9:43capabilities around dashboarding and

9:45analytics. But with Tableau paths, with

9:47Tableau GPT,

9:48these are features geared towards everyone

9:51in the organization's allowing them to ask

9:54their own

9:54questions, allowing them to build their own

9:56insights, allowing them to curate the kind

9:58of

9:58metrics they want to curate without having

10:00to have an analyst as not necessarily a

10:02gatekeeper,

10:02but as a person to go through, right? How

10:04can analysts be empowered to empower

10:06everyone in

10:07the business to basically self serve that

10:09term, I hate it, but that is essentially

10:12potentially the

10:12direction that's going here. So this is,

10:14again, another framing of this Tableau

10:16saying in lots of

10:17different ways and lots of different

10:18languages, just to make sure it's

10:20absolutely clear. And

10:21it's super important, because, you know, he

10:23then goes on to say, look, you know, we do

10:25appreciate

10:26the work the community has done, we

10:27appreciate where we've got to. But this

10:29this point here just

10:30keeps coming again and again. It's also

10:33having data and being able to be data

10:35driven for everyone

10:36inside your company. And we'll talk about

10:38that today. And of course, this only, as I

10:41mentioned

10:41before, accelerates, this need becomes more

10:45urgent in the world of the AI revolution

10:48that we're going

10:48through. Now, this is all exciting. And I

10:52am very lucky to be up here as the CEO of

10:54Tableau talking

10:55to you about where we're going, because it

10:57's exciting, right? Who wouldn't want to do

11:00this?

11:00What's not as exciting, the amount of data

11:02that's coming at us. There's so much data

11:06coming at us

11:06every day, billions upon billions of gig

11:08abytes of data. And this thing, that's like

11:12an interesting

11:12sort of tone change, isn't it like when if

11:15you're like a musician or creative, and you

11:18're trying to

11:19move the listener, if you're making a film,

11:21you're trying to move the audience through

11:23emotions.

11:24And that's kind of what's just happened.

11:25There's sort of all this opportunity, all

11:27this excitement

11:28around my table is going. But then there's

11:30a subtle shift in like the the tone. And

11:33what's not so

11:34exciting is the overwhelming amount of data

11:36that's coming across. And this, this is, I

11:39think he's

11:39tearing up a point here. I mean, I know he

11:41's taking up a point, I've watched the

11:42keynote, I was in it.

11:44But he's taking up a point here. And it's

11:46essentially this sort of point that I think

11:51we're

11:52all aware of. And we are not necessarily

11:54choosing to address head on, which is,

11:57there is always a

11:58growing amount of data, I think, if there's

12:01, if you look in, if you look on the

12:03internet, there's

12:04this, this, this statistic about how much

12:07data is created every year, versus previous

12:09years. And

12:10there have been a couple of inflection

12:13points where, in one or two years, we

12:15created more data

12:16than the previous entire history of

12:18humankind, right. And that's all to do with

12:20the capabilities

12:21around storage, compute, things like mobile

12:24phones, devices, the things they do, the

12:27scale and

12:27quantity of information they collect, but

12:30then also the opportunity that's available,

12:32because

12:33a lot of data is being captured all the

12:34time, but it's not necessarily being stored

12:36. So in a sort of

12:38push to help businesses understand this, I

12:41think this is an important sort of piece of

12:42context,

12:43there's a world of opportunity in data, but

12:45a lot of it sort of goes amiss. And that's

12:48because it's

12:48quite overwhelming. It's sort of a lot of

12:50it, there's a lot of out there, and it's

12:51hard to

12:52process. And it's hard to do it. Well, I

12:54work as a consultant, it's literally my job

12:57, helping people

12:58do better with how they harness and turn

13:01this data into meaningful outcomes, not

13:05just, you know,

13:06nice charts and beautiful visualizations,

13:08actual impact, actual bottom line,

13:10development,

13:12you know, improvement and quality of life,

13:15and not just customers, but employees as

13:17well, you have to

13:18sort of wrangle all this stuff. Data, it's

13:21collected, it's stored, it's processed, and

13:26it is

13:27consuming us. There's so much information

13:30that is out there. This data also continues

13:34to be,

13:34you know, in different places. And when I

13:37say that, I talk about this concept of it

13:40being

13:40disconnected. All of this data can be in a

13:43data warehouse, it could be on your phone,

13:47it could be

13:47on some type of local device, or maybe it's

13:49on premise. It really could be anywhere.

13:52And that

13:53is concerning to me as the CEO of Tableau,

13:55because we want to help you, of course,

13:57solve this problem.

13:59Now, I think it's also important to talk

14:01about, you know,

14:03It's super interesting that Snowflake and

14:06Databricks are the two databases mentioned

14:08here.

14:09I never think that's like a slip of hand. I

14:11don't think the person who's making this

14:13deck just went

14:14and got the just Google databases and Snow

14:16flake and Databricks came up first, and they

14:18just grabbed

14:18two icons. I think it's a deliberate choice

14:21. It might be to do with Tableau aligning

14:24itself with

14:25the databases it thinks are best suited to

14:27its own platform, Tableau Cloud. And, you

14:30know, these two

14:31platforms are kind of cloud native and Snow

14:33flake is, I don't know too much about Datab

14:36ricks, but

14:36I believe Databricks is also quite cloud

14:40native. And so it's a subtle but important

14:44sort of

14:44detail to make sure you don't skip. Data is

14:47so important, as you know, we're here to

14:50talk about it.

14:50I'm wearing a shirt. The shirt says like, I

14:53love data, I have shoes on it. I love data.

14:55A lot of us love data. And I keep asking

14:58the question of like, why do we love data

15:01so much?

15:02Why are we so consumed by all of this data?

15:04And we're consumed by this data. Well,

15:08Tableau has

15:08always been about data. We are data people.

15:10And I'll say again, thank you to our

15:12community to

15:13helping us get there. We are data people,

15:15but you can easily play like a data bingo.

15:20How many times

15:20can we head where to term data in this key?

15:23If you know the answer to that, if you're

15:26willing to watch

15:26this entire video, tell me the exact number

15:29in the comments below. I would love to know

15:31and I'll think

15:31of some sort of suitable reward of some

15:33kind. Let me know in the comments. Not

15:36everyone has been on

15:37this journey for 20 years. And so it's

15:39important to give you an example. I like to

15:41think of an

15:42example from my personal life and from our

15:44personal lives in general, because our

15:46personal lives, well,

15:47they run on data too. Now they run on data.

15:51How many of us have had an Apple Watch or

15:54glucose

15:54monitors, sleep ring, you know, maybe give

15:56me a show of hands that you've had some

15:58type of device

15:59that tracks your personal data? Well, that

16:01's great because your data tells you what

16:03you're doing,

16:04what's going on. Is it going up or is it

16:06going down? It allows you to make decisions

16:09about your life.

16:10And really in that case, your life, of

16:12course, whatever decisions you have, that

16:15ultimately

16:15determines your level of success. Now for

16:18me, I guess they convinced me to put my St

16:20rava data up

16:21here. I am a runner and I like to be out

16:23running around. That's why I'm taking all

16:26these steps, by

16:27the way, because I'm trying to get more

16:29activity. I am out doing these things

16:31because for me, I like

16:32to see what's happening. I collect data,

16:34everything from my steps to my elevation to

16:38my calories. And

16:39I look at this across many different apps,

16:41including Tableau, which that is Tableau up

16:43there.

16:43I just wanted to stop here because this

16:46made me think of a point. I think a decade

16:49ago,

16:50I was into this craze. I was even part of

16:51this movement called Quantified Self. I did

16:53lots of

16:54talks about this exact thing, collecting

16:56data, scraping it, running it through. I

16:58have

16:58visits about my music listening data. You

17:01'll hear more about that later. This is how

17:03I got my sort

17:04of starting data and is actually

17:06understanding my own data that led me to

17:08realize why businesses

17:09themselves were passionate about their own

17:11data and how you could start to sort of

17:13make sense of

17:14it. It kind of taught me a lot about wr

17:15angling data because I knew a lot about

17:17myself so I could

17:18immediately spot mistakes. What is super

17:20interesting that we live in a world today

17:22where what I was

17:23doing 10 years ago is just a standard

17:25feature of most devices now. Phones,

17:28watches, these all track

17:30step data. They just offer it to you even

17:32without asking. If you buy an iPhone as an

17:35example and you

17:36go into Apple Help, you might know this

17:38that even without an Apple Watch, it's

17:39tracking your step

17:40count. It's tracking a whole bunch of

17:42metrics and it's only if you choose to do

17:44something with it

17:44does it actually start to share that out to

17:47a place of choice. I think it's a super

17:49interesting

17:50thing and actually I've thought about this

17:52many many times. It's a really good way of

17:54contextualizing the challenge of data to

17:56everyday people because steps, heart rate,

18:00these are personal metrics, these are

18:02personal health metrics and I think health

18:04is a good way,

18:04it's a good analogy actually to make people

18:06understand the importance of doing

18:08something

18:08and understanding these simple metrics kind

18:11of leads you down this path where you start

18:13to think

18:13about how you wrangle it, how you work with

18:15it and it doesn't connect with everyone.

18:17Not everyone's

18:17super passionate about this data because

18:19there might not be an athlete, there might

18:21not be super,

18:21they might not have issues that they're

18:23tracking and therefore don't need to pay

18:24attention to these

18:25metrics in the first place but everyone can

18:27relate to a simple metric like the number

18:30of steps,

18:30the distance you walk, the quality of your

18:32sleep. These are things that we talk about

18:34day to day

18:35right that we can all connect with and so

18:38talking about the scale there and the scale

18:40in businesses

18:41and giving that as an analogy and using

18:44this to help essentially contextualize the

18:46challenge I

18:47think is a really good mechanism. This is

18:50how I run my life right, my goal is to run

18:52an ultra

18:53marathon, that's a goal for me, my version

18:55of success is actually to get there but I

18:58have to

18:58tune all of the data and all of the data

19:00that is happening and being pulled inside

19:04of these various

19:05applications helps me to actually

19:07understand what I need to do and what I

19:09need to tune.

19:10So this is the personal experience and that

19:13's great but what about companies because I

19:16think

19:16companies also need to be data driven, I'm

19:19sure you'd all agree. Now we believe at

19:22Tableau that

19:24data is the heartbeat of the modern

19:27organization and that's exciting. What's

19:31not exciting though

19:32is that we studied and interviewed and

19:34spoke to roughly 10,000 IT professionals

19:37like people

19:39online, like people here in the room and we

19:41found that actually less than 10% of them.

19:44So there's a little report here on the

19:46bottom left, Salesforce State of Data and

19:48Analytics Report.

19:49We'll have to have a look at that but it's

19:51an interesting, you've got to be careful

19:53with these

19:53surveys, not saying that data is inaccurate

19:56whatever but like if this is called the

19:59Salesforce State of Data and Analytics

20:02Report, if Salesforce queried people in the

20:06sector and

20:07the majority of them were Salesforce

20:09customers, it's more of a reflection on

20:11their customer base,

20:12right? It's not necessarily a reflection on

20:14the general industry but in the way that

20:16this is

20:17presented obviously no one's looking at

20:19that bottom left hand side so I think it's

20:20always

20:21useful just to go to that report and make

20:23sure that you look at it for yourself

20:25before you know

20:26really buying into these facts and you

20:28understand the context of what that survey

20:30was trying to do,

20:31the questions asked and you just make sure

20:33that you frame your understanding of these

20:36numbers

20:36with that context. I'm not saying these

20:39numbers aren't true and I'm not saying this

20:41context isn't

20:41true, I'm just saying make sure you frame

20:44that for yourself because those same

20:45numbers and those same

20:47bits of information might mean something

20:49different to you if you understand the kind

20:51of people

20:52Salesforce were surveying and the responses

20:54those people gave.

20:55[Dana] Are getting a full value of...

20:57[Alistair] Oh let's not go double speed.

20:59[Dana] That's concerning to me.

21:02[Alistair] We could try it double speed.

21:03[Dana] It's very concerning and it's really

21:05concerning in the world where we are being

21:06accelerated with AI, right? You must go

21:09faster, you must keep up. Now some good

21:12news in that survey.

21:13[Alistair] So yeah less than 10% of

21:17companies get full value from their data.

21:19Again I haven't

21:20read the report but I'm not surprised by

21:22the number. I know it's concerning but I'm

21:24not

21:24surprised by the number because if like for

21:27let's say the 10% who think they're getting

21:31the full

21:31value of the data, how do they know? Are

21:34they doing everything they possibly can

21:36with their

21:36data to gain insight? I don't think the

21:39answer is yes to that, right? So you know

21:42if you ask the

21:43opposite question and you sort of break

21:45down what is the question full value? Does

21:48full value mean

21:50that it's driving business aims and

21:51objectives? It's literally helping you

21:53increase income,

21:54revenue, ROI. Or is full value just meaning

21:57that you are getting a useful insight that

22:00's helping

22:00you make decisions but it doesn't mean your

22:02business necessarily is sort of turning a

22:03corner

22:04and doing better. So that term full value I

22:07think needs a bit of contextualization. 87%

22:10of leaders

22:11in contrast though see AI accelerates the

22:13data initiatives and that is concerning for

22:16me, right?

22:16Like this technology has basically been

22:18here for like half a minute and when I say

22:21half a minute

22:22that's strictly not true. AI has been in

22:24the technology sphere for quite some time

22:26but I

22:27think when you ask this question in 2023

22:29immediately after we've had things like

22:31chatgpt explode and

22:33then leaders are seeing the opportunity

22:36that's available, how do they know it's

22:39accelerated

22:40their data initiatives? I don't think in

22:42the time that chatgpt has been available

22:44that they could

22:46really claim that, right? I think it takes

22:48six months to really say hey this is what's

22:50possible.

22:50Unless what they mean is they're seeing

22:52projects that are now possible where they

22:54are, they're

22:54seeing opportunities that are now possible.

22:56I think that's an important context here.

22:58Is it that

22:59their data initiatives are being

23:01accelerated because there are small

23:03opportunities that

23:03are now more possible? Is it that AI is

23:05helping them crunch through data faster?

23:08Again there's

23:09all this detail and context, really good

23:10questions I think worth asking. Again we've

23:12got to go to the

23:13report and find out but I think it's useful

23:15just to have that in mind when you sort of

23:18see these

23:18numbers. It's actually quite easy during

23:19the keynotes to just look at these numbers

23:21and just

23:22buy into it because you're in the flow. You

23:24're seeing a keynote speaker talk and you're

23:26just

23:26buying into what they're saying. You're

23:27trusting what they're saying. Of course you

23:29trust the

23:29products, you trust the the CEO saying

23:32these things. But me here, I'm sat after

23:36the keynote

23:37in my room with a nice camera, plenty of

23:39time to think. I've watched this three or

23:41four times.

23:41I get a little bit more sort of, not

23:43skeptical, but I start to interrogate the

23:46facts a bit more

23:46and you can start to ask really good

23:48questions that might lead you to the actual

23:50answers

23:50that actually do validate these claims or

23:52actually make you think about them in your

23:54own context,

23:54which is super important. And that study,

23:58and it's actually called the state of data

23:59and analytics

24:00report that you can all get your hands on,

24:03is that 87% of leaders said they are now,

24:06these IT professionals, accelerating their

24:08investment. They are now investing more

24:11aggressively in data management and their

24:14data priorities to fuel their AI strategy.

24:17And that's exciting. But of course nothing

24:20's that easy, right? You can't just solve it

24:23by buying

24:23tools because the journey is complex. It's

24:27not easy to go through this data revolution

24:30and AI

24:31revolution because some of us here maybe

24:34are analysts. Some of us are. I love this

24:37by the way.

24:37When I saw this I was like, ah yes, this is

24:40, I love this, I absolutely love this. And

24:43let me

24:43explain why. What you're seeing is

24:45essentially three ways of framing a

24:47business. You do it by

24:49roles, by the departments, or by the

24:51industries they come from. And the reason I

24:53love this is

24:54as a consultant, this is basically what I

24:57do every time I turn up to a client. I'm

25:00like, okay,

25:00who am I working with? They're the data

25:02analysts. Okay, which department do they

25:03work in? Accounting.

25:04And in which sectors they work in?

25:06Education. Okay. And the combination of

25:08those three things

25:09literally configures my brain as to how I

25:12communicate with them. And for every single

25:14combination of role department in industry,

25:17there's a completely different way of

25:19handling

25:20the context, credentializing yourself, and

25:22working through problems, contextualizing

25:25problems,

25:25communicating with them. And it kind of

25:27presents a challenge to marketeers. Market

25:30eers are constantly

25:30working in these sort of buckets and

25:32demographics, right? But I've just never

25:35seen a company just

25:36boldly just flatter and say, hey, here are

25:37three ways of splitting our customer base

25:39and showing

25:39it to you as a way of sort of showcasing

25:41how they think about this problem. And

25:44essentially, what I

25:45think Tableau is saying is this journey is

25:47complex. And depending on who you are and

25:49how you look at

25:49yourself, role, department, industries,

25:51there's a different way of doing this. And

25:53I think if you

25:54were just to take all the items in this

25:56list, you probably come up with a whole

25:58array of different

25:59groups and combinations. And the industries

26:02one is I always think that's sort of the

26:04least important

26:04one because let's say you are an analyst,

26:07you work in accounting in education, that's

26:10not going to be

26:11too dissimilar to a, let's say, a team

26:14leader in human resources who works for a

26:17bank, let's say,

26:18okay, and you might think, well, what do

26:20you mean? Well, the context of what you're

26:22doing doesn't

26:23necessarily sit under your role or your

26:25department, it sometimes sits in the nature

26:28of your business

26:28department and exactly the activities they

26:31're doing. And so it's, you'll probably find

26:34some

26:34of these groupings sort of group up into

26:36the same buckets anyway, but I just really

26:38like the way

26:38this contextualizes everything. Now, little

26:41subtle hints. I love the details on these

26:44slides. I don't

26:45know, maybe it's just me, but I love this

26:47little thing on the corner, data, CRM,

26:49trust, AI. AI is

26:51like very gently faded out, you could just

26:54pick it out. And I always think like some

26:57designers

26:57sat there and thought about the different

26:59fades he was going to apply or she was

27:01going to apply

27:02to these rocks, right? And I just think

27:04that's fascinating. Data is the boldest one

27:07. AI is like

27:09faded out, CRM and trust. I don't know if

27:12these are sort of subconscious things that

27:14Tableau is

27:15sort of trying to pepper into the keynote,

27:17but I just find those really interesting.

27:19Anyway,

27:19let's carry on. >> But here, maybe

27:21executives. Some of us here, maybe

27:24individual contributors

27:25or operational folks. And so we want to be

27:28helping you in that regard. Now, you may be

27:30in different

27:30departments, HR, finance, et cetera, or you

27:33may be in different industries where there

27:35's regulatory

27:36compliance, data governance issues, data

27:38residency requirements. You may be in

27:41financial services,

27:42you may be in healthcare. It really is a

27:45complex world. And so the most important

27:48thing that I can

27:48say, and I really mean this, is that we are

27:52your guide within the Salesforce and Table

27:55au world.

27:56We are your guide. We are your trusted AI

27:59and data partner. And that's extremely

28:02important

28:03to communicate because we are going beyond.

28:06We are going beyond seeing and

28:08understanding your data.

28:09Because the journey, as I mentioned, is

28:12complicated. We will start by helping you

28:15connect

28:16and of course harmonize your data with, of

28:18course, Tableau data management and data

28:21prep,

28:21but also with amazing tools like Mealsoft.

28:24And also with tools and solutions like the

28:27Salesforce data cloud, which connect to

28:28third party applications and data

28:30warehouses, et cetera.

28:31Step one, if you will. And then going

28:34beyond that, we will continue to help you

28:36see and understand

28:37your data, whether you're using Tableau,

28:39whether you're using the Salesforce

28:41intelligent app suite,

28:42like revenue intelligence, or of course,

28:45right, any allowed there to go to the next

28:48step, which is

28:50taking action with your data. In my

28:52personal example, I explained how I'm

28:53taking action

28:54with my data. Well, you need to take action

28:57to be successful. And to be successful,

28:59right,

28:59you can use some of these new exciting

29:01things like Tableau Pulse, which you'll

29:03hear about today,

29:04or Slack, or third party applications. We

29:06want to help you across that whole journey.

29:09Now what's actually in my experience of

29:11roughly three and a half months in this

29:13role,

29:14one of the most exciting things that I get

29:16to do is work with our amazing community.

29:18And these folks

29:19are here to also help you. This is our

29:21unique differentiator, right? And yes, let

29:24's give them

29:24a round of applause again. Because our

29:27community is here to help you be successful

29:32. These are the

29:34individuals who've spent so much time. They

29:37are data people. They know how to make a

29:39data-driven

29:40culture work. They have made themselves

29:42successful. They have made their company

29:44successful,

29:44and they have made the world more aware of

29:47data. And this is incredible. And this is a

29:49really great

29:50tool and a very unique differentiator to

29:52Salesforce and Tableau. So I want to thank

29:55them again.

29:56And then the other thing I would say as we

29:58move forward here is we want to deep dive

30:00as we move

30:01into this next chapter, which is how do we

30:04go from our current environment of focusing

30:07on our beloved

30:08analyst, but also talk about, of course,

30:11the business user, maybe someone like

30:13myself, or the

30:14Salesforce CRM user. And so that I'm

30:17incredibly excited to bring up my really

30:20good friend

30:21and colleague who has nice shoes on,

30:23Francois Aginstadt.

30:24So I'm going to pause it right there. I'm

30:27also going to take a break. But business

30:29users,

30:30analysts, CRM users, this is a conceptual

30:32ization of what they mean by everyone. So

30:35the analysts

30:35are right there at the core. But then at

30:38least in the Salesforce context, CRM users

30:40are just people

30:41who are out there using Salesforce. And

30:43then the business users are people who sit

30:45in the business.

30:46You could almost see these two is basically

30:48the same thing. But they're just contextual

30:49izing it

30:50this way. And then Francois is a chief

30:51customer officer, he recently changed from

30:54being a chief

30:54product officer had been for quite some

30:56time, the chief customer officer. And so he

30:58starts to go

30:59through some features. Anyway, I'm going to

31:00take a break, I'm going to watch some

31:02Formula One,

31:02I'm going to come back and finish recording

31:05this video. Okay, we're back. I watched

31:08Formula One,

31:09what an ending for context signs won the

31:11race, if you don't know which race I'm

31:14talking about,

31:14he's won two races. So that should tell you

31:17which race. Okay, let's carry on with this

31:19particular

31:20keynote. All right. No, I got my thanks,

31:25Ryan, I gotta say your shoes. Pretty

31:29stylish.

31:30Got good taste. Well, hello everybody. It

31:33is so great to be here with all of you to

31:35talk about

31:35the latest innovations coming to Tableau

31:38and how you'll be able to bring data to

31:40even more people.

31:41Because this is the era of data. We know

31:45that companies that use data are just more

31:49successful

31:49companies than those that don't. And we

31:52know that employees who are empowered with

31:55data

31:55are just more satisfied employees, they're

31:58able to get their jobs done better. So the

32:00opportunity

32:01is truly to bring data to everyone and make

32:04every company a data driven company and

32:07every person

32:08a data. So if you ever watching these key

32:11notes, a bit of behind the scenes, you can

32:14see that two

32:15people by this log. And one of them is

32:18April, I know her because she helps chore

32:23ograph all these

32:24wonderful demos that we see here on stage.

32:27That is one of her roles in Tableau. And

32:29then we

32:30think we have a product manager, we're

32:32going to get an introduction. I think we

32:34have a product

32:34manager who's about to walk through the

32:36demos. And you can see a whole bank of

32:37laptops that are

32:39set up ready to go. I'm sure they're all

32:41backups of backups. But yeah, that's

32:43basically what's

32:44about to happen. So if you ever watching a

32:46keynote and you see two people get up to go

32:48to this log,

32:49now there's about to be a demo, which is

32:50probably what's going to come next person.

32:53And this is what we're going to talk about

32:54in this session. We're going to talk about

32:57how we're

32:57bringing data to everyone, empowering every

33:00single user, whether you're a business user

33:03or an analyst

33:04or a CRM user, everybody should be able to

33:06use data to make better decisions faster.

33:09And this

33:10is key because our mission is to help

33:12people see and understand data. And that

33:14truly does mean

33:15all people. And you'll see in this keynote

33:18how we're going to be bringing new

33:20experiences for

33:21our users with new metrics and insights

33:24that makes data easy, approachable, and

33:26contextual.

33:28How we're going to make everyone be able to

33:30explore data in an easier way than ever

33:33before.

33:34And how we're going to deliver analytics in

33:36the flow of work so you have smart

33:38applications

33:39that are actionable. This is what this

33:41keynote is all about. And of course, we're

33:45going to sprinkle

33:46a lot of AI right at the heart of the

33:49keynote because the AI is an opportunity.

33:52[JT] So this is an interesting slide. I

33:54feel like there's some firm dates in here.

33:57So Tableau Pulse

33:58general availability December 23, which

34:04means 23.4 will come in December 23, which

34:08also means the gap

34:10between 23.3 and 23.4 is going to be tiny.

34:14I wonder if they're just going to sneak it

34:17in right before

34:18the end of the year, right? Like just after

34:21Christmas. Not before Christmas, because

34:24that

34:24would essentially mean if 23.3 comes out,

34:27let's say at some point this month, let's

34:30say in the

34:31next three weeks, is my guess. Actually, no

34:34, let's say 23.3 comes out the beginning of

34:37October,

34:37right at the beginning of October. Then you

34:39've got October, November, end of December

34:41would kind

34:42of work. And it doesn't have to be a big

34:43release. It'd be a quiet release for 23.4.

34:46But that is

34:46interesting. It's also the first time we

34:49get to see Tableau Pulse in public. I feel

34:51like what's

34:51happening right now is that Tableau Pulse

34:53is being rolled out sort of behind the

34:55scenes. But 23.4

34:57might be when we then get Tableau Pulse

35:00going out. Einstein Co-Pilot for Tableau.

35:04Now, this

35:04feels like something new. I think this is

35:07an official name. At conference, we

35:09previously

35:10saw something called Tableau GPT. And that

35:12was a pretty interesting set of

35:14capabilities. It feels

35:17like Einstein Co-Pilot is the sort of brand

35:19name that sits on top of that technology.

35:21So Tableau

35:22GPT is the underlying technology. The

35:25product that uses that is Einstein Co-Pilot

35:28. Co-Pilot is

35:29an interesting brand name because of course

35:30that's been made famous by Microsoft,

35:32specifically GitHub

35:33Co-Pilot. And then Microsoft deployed that

35:35more widely across its platform to then

35:37make Microsoft

35:38Co-Pilot and Excel Co-Pilot and all of

35:40these wonderful co-pilots. Einstein Co-P

35:42ilot feels

35:43like Salesforce to take across the whole

35:45entire platform. But specifically in Table

35:48au, it's

35:48obviously going to help you do certain

35:50things. What is not clear is whether it's

35:52going to need you to

35:52have a Tableau Cloud or specific

35:54capabilities around Salesforce. I assume it

35:57will need

35:58something like Tableau Cloud for it to work

36:00because for you to be able to deploy that

36:03kind

36:03of model capability in your own Tableau

36:05server would be kind of tricky. Or if you

36:07do get to

36:08deploy it on your own Tableau server, they

36:11might ask you to open up specific ports and

36:13specific

36:14IP addresses to Co-Pilot so it can process

36:16all the requests, but then use some of that

36:19magic with your

36:20on-premise tech capabilities to start to

36:22understand what's going on. But there's no

36:24way Co-Pilot is

36:25working without understanding what's going

36:27on on your infrastructure somewhere in the

36:29cloud. That's

36:29almost a dead certain thing. If they are

36:31able to deploy that kind of thing onto

36:33server without that

36:34capability, then wow, that's a huge sort of

36:37Apple-like approach to data and privacy,

36:40running all the modeling on your own

36:42servers. But again, as Apple does, you need

36:45incredible

36:46integration between hardware and software

36:49to be able to get the optimizations that

36:51make that sort

36:52of make sense. Anyway, intelligent apps,

36:55general availability now. Be interested to

36:58see what that

36:58is. I don't know what that is. I might be

37:01missing something, but let's wait and see

37:04what that is.

37:0533% faster time to insight. That is an

37:08interesting metric. There's no context for

37:10it. So let's hope

37:11we hope we get that. And there's some logo

37:13for some companies. So you've got source,

37:16financial

37:16year, FY24, Salesforce, customer success

37:19metrics. So I don't know if these customer

37:22success metrics

37:23that 33% faster time to insight relates to

37:26these three companies sort of speaking to

37:30that number.

37:30So if it is 33% faster time to insight for

37:33these three companies, that is what the

37:36star is. And I

37:36think that's what that source is pointing

37:39to. Maybe there's somewhere at the end with

37:41an appendix,

37:41who knows, but unnecessarily detailed

37:44amounts of tearing down this slide. Let's

37:46move on.

37:47There is no AI without data. And this

37:51community is all about data. So we are at

37:54the heart of the AI

37:56revolution. Right. And that's true. I bring

37:59AI into our, I'll stop it here. I think the

38:01Nvidia CEO

38:03said this himself, when they announced

38:04their partnership with Snowflake, which I

38:06covered

38:06separately. He said, for AI to work, you've

38:10got to bring the most capable AI platform

38:13to the world's

38:14best data platform, which is that time on

38:16Snowflake. And I think Tableau is right to

38:18contextualize that here. Tableau is a

38:20product that sits in the heart of data,

38:22naturally means

38:23that Tableau has to have a pivotal role in

38:26some sort of AI capabilities. What is

38:28interesting is

38:29that what that AI does is really the

38:31fundamental question. Nvidia can generally

38:34just talk about its

38:35hardware that enables AI capabilities, a

38:38sort of bare metal kind of innovation,

38:41whereas Tableau

38:41really have to come at it from a SaaS

38:43perspective. So software as a service is S

38:45aaS. So therefore

38:46their innovation here has to be software

38:48led. So they have to come up with products

38:49that sit on top

38:50of that capability and see how that gets

38:52deployed. So that's why I'm here.

38:55- Products, it'll make the product

38:57experiences easier to use because we're

39:00going to make the

39:00complex simple. It's going to enable us to

39:03bring new experiences that broaden the

39:06reach

39:06of analytics for more people. And

39:08ultimately it'll enable everyone to be

39:10successful with data.

39:12This is really the opportunity. And you're

39:14going to see it in the new Tableau Pulse.

39:16You're going to see it in the new Einstein

39:19Co-Pilot for Tableau. And you're going to

39:21see

39:21it across every single one of our

39:23applications. This is the beginning of the

39:26new generation of

39:28Tableau and the new opportunity for

39:30everybody to become data powered.

39:32- And just to call it out there again,

39:35Francois saying it himself, it's the

39:37beginning of the new

39:38generation for Tableau, right? They kind of

39:40bookended the chapter and I think Sales

39:42force

39:42acquired Tableau maybe at a good

39:44opportunity because I don't think the

39:45chapter was clear

39:46that it was going to be sort of bookended

39:48like that. Maybe it was and we didn't know

39:50it and we

39:51didn't see it coming. Salesforce did. But

39:53nonetheless, at least for Tableau, it feels

39:56like they are realizing this AI revolution

39:58is going to fundamentally change the way

40:00they do

40:00their work. And in some way, I agree. I

40:02agree for lots of reasons. I'll come to

40:04that later.

40:05Anyway, let's carry on. - And of course,

40:07whenever we talk

40:08about data and AI, we have to talk about

40:11trust. It's important that the AI is

40:14trusted and ethical,

40:15that your data stays your data and is not

40:18used to train other applications. And that

40:21we're

40:22transparent about how we use your data and

40:24how AI is done. - It's actually super

40:27important that

40:28you have this slide. I think Tableau have

40:31to basically lead every discussion with AI

40:33about

40:33this. So number one, your data is not a

40:35product. You control access. We prioritize

40:39accurate,

40:39verifiable results. That is the boldest

40:42claim on this one. I'm not saying that it

40:44can't be done.

40:45I'm just saying that's easy to hold them to

40:47account to. You can literally just simply

40:49take the AI

40:50output and say, is this accurate? Go and

40:52test it and find out the answer. The others

40:54are based on

40:56trust. You have to take Tableau's

40:57reputation and they have to come up with

40:59white papers and

40:59evidence for that. That's quite easy for

41:01them to do. And I believe them because

41:03Tableau has a

41:04history of meeting tons of really difficult

41:07compliance standards. Our product policies

41:10protect human rights. That one is open to

41:14debate because as these things go, policies

41:17are always

41:18open to interpretation and the policies

41:20change over time as well. So it's kind of

41:22like a moving

41:23goalpost. So you can say that now, but then

41:25something might change in the future. The

41:27standards

41:27might get lifted. The standard might get

41:29lower and you can either choose to uphold

41:31higher standards

41:32or just keep the bare minimum or drop the

41:34standards. So there's always sort of three

41:36routes

41:36you can take in this field. We advance

41:39responsible AI globally. That is also an

41:42interesting claim.

41:44What does advancement mean? Is that sort of

41:46showing product leadership, product

41:47innovation,

41:48showing the right way to do something,

41:50contributing to projects that support this

41:52kind of thinking.

41:53And then transparency builds trust. Yes,

41:55actually being transparent about this path

41:58and how it's

41:59being built is super important. Of course,

42:01it can't be fully transparent because this

42:03is a

42:04product. And if they tell you how they're

42:06doing everything, then the competitors can

42:08just go ahead

42:08and copy it. And we work in a world and

42:11space where ultimately all these platforms

42:14are SaaS

42:14platforms, which means they sit on top of

42:17things like AWS and therefore the core

42:19underlying

42:20capability can be leveraged by other people

42:22. Once you've seen how something is done, it

42:24's very easy

42:25for you to find several ways of achieving

42:26the same thing, even if you don't

42:28necessarily have the same

42:29front-end or back-end platform. So super

42:32interesting breakdown there. And this is

42:35key and it is core to

42:36the Einstein platform. All right. As I

42:40mentioned, the goal is to reach more people

42:44with data.

42:45I love it. I love this slide. Then dash

42:48boards and charts, gauges. I can't even

42:50build a gauge

42:51in tablet easily. So I don't know why that

42:53's there. Nonetheless, now none of the dash

42:56boards,

42:56now none of the charts. Here we're going to

42:58give you metrics. So this is super

43:00interesting, isn't

43:01it? Reaching more people with AI. This is

43:03what AI is going to let us do. Search and

43:06metrics. They're

43:07literally spelling it out. This is how you

43:08're going to be, you know, the product is

43:10going to change.

43:11I think this could be the defining sort of

43:13direction of Tableau. The search box, the

43:16Omni

43:16search is going to become a core part of

43:18the product. And in a way, I've said this

43:19myself,

43:20when I was playing around with chat GPT, I

43:22said, you can't fit everything in this box,

43:24you might as well give up. And it kind of

43:26feels like Tableau semi agrees with some of

43:28that, but

43:28also understands that its core interface is

43:31also going to have to change to make it

43:33more accessible.

43:34And using AI along the way. Now today, the

43:40way most people consume data is by getting

43:42dashboards.

43:43Everybody getting dashboards today? Do you

43:45want more dashboards? Not really. You want

43:50more

43:50insights. You want more action. You want

43:52just kill the life of hundreds of analysts

43:56out there building

43:56dashboards. Yeah, you're building dash

43:58boards. You feel like you're killing the

44:00dream. You're, you're,

44:01you feel like you're delivering and your

44:03end users are sat here at conference going,

44:06I hate these

44:06things. I don't want any more of them. Love

44:08it. I love the fact that someone in the

44:10background

44:10shake, shook their head without even the

44:12prompt. He asked the question, they're

44:14immediately going

44:15like that. Love it. Love it. Love it. For

44:18empowerment. Now the opportunity is to

44:24simplify

44:24the experiences and make data as easy as a

44:27Google search, make data as approachable as

44:31every single

44:31widget you have on your phone. If your

44:34business data gets used, that creates

44:37opportunity. And we

44:39want everybody to use that data. And this

44:41is why today we're really excited to share

44:43with you our

44:44newest application Tableau Pulse. Tableau

44:47Pulse is a whole new experience for data

44:50that's powered by

44:51generative AI. It's personalized for every

44:54single user, contextual to the task at hand

44:57. And it is

44:58smart. It enables you to pull out insights

45:01from your data automatically. I'm going to

45:04pause this

45:04here and just say I've actually done a

45:06video breaking down the Tableau Pulse

45:08announcement

45:08from Tableau Conference. It goes into it in

45:11pretty much the same detail. If there's

45:13anything new in

45:14here, I will pull it out now. But I think

45:16this is going to be the same demo as we saw

45:18back then. So

45:19let's have a look. So you guys want to see

45:21it? Yeah. All right. Homer, are you ready

45:25to show it?

45:26Absolutely. All right. Please welcome Homer

45:28Wang.

45:28All right. Thank you Francois. And hello

45:34Dreamforce. I'm Homer, a product manager

45:38here building Tableau Pulse and Tableau AI

45:40for all of you. Now I hope you're just as

45:44thrilled as I am.

45:45So this is an interesting starting point.

45:48Number one, we're in Slack. This is what

45:51this interface

45:52looks like. I'm actually a customer of

45:54Slack. And this looks like a new interface.

45:56It looks like an

45:57updated version of Slack that is, I think,

45:59trying to replicate Teams. I think they

46:01announced some

46:02sort of change that makes Slack look more

46:04like Teams for the companies that kind of

46:06really love

46:07Teams but don't want to let go of it and

46:09want something that looks like that in

46:11order to switch.

46:13So really nice. Obviously, I think it's

46:15quite a polished setup. So what we have

46:18here is a Tableau

46:18app inside of Slack. And it's obviously the

46:21way Slack works with Tableau is... My watch

46:24is coming

46:25out. The way Slack works is that there's a

46:27Tableau app which creates a one-to-one chat

46:30with you as an

46:30app. So that's what you can see here on the

46:33left-hand side. And in essence, in there is

46:35where it delivers you personalized messages

46:38and notifications. It doesn't yet feed

46:41those in

46:42entirely into the channels setup. I think

46:45it can post alerts into channels, but

46:47specifically,

46:48these sorts of insights come to you via the

46:50app. To show you how we're reimagining the

46:54way

46:54anyone interacts with data. So demo hat's

46:58on and let's dive right in.

47:00Like many of you, I start my day in Slack.

47:04With Pulse, I can get a personalized digest

47:08on the key metrics that I follow and matter

47:11most to me. Now, AI looks at what's

47:14happening across

47:14the board and delivers a crisp summary

47:16upfront so that I know what to focus on in

47:18just two seconds.

47:19And here, I see an unusual uptick in... So

47:23let's just interrogate what this is doing.

47:27So

47:27it's interesting because there's no visual

47:29element, but we do get this sort of call-

47:33out

47:33of this unit indicator that we saw earlier

47:36on. So device sales, 1,675 units. You get

47:39this prompt

47:40that says, is it helpful or not? And at the

47:42very top, we've got this Tableau Pulse

47:44Digest for

47:45September 12th, 2023. So this is something

47:48new. A Pulse Digest suggests that it's

47:51going to be

47:52sending you a daily update and it's a

47:53digest of the things you care about. So

47:56device sales,

47:57campaign ROI, and regional revenue seem

48:00like three separate metrics, hence they're

48:02bold.

48:03So device sales are seeing an unusual spike

48:05since the beginning of this week,

48:07while quarterly regional revenue and

48:09monthly campaign ROI are steadily

48:10increasing. So these

48:12are three metrics and it's basically

48:13telling you what's going on. Additionally,

48:16seven of your 12

48:17other metrics have changed, four favorably

48:20and three unfavorably. So that's a pretty

48:22nice

48:23summary. What is interesting is there's no

48:25visual element here. So it would be nice to

48:28be able to

48:28see those metrics like as a scorecard here,

48:32like you show me all 12, show me the four

48:35favorable,

48:36the three unfavorable at the top, and then

48:39talk about the other seven that are just

48:43doing nothing

48:43basically. So I think that would be kind of

48:47nice and maybe this is where the product is

48:50heading,

48:50but immediately here, well, maybe that's

48:53what this... Actually, no, that is what

48:56this

48:57breakdown is below. The line below it makes

48:59it feel like it's something separate,

49:01but actually it's one thing. So then we do

49:03get a breakdown. Again, nothing visual,

49:05which is a shame.

49:06You get a text breakdown. So device sales,

49:08that's what we've got there. Regional

49:10revenue

49:11got there as well. And so let's see what

49:13else is in this. Let's keep carrying on.

49:15Device sales. Well, as a regional sales

49:18manager, this is great news,

49:20but I do need to understand a little more

49:22than that. So let's click in.

49:24Cool. So clicked on Slack and it took them

49:32straight to a page in Tableau Pass. So I'm

49:34sure it won't be that smooth. It kind of

49:36animated in between the two. That's

49:38definitely something

49:38like a Figma interaction, but ultimately

49:42you go to Tableau Pass. This Tableau Pass

49:45interface feels

49:45almost entirely new. It doesn't feel nice

49:48to Tableau. Super interesting. There's a

49:50whole

49:50social element to this. There's a whole

49:52profile element to this. If this is the new

49:54interface

49:54coming to Tableau Cloud, that's going to be

49:56super interesting. But you do get this lead

50:00in with this

50:00sort of visual element that I was talking

50:02about. So you can actually see the general

50:04trend, the kind

50:05of general area that your data performs in,

50:08and you kind of see a breakdown at the top.

50:10If you

50:11use something like Alteryx Auto Insights,

50:13this is very similar, but of course Tableau

50:16definitely has

50:17an upper hand here, is that they've got the

50:20entire Tableau server and charting

50:22capability to boot as

50:24well. So with something like Alteryx Auto

50:27Insights, what you have to do is

50:30essentially push the data

50:33into Auto Insights from the output of an Al

50:35teryx flow, or you can feed it specific data

50:38sources to

50:39look at. But even then, this just feels a

50:42little bit more sort of smoother. The

50:45experience feels

50:46smoother, even though it might actually be

50:47doing the same thing, but just not as

50:49advanced. Anyway,

50:50let's keep having a look. You can clearly

50:52see this latest anomaly picked up by Table

50:55au and visually

50:55explained to me. I can also export metadata

50:58on the metric that our endless friends

51:00helped define

51:01so that I can trust what I'm seeing. And if

51:04I want, I can... I think the fact that

51:07metadata

51:07is buried there is actually a bad thing,

51:09right? The device sells no one clicks on

51:11that eye

51:12indicator. How often do you see indicators

51:14like that? You just blow right past them.

51:16If anything,

51:17if it's like any sort of social or digital

51:19feed, what you do is you look at the things

51:21that are

51:21calling out at you. So the green and the

51:24big numbers, that's where you go to. If you

51:27told me,

51:27where do I find the information about this

51:30data source? You'd look around a little bit

51:33and you'd

51:34spot that eye and then you'd click on it,

51:35but it's not natural. I kind of feel like

51:37it should just

51:38sit up there at the top, device sales,

51:40published by, owned by, and the kind of

51:43general metrics,

51:44and then like show data metadata, like just

51:47call out that you are going to see the

51:50metadata for

51:50this data set rather than this very sort of

51:52clean interface, which feels like a UX win,

51:54but it's not a very useful... Like in

51:57business, you need context up front. You

51:59don't have to

51:59click in to see context. That makes it two

52:02activities that are unnecessary, one to

52:04click in,

52:04one to click out. So yeah, nice interface,

52:08but I just think, just put that metadata

52:11right up front.

52:11Tableau Cloud, Tableau Server does that

52:13already with certified data sources. You

52:15can just go and

52:16see some of that metadata immediately. So

52:18we'd love to see that here. Don't do this

52:20view to my

52:21own liking while respecting my security

52:23context. Okay. Now that I know what I'm

52:26looking at,

52:28what about the why? Well, guided questions

52:32just by AI help me phrase what I want to

52:34ask,

52:35but don't necessarily know how. And here I

52:38'm interested to know which products drove

52:40the

52:40sudden increase. And with one click, I get

52:44a plain natural language insight

52:47accompanied

52:47by visualization, all coming from Tableau.

52:50There was a subtle sort of broken user

52:57interface there.

52:58And when you clicked, the activity happened

53:02below. So if we just go back a few seconds,

53:06this is like ridiculously detailed product

53:09feedback, but hey, this is the keynote

53:11breakdown.

53:11We can do this here. So here you can see

53:13the window, the fold as it were. So

53:15everything in

53:16the window is as is, but then if you click

53:17which products drove this sudden increase,

53:19essentially the activity happens out of the

53:21fold, right? It kind of feels like the page

53:24should scroll. So you see that it's one of

53:26those sort of, sort of, let's say slightly

53:29difficult,

53:30but annoying things to achieve in web

53:32design, where you're always keeping the

53:34user in context

53:34of what they're asking for. And when you

53:37click on that, I kind of feel like maybe

53:40the thing should

53:41load the new inside of the top and the old

53:43thing should go at the bottom. It's kind of

53:46unintuitive

53:47or click on it, but then scroll the page

53:49down. So the user sort of sees that story

53:51being told

53:52sort of vertically or goes to the right or

53:54goes to the left, go wherever you want,

53:56just make it

53:56more obvious. If you see this interaction,

53:59he clicks and then it goes over to the

54:02bottom,

54:02but we have to then scroll down to see it.

54:04So watch this. When I want to ask, but don

54:06't

54:06necessarily know how. So here we go. And

54:09here, I'm interested to know which products

54:11drove the

54:12sudden increase. So click, I get a plain

54:15natural language, a company by

54:18visualization, all coming

54:20from Tableau. Now this inside here shows me

54:23the top drivers behind this change,

54:26e-phones and Simpson phones. Well, like

54:29Ryan said earlier, these smart devices,

54:32they're all the buzz these days. And of

54:35course, we all have our own questions too.

54:38Now, in this case, I'm wondering if we'll

54:40fulfill these orders. So I can simply start

54:43by typing in

54:44pulse where smart and contextual

54:45recommendations come up at every step with

54:48the help of AI.

54:49>> Okay. So let's just stop this for a

54:51second and make sure. Will we fulfill phone

54:55orders?

54:56Inventory fill rate, North American phone.

55:01What is projected inventory fill rate? Is

55:04there

55:04seasonality in inventory fill rate? Okay.

55:07So basically the context here is

55:14you've seen an increase in certain sales.

55:16And then you're basically asking, hey, are

55:19we going to be

55:19able to meet these orders? And by asking

55:22the question Tableau, let's call it pulse,

55:26but it's

55:26actually GPT has come up with three

55:29perspectives. Let's say inventory full rate

55:33, North America

55:35phone inventory, full rate, North America

55:37tablet, and then pending orders, North

55:40America. So three

55:41ways that you could find out this answer

55:44based on metrics that exist. I have to

55:47assume that it's

55:47based on metrics that already exists. And

55:51yeah, let's play this out to see what's

55:54going on. I like

55:54the social element telling you which one

55:56most people follow. Maybe it should be

55:58sorted that way.

55:58Right. So most people are probably just

56:00going to go to inventory fill rate, North

56:02America tablet,

56:03but the context of tablet and phone is

56:05super interesting. And I wonder if there's

56:07an inventory

56:08fill rate North America view of which

56:10tablet and phone are actual filters. And so

56:13maybe there's

56:14some nesting and some hierarchy work that

56:16could work here to say that these are the

56:18same metric

56:19with different options. That could be a

56:21nice little interaction, but anyway, let's

56:23play this

56:24through. And there you have it. Another

56:28metric, another insight all answer my

56:31question. So that

56:33kind of does answer my question. Inventory

56:35for it is the metric North America is a

56:37regional

56:38filter category, the regional filter and

56:40the monster month comparison is again,

56:43contextual.

56:44Now, now what is interesting is I think

56:45they've kept the monster month comparison

56:47consistent

56:49across these, but you do get an indicator

56:52says low. I assume you have like certain

56:55thresholds

56:55where you can set that when you build the

56:58metric in here, you have an overview and

57:00then you have

57:01a breakdown view. We'll see what that is in

57:03a second. You've got this little light

57:05bulbs

57:06inventory for right versus of that is now

57:0891% a drop below the expected range of nice

57:10five to 92.

57:11That doesn't mean it's bad. A new

57:13unfavorable trend has been detected for

57:15inventory for rate.

57:16It's trending down compared to the previous

57:17trend. You see, that's an interesting one.

57:19Like

57:20if you have a blockbuster sales event, then

57:23you will get a unfavorable inventory for

57:25right. And so

57:26that is maybe like a false positive, right?

57:29It is bad. But if it's off the back of a

57:33really

57:33successful sales period, let's say you sell

57:35out of a product. That's not a bad thing.

57:37Especially if

57:39you saw everyone that you made and you can

57:41't make them fast enough. That maybe leads

57:44to other

57:45questions around price and availability and

57:47resources and material, right? So that

57:51could

57:51be something that scales on favor and

57:53favorably with Habra metrics, right? Lots

57:55of false positive

57:56where people are going in, it's saying

57:57something's unfavorable, but you go and

57:59look at it and you're

57:59like, well, that makes total sense. In fact

58:01, that happens all the time because I just

58:02ran a promotion

58:03and I expect my inventory for it to be

58:05super low because you build up stock to,

58:07you know, release

58:08them all at once Easter eggs, Christmas,

58:11Halloween, these are all events that will

58:13have incredibly low

58:15info inventory fill rates right after they

58:16've happened. You're not going to have a

58:18high fill

58:19rate for something like turkeys. So how

58:21does the system understand that content?

58:23How do you feed

58:24that content into the system into these

58:26metrics to help you sort of understand that

58:28? So super

58:28interesting challenge for tablets figure

58:30out that since at the speed of thought,

58:32now what's happening behind the scenes is

58:35post detects business critical insights,

58:38such as

58:39drivers, trend, forecast, and outliers, all

58:41with trust, this statistical calculations

58:43from Tableau

58:44where AI can make the language more consum

58:47able and deliver them to you pro actively so

58:50that all

58:51of us can see and understand data. So I

58:54actually agree with that framing of AI. I

58:56think it's a

58:57really good way of putting it. AI is like

59:00an interface to the data. It allows you to

59:02ask a

59:03question and find the answer and it's

59:05helping sort of synthesize what you want

59:07out of the data.

59:08What is interesting is in the background,

59:12is AI doing the hard work or is it actually

59:15not just AI

59:16doing the hard work? Is there actually some

59:18other more complex thing going on? Are

59:20analysts setting

59:20up these metrics and building out these

59:22contexts like we've seen in the past with

59:24our data?

59:25Who is doing that work? Who is setting the

59:27contextual landscape for this AI machine to

59:30go off and understand how things work? So

59:32that is super interesting. Now we do have a

59:35new chart

59:35here. This one's amber. So let's try and

59:37figure out why. You ask what is the

59:39projected inventory

59:40file rate. So I think it's drawn this amber

59:43chart. Again, this was out of the fold,

59:46wasn't it?

59:47We landed on this page and it's only now

59:49that we've scrolled down, it might be

59:50because the demo

59:51screen is small and it's a laptop, but kind

59:54of vilifies my point earlier on. This nice

59:58little

59:58chart was out of the fold. So we didn't

60:00actually see it until you scroll down. But

60:02if this trend

60:04continues, inventory fill rate for phones

60:07is predicted to be 89% call. Is that a good

60:10or bad?

60:10That's fine. That might be fine if you've

60:12just had a new iPhone, for example. The new

60:15iPhone has just

60:15come out and you can't get any until

60:17November already, as is always the case

60:19because people

60:20just want new friends every year,

60:22apparently. But that's not a bad thing when

60:24you can't fill

60:25inventory that fast for new products. That

60:28happens every single year. So who sets the

60:30context to say

60:32this is good or bad? Can you tell the

60:34system that only let me know if it's below

60:3710%?

60:37Set a 10% threshold for inventory fill rate

60:40. If it falls below this, then we're going

60:42to have

60:43long-term supply issues and long-term

60:45demand issues and long-term fulfillment

60:47issues. Where

60:49is that sort of capability? Anyway, let's

60:51play on. But it doesn't stop there. What if

60:56I want to stay

60:56on top of these changes and share my

60:59findings? Okay. Well, in just two clicks

61:02here, my entire

61:04team now is following inventory fill rate

61:06so that they can start tracking and acting

61:08on it too.

61:08So if this is Tableau Pulse and this is

61:11going to sit inside a Tableau Cloud, that

61:14interface we just

61:15saw there is completely different to the

61:17sharing interface we see today. Side by

61:19side, they're just

61:20not the same. I'll try and put a screenshot

61:21of it up on screen. Yeah, they're just not

61:24the same. So

61:24there's a whole lot of, let's say, staging

61:28here that if it's genuinely going to be

61:33available in

61:34December as we're seeing it like this, wow,

61:37there's a lot of change coming to the Table

61:39au platform,

61:40definitely. And to confirm that, I can back

61:43out to my personal homepage where I'll be

61:45able to see

61:45inventory fill rate and I'll say other

61:47metrics that I care about. So let's look at

61:51this as well.

61:52Device sales had an unusual spike. Yeah.

61:55These are the same three summaries that we

61:58saw before.

61:58There's always this sort of, was this

62:01helpful up or down? It feels like

62:03this is sort of like you're training the

62:06model on what you find useful. And if you

62:09do enough

62:09of these, eventually over time, they'll

62:11just probably stop asking you because they

62:13know,

62:14right? Google used to do this. They used to

62:16ask you what you thought of a link, and

62:18then they

62:18just eventually stopped because they

62:19figured out a way of figuring it out

62:20without having to ask you.

62:22So maybe something we see in the early

62:24stages of this kind of technology, but in

62:26the future,

62:26it will just know. It will just know based

62:28on metrics, like how much time you spent,

62:30how much you engaged with it. And then they

62:32'll start to sort of train analysts on how

62:34to create

62:35engaging metrics that actually drive

62:37actions and stuff like that. So super

62:39interesting.

62:39There are three metrics. Who knows what's

62:42below the fold. All metrics means probably

62:46everything

62:46that's shared with you, whereas the ones

62:47you're following are the ones you

62:49passionately care about.

62:50The best part is Pulse isn't just limited

62:54to Tableau. It's everywhere you and your

62:57teams

62:58work today. Like an email where you can

63:01receive a daily dose of your metrics. Let's

63:03ask April

63:04from my team here. Did you get it? Awesome.

63:06There you go. So you get a nice email and

63:12Tableau Pulse. Oh, interesting. That looks

63:18like a Pulse tab inside of the Tableau

63:20mobile app,

63:21right? That's what that's got to be. That's

63:24all I can assume it is. You have Home and

63:27Explore,

63:28which are two tabs you get in Tableau today

63:30. Home is like the favorites and everything

63:31you follow.

63:32Explore is that classic Explore tab. Pulse

63:34feels like it's going to be a new tab,

63:36which means

63:36probably when it comes to cloud, it will

63:38also be its own little slide tab. It might

63:40even be the

63:40default place you start to go then to find

63:43other things. So it all makes sense. On

63:45your phone,

63:46where you can receive insights on the go. I

63:48am scrolling through them right now.

63:52And of course, in Slack, as you saw earlier

63:55, we get to collaborate more easily.

64:00But here at Dreamforce, we all care about

64:03Salesforce, don't we?

64:05You guessed it. I land in Salesforce and

64:13find a-

64:14- Silence. The traditional Tableau kind of

64:18customer doesn't care about Salesforce.

64:20Maybe I'm wrong.

64:21Let me know. But yes, he was kind of hoping

64:24they'd all say Salesforce and didn't quite

64:26land that one.

64:27But this is Tableau Pulse inside of Sales

64:29force. Now it's interesting that as soon as

64:31it goes to

64:32Salesforce, it feels different, doesn't it?

64:34It looks slightly different. And Einstein

64:36has popped

64:36up already and you've got slightly

64:38different sort of layout. It feels like a

64:40more Salesforce-centric

64:42version of Pulse. So that's interesting.

64:45There's something called Omni-channel

64:47macros. I love

64:48seeing what else is on screen whenever you

64:49take like a screen grab like this, because

64:51it kind of

64:51shows you the context of what people are

64:53thinking. Einstein is sitting in the corner

64:55there so you can

64:55ask some questions, but it's more or less

64:58the same thing. This might be like a new

65:00lightning web

65:01component that sits specifically inside of

65:03Tableau. There is a new one for Tableau in

65:0623.3. And so

65:08this might be like a Pulse extension of

65:10that, which is kind of nice. - Pulse

65:12homepage deeply

65:13embedded with all my favorite metrics in

65:15one place. And clicking into the metric

65:17here, I see

65:18the same detailed view with the same

65:20interactions that we've seen. Okay. - So

65:24that answers the

65:24question. You have to have Tableau Cloud to

65:26have that kind of interaction. I don't see

65:28how you

65:28achieve that with the Tableau server behind

65:31firewalls. It just doesn't seem to work. So

65:35that might be a nicety you only get inside

65:39of Tableau Cloud. But with Tableau Server,

65:44you might

65:44be able to use something like connected

65:46apps to allow this to happen and get that

65:49working. But

65:50yeah. - That's off. Let's rewind. We trust

65:53it. We're familiar with it. It's quick to

65:57find and

65:57easy to use. Now all of us can truly

66:00succeed from anywhere. And this is Tableau

66:04Pulse. Metrics and

66:07insights reimagined, powered by Tableau AI

66:09to enhance and accelerate everyone, to make

66:13informed

66:13decisions fast and take actions on data.

66:17Thanks for staying with me and passing back

66:21to you Francois.

66:22- All right. Great job Homer. I love Paul.

66:28- I'm gonna pause there. I'm recording this

66:31on a weekend,

66:31so I have to keep popping off to do family

66:34and personal DTS. I'm gonna hop away and I

66:38'm gonna

66:38come back and we'll carry on with the rest

66:40of this a little later on. But so far,

66:41super interesting.

66:42I'm kind of keen where the rest of this

66:44keynote goes. So yeah, I'll see you in a

66:46second. Okay,

66:47we're back. Let's carry on. - I use it now

66:50every day and it's really made my daily

66:52experience

66:52easier. I get all of the information at a

66:55glance and I can take action on it really,

66:58really quickly.

66:58But you know, when you want to explore data

67:01, this is really... - Again, this is another

67:04sort of

67:04framing. Then use the build dashboards. Now

67:07you ask questions. And again, this is

67:10really important.

67:11I cannot... Like if you build dashboards

67:14today, if you're becoming a data analyst

67:17today and you think

67:18learning how to build dashboards will be

67:21the way that Tableau is heading, I think it

67:23's changing.

67:24I think you need to become more of a data

67:26engineer, data modeling expert, metadata

67:28expert

67:29in order to enable people to do what Table

67:32au is talking about now. How to answer, how

67:35to frame,

67:35how to contextualize questions. And

67:38ultimately the data is going to be treated.

67:40It's going to need

67:41to be treated. It's not just going to work

67:43out of the box. When you get, let's say, a

67:45purchase data

67:45from a superstore, it doesn't come ready to

67:48do this. When you get transaction data from

67:52your bank

67:52or from an online store, it does not come

67:55ready to do this. In order to get it to

67:57this place where

67:58you can actually use AI on top of it, you

67:59're going to need to clean it, prep it, put

68:01it into a data

68:02model, warehouse it, make it available,

68:05think about security, think about all these

68:07different

68:08things. That is where being a data analyst

68:10is going to be heading to. And that's where

68:12being

68:12a data engineer or being a data modeler

68:14will sort of pay dividends. I think the

68:17people who

68:17build dashboards today are just going to

68:19move further back into the stack and do

68:21more things

68:21to enable these kinds of experiences. So

68:24everyday people can just go and ask

68:25questions.

68:26This is where the Tableau superpowers come

68:29in. Tableau is easy to use. You can easily

68:32slice and

68:32dice your data however way you want. But

68:36normally, when you start in Tableau, you

68:38got a blank screen

68:39and some data, and you have to kind of

68:41learn the product. You have to know how to

68:43drag and drop,

68:43where the features are. You need a

68:45visionary around you. You need the

68:48documentation. You

68:49need experts to help you get successful.

68:52But what if we brought AI into the

68:55experience?

68:56What if you had Einstein with you that

68:58understood Tableau, could help you answer

69:00your questions more

69:01easily? Well, the future of Tableau is to

69:05have AI embedded in the exploration

69:07experience,

69:08where AI fully understands how to use the

69:11product, where AI understands the meaning

69:15of your questions

69:16and can help essentially do the drag and

69:19drop for you. So it's not about either or,

69:21it's an and.

69:22It's augmenting the experience, making it

69:2510 to 100 times easier, and making all of

69:30you more

69:31productive. That is the goal. And so today,

69:34I'm pleased to announce the Einstein Co-p

69:37ilot for Tableau.

69:41So Einstein Co-pilot, June 24th, that's

69:44almost a year away, just under a year away.

69:47This is going to be sensational. I think

69:51this is why it has such a long run up. So

69:54much has to

69:55happen. And I think there's some

69:56fundamental questions that Tableau have to

69:58answer before

69:59this kind of tool gets deployed. And yes,

70:01you guessed it, it'll probably be cloud

70:04first. You

70:04won't get this for Tableau server. No way.

70:07Tableau server's probably got a 25 release,

70:09if that makes sense. I just, the more I

70:12think about sort of Salesforce being a SaaS

70:15company,

70:16the more I think about where Tableau is

70:18heading, I just cannot see how some of the

70:21ideas they're

70:22thinking about here come to Tableau server

70:25in an expedient way. Because like I said

70:27before,

70:28for these things to work on Tableau server,

70:31the requirements on infrastructure are just

70:34going to

70:34keep going up and up and up until the cost

70:37of doing those things kind of pushes you to

70:40the

70:40cloud, honestly. I've not sort of been in

70:43touch with Tableau server for the last

70:45couple of years

70:46now, because I just haven't needed to use

70:48it as much. Tableau cloud has been the

70:50predominant

70:51side where clients are working. And so

70:53actually knowing how to manage the backend

70:55and infrastructure

70:56of that is almost sort of non-existent

70:59because all you have to do is go into

71:01online.tableau.com

71:03and manage the front-end user face there.

71:05So I'd be really intrigued to know what are

71:09the

71:09server requirements, what is the server

71:11usage of a server today, and what are the

71:14features that we

71:15get in the cloud that aren't available yet

71:17that would sort of increase that and as

71:18Tableau roll

71:19out these AI features, how is that going to

71:22play out long term? Anyway, interesting,

71:24interesting,

71:25interesting sort of thing to see play out.

71:30Oh, yes. The Einstein Copilot is really

71:34going to be

71:34a core part of the Tableau experience that

71:37enables you to ask questions of your data

71:41and it will

71:41basically explore it for you. It'll give

71:44you better results because it understands a

71:47lot of

71:47the context. You'll have better best

71:50practices built in and ultimately you'll

71:52just be more

71:52successful. You'll be able to ask more

71:55questions and drive more value to your

71:57organization or you

71:58can just put your feet back up and enjoy

72:01the rest of your day. So let's see Einstein

72:05Copilot in

72:06action. So for that, please welcome Hanto

72:08Mei. Hanto. Thanks, Francois. In the next

72:13four minutes,

72:14I'm going to show you how Einstein Copilot

72:16backed by the Einstein Trust Layer can

72:19speed up and

72:20improve the quality of your data analysis.

72:23Here in Tableau prep, I have customer

72:25purchase data

72:25for nationwide change. I want to use this

72:28data to create personalized experiences for

72:32my customers

72:33that will drive incremental revenue. To do

72:35so, I need to know where my customers are.

72:38So I need

72:38their postal codes. Unfortunately, my

72:40postal codes are trapped in this customer

72:43mailing address

72:43column. Normally, I'd have to figure out

72:46how to write a calculation to extract this,

72:49but with

72:49Einstein, all I need to do is ask in

72:53natural language. And on the fly, Einstein

73:01is able to

73:02create this calculation for us. Now, all I

73:05need to do. Now, I think people have seen

73:08this demo

73:09before, which is why the crowd didn't have

73:11sort of a big reaction. A lot of the people

73:13at this

73:13conference were at Tableau conference. And

73:15so they've seen this demo, they've seen

73:17this example

73:18before. That said, it doesn't take away

73:20from the, let's say, awesomeness of this

73:22because this is

73:24sort of why I believe AI is fundamentally

73:26going to change the way analytics is done.

73:29You see previously, if you just looked at

73:31this problem, let's say you're a data

73:33analyst and

73:33you're not like an experienced data analyst

73:36, you've been working the field maybe a year

73:38or

73:38two years, okay? And this is the data that

73:40you get. And someone asked you, "Hey, how

73:42do you

73:43extract the postcode from this column?"

73:47Your first instinct might be to pass out

73:50the commas to get

73:52the final field of each column, which is

73:55still leave you with Florida, 322, 444, USA

73:59.

73:59And then the next thing you might do is to

74:01say, okay, if you find any one of these

74:04states,

74:04go ahead and remove that, which will leave

74:07you with 9, 9, 2, 1, 0, and then USA. But

74:11you see,

74:11that's not always consistent. You see North

74:14Carolina down here is 28405 and doesn't

74:18have

74:18a country on the end. So sometimes it's USA

74:20, sometimes it's US. And you get into this

74:23really

74:23messy world where if you really have to

74:25clean this, you kind of use brute force

74:27method and you

74:28apply like a really convoluted way of

74:30passing this in steps, maybe 15, 16 steps

74:33to get to where you

74:33need to. And then you kind of go from there

74:36. Well, the experienced analysts will be

74:39able to

74:39look at this text and say, "Hmm, there's a

74:41pattern here." The postcode is essentially

74:43a certain number of digits followed by

74:45letters essentially. And so we can actually

74:49go and find

74:50that in the string by just looking for that

74:52, using that pattern. And the technology

74:55that helps you do

74:56that is called regex. Now you wouldn't know

74:58the term regex. You wouldn't even, you

75:00might stumble

75:00across it if you let's go into, if say, let

75:02's say you go into Reddit and you ask a

75:03question,

75:03you wait a few days, someone replies, or

75:05you Google, you come across this thing and

75:07it's

75:07called regex. Then you go to regex 101 and

75:09you start trying to use it. And you're like

75:11, okay,

75:11this is interesting. You go down a rabbit

75:13hole, 30 minutes later, you're now figuring

75:15out how to

75:16write regex for this and it works, but it

75:17doesn't work some of the time. And other

75:19times you go test

75:20it. And so you're not so confident. So you

75:22try this thing and you kind of move on.

75:24That whole

75:25flow probably has taken 40 minutes, maybe

75:2730 minutes. If you're super fast, you know

75:29what to

75:29search and you're kind of adept and you're

75:31kind of really going down this route of

75:33exploratory

75:33sort of data analysis. That said, the

75:36simple fact that you can go in and type the

75:38question, say,

75:39I want the postcodes. You don't have to

75:41know the term regex. You don't have to know

75:44regex pattern matching. You don't even have

75:46to go to regex 101 or even ask the question

75:48. You can just

75:49go and type the ask and the AI tool helps

75:52you figure out what you need to know. And

75:55here's

75:55the added bit. Now that you see that term

75:58regex p extract or whatever, that should p

76:02ique your

76:03interest. If you're a good data analyst,

76:04that will pique your interest and go, huh,

76:06what is this?

76:07And so you then go Google that thing and

76:09you understand what it is. And now not AI

76:12hasn't

76:12just solved the problem. It's also given

76:14you a shortcut directly to the thing you

76:16need to learn

76:16and the way it's working in order to

76:18enhance that. So now you kind of start to

76:21use AI as a way of

76:22discovering things you need to learn as

76:23well as a way of helping you, which is sort

76:25of a double-edged

76:26thing. The next time you come to this, you

76:28'll ask specifically, hey, can you use regex

76:31to solve this

76:32kind of problem? It's quite complex. And

76:34now you're having a much higher level

76:35discussion

76:36with AI. You're still using AI, but you

76:38still understand what's going on and you're

76:40building

76:40your understanding as you go along. So it's

76:42also helping you with data literacy. So I

76:44think this

76:44is a, this to me is probably the biggest

76:47opportunity that Tableau has just to help

76:50everyday data analysts who actually still

76:53do build data sets and or data models and

76:55or visualizations.

76:56And more importantly, it's also going to

76:58help bring the skill level up for everyone

77:01who's

77:02already doing this stuff. It's going to

77:03bring them right up so they too have access

77:04and awareness of

77:05things like LODs, all these complex terms

77:08like set actions. It's not going to solve

77:10the problem,

77:11but it might just alert you to the

77:13capabilities behind these things. Anyway,

77:15let's keep seeing

77:16the demos and see the examples. Fortunately

77:18, my poster codes are trapped in this

77:20customer

77:20mailing address column. Normally I'd have

77:23to figure out how to write a calculation to

77:25extract

77:25this, but with Einstein, all I need to do

77:29is ask in natural language.

77:32And on the fly, Einstein is able to create

77:38this calculation for us.

77:41Now all I need to do is give it a name.

77:46And voila, a new column in my data with the

77:52customer poster.

77:54What I would be really interested to know

77:58is what's in the reference tab here. I was

78:00just

78:00thinking about it. I was like, huh, there's

78:02a reference tab. Is a reference tab showing

78:04you

78:05what it's doing? Like the nice thing with

78:07websites like Regex 101 is that it shows

78:10you how it's

78:11working. What I would love Einstein Co-P

78:13ilot to do is to almost play through an

78:15example of a

78:16calculation in the context of Tableau to

78:18show you what's happening and even show you

78:20what's going

78:20on. I've put this here. I've done this

78:23there. Almost guide you through the steps.

78:25You know,

78:26ChatGPT can do this today. It will tell you

78:28do this, do this, do that. Obviously it's

78:30not

78:31perfect, but if you're training a model

78:33specifically around Tableau, then actually

78:35it

78:35should be possible to be able to instruct

78:37it and give you instructions on what

78:39exactly is going on.

78:40Almost reverse write the blog post that you

78:42would write if you'd figured out how to do

78:45this.

78:48Now all I need to do is give it a name

78:52and voila, a new column in my data with the

78:57customer postal code.

79:00And I did all of this in a matter of

79:05seconds and without writing one line of

79:09calculation code.

79:10So with Einstein, you and anybody can use

79:15Tableau Prep to transform the data they

79:17need into the

79:17format they want. So how are we going to

79:20reach these customers though? Well, did you

79:23know

79:23that you can use Tableau to visually

79:25explore your audience data and to create

79:28audience segments in

79:29Data Cloud? Wrong screen.

79:34We're good? Okay, there we are. A little

79:38excitement there. So I have just connected

79:43with the help of my

79:44friends back there to Data Cloud. Now, when

79:48I'm presented with a blank slate like this,

79:51I ask

79:52myself, where the heck do I begin? The

79:55really interesting thing here is that

79:58Einstein is already

79:58available on the right hand side and the

80:00right hand side has become this sort of

80:01contextual

80:02place to find out more about the data set,

80:04more about what's going on, more about the

80:06metadata

80:07inside of Tableau. And it's sort of grown

80:09legs. This is also where explain data used

80:11to be. I kind

80:12of feel like that's going to get sort of

80:14pushed to the side now because you'll have

80:16our state

80:16explained as those are all going to go away

80:19and Einstein and Tableau GPT and we'll pass

80:21and metrics

80:22will sort of sit in this space more

80:24squarely ready to help you sort of answer

80:25questions and pull out

80:27insights rather than sort of forcing you to

80:29come up with the answer question yourself.

80:32The other

80:33nice thing here is obviously this is a web

80:35edit experience and this is a draft. So

80:37that means

80:37he's exclusively using the authoring

80:40experience in web edit. And last edited the

80:43September the 12th,

80:45that would have been the time of the demo.

80:47So he's using a sort of live take of this,

80:49then if that makes sense. All these details

80:51do matter because I think a lot of people

80:54think

80:54about Tableau in the desktop sort of setup

80:57when in reality, that's not how most of it

80:59works. That's

81:00not how most of Tableau demos anything

81:01anymore. It's all done in the browser. And

81:04so it's

81:05interesting to see that I don't think you

81:07'll get the same experience in desktop. I

81:09just don't think

81:10that will pull through unless you're a

81:11Tableau cloud customer. I kind of think

81:13this is going to

81:14be a really a smoother experience in the

81:16web because that's essentially where this

81:18will be

81:19running. Otherwise, you can just imagine

81:21sort of the back and forth between your

81:23local client and

81:24your laptop and Tableau servers when this

81:26stuff is running. But it will still be

81:28interesting to see.

81:30Now the customer purchase history, what is

81:33not clear is if this is a data set in Sales

81:36force,

81:36and that is why Einstein copilot and this

81:39technology is working really well, or if

81:41this

81:41is going to work across non Salesforce

81:43based data sources as well. So things like

81:46in your Snowflake,

81:46in your Databricks database, whatever those

81:49are, and that detail is still not clear. I

81:52assume it

81:52will work everywhere. And Tableau will be

81:54running this technology on the cloud

81:56looking at these

81:57data sets and sort of processing them. That

81:59's how some of the past features have worked

82:01. There's

82:02something called data change radar, which

82:03has essentially been taking snapshots on

82:05your server

82:06and then analyzing that on your cloud

82:07instance and then pushing you alerts when

82:09something changes

82:10that shouldn't have changed. So super

82:13interesting little nugget. Let's see how it

82:15actually works.

82:16Using generative AI and statistical

82:20analysis, Einstein is able to understand

82:23the context of

82:24your data. And in doing so, Einstein is

82:27able to suggest relevant business questions

82:30to kickstart

82:31your analysis. Let's take a look at this

82:33one about patterns of my sales over

82:36different product

82:37categories. And look, with one click and

82:40without having to drag a single pill onto a

82:44shelf,

82:45I'm able to see the viz that shows my sales

82:47for all my different product categories.

82:50But what

82:50about this pattern Einstein was looking at?

82:53I can see, yeah, that's right, outdoor

82:56sporting goods

82:57are popular in the summertime. That's not a

82:59surprise, but that gives me an idea. I know

83:02that our in-store experiences drive bigger

83:05purchases compared to online. What if we

83:09invite

83:09these outdoorsy and sporty people back into

83:13a store with an event like an outdoor pet

83:16first

83:16aid class? Well, how am I going to do that?

83:19First of all, we're going to use those

83:21postal

83:21codes we extracted earlier. And you guessed

83:24it, we're going to ask Einstein. Yeah.

83:27Show me the location of customers who

83:33bought sporting goods in the last three

83:36months by zip code.

83:37All right, I see all my customers on a map.

83:44All right, but what about my stores?

83:46How far are these customers from store

83:49locations?

83:51And look, without knowing anything about

83:58map layers.

83:59I'd seen this before, but I was paying

84:02attention to what was going on and what

84:04changed. And

84:05the great thing about this,

84:11so see customer transactions, store

84:14locations, store locations is in which

84:19there we go. So

84:21there's two data sets in this data model.

84:24One is customer transactions, one is store

84:27locations,

84:28and essentially what they're doing is

84:30relating the customer transaction to the

84:32store level data,

84:33which gives us two spatial fields. The city

84:38or yeah, there's a customer location field.

84:42So it

84:42could be the city or the whatever of the

84:44customer, their dress. And then you have

84:48the location from

84:51the store. And so to bring these two

84:52together, you are creating map layers. They

84:55're putting the

84:56two on top of each other because they're in

84:59the same data set. They have a data model

85:01relationship.

85:02So it should naturally work nicely. You can

85:04do that without a relationship. You can

85:07just bring

85:08on your store locations as a separate data

85:10source. And a new feature, nearly a year

85:13ago now,

85:14allowed you to basically overlay two

85:16separate data sets on a map without having

85:18to do any sort

85:19of join or relation to them, which is kind

85:21of powerful actually, because it allows you

85:23to bring

85:23contextual sort of map layers without

85:25having to do the dirty work of blending it

85:28or doing whatever

85:28you need it to do to make the map work. So

85:31this is quite nice. Now, why I like this

85:33demo is because

85:34it kind of shows that iteration. It kind of

85:36shows a Tableau kind of going through the

85:38steps. And

85:38again, I believe the language that's being

85:40used here is generally authentic. It's kind

85:42of what

85:42you'd ask in terms of the analysis. If you

85:44're asking good questions, that is a skill

85:46in itself,

85:47but I think this is a fair reflection of

85:49what people would actually do with AI. And

85:51it's doing

85:52what you'd expect it to do. Now, what we

85:54can't tell by this demo is how often is it

85:56good at

85:57doing this? Because you know, sometimes

85:58with AI, these things just, you know, 90%

86:00of the time,

86:00they're okay. And sorry, 90% they're good.

86:0310% of the time, they're okay. And when

86:05they fail,

86:06they fail epically, right? So this looks

86:09pretty good. It's doing a few complex

86:11things, latitude,

86:12longitude, bring it all in. There's levels

86:15of detail here that are on there on both

86:17the customer

86:17and the store locations and the marks pane.

86:21There is coloring going on. You could argue

86:25potentially

86:25there's some buffering going on. I don't

86:28know if the size of the circle, the size of

86:30the circle

86:30represents the customer count from that

86:33store. So I actually think it's sort of

86:35interesting because

86:37the customer, you know, the customer count

86:39maybe relates to like, maybe it's a

86:41specific town for

86:43these customers. And it's just showing you

86:45where those people are coming from in those

86:48towns. And

86:48that's why certain towns have bigger or

86:51smaller circles. But you have customer city

86:54, customer,

86:55I don't know, something. I can't see the

86:57location hierarchy. So again, unnecessary

87:00levels of details,

87:02unnecessary level of breakdown of the

87:04details here, but it seems pretty good. And

87:08the demo

87:09adds up. That's all I'm trying to make sure

87:11. Like, is this a file fetch demo? No, it's

87:13not.

87:14And it's probably something you'd get asked

87:15to do. And you'd be asked to put this in a

87:17dashboard.

87:18Now, the super interesting thing here is

87:22just imagine that the whole of the left-

87:24hand side

87:24doesn't exist. And all you have is the

87:26Einstein column on the right, and the

87:28customer location

87:29chart. And that's all you get. What if that

87:30's the experience of Tableau going forward,

87:32right?

87:33For every person, for everyone, that

87:34becomes the experience. But they're trying

87:36it here first with

87:37data analysts to kind of test if it's good

87:39and if it's bad, but slowly over time, this

87:43experience

87:44where you type and you see charts will be

87:45pretty much the core experience of Tableau.

87:48What do you

87:49think? Let's carry on. Or geographic roles

87:55in Tableau, I was able to create this

87:56intuitive yet

87:57complex map viz with Einstein. But don't

88:00forget, Tableau at its heart is an

88:02interactive and visual

88:04tool. I can easily grab these customers

88:07from San Francisco and with our integration

88:11with Salesforce

88:14send these customers up as an audience

88:18segment. Now that is a pretty sick feature

88:23if you say a

88:24Salesforce customer. Just being able to do

88:26the analysis, select the customers, create

88:30a segment,

88:31and push it back into Salesforce. I mean,

88:34if you're a Salesforce customer and a Table

88:37au customer,

88:37that is chef's kiss. That is perfect, right

88:40? That is the dream. Now in reality,

88:44not many people have the ability to do that

88:46or sort of trust to do that. But in reality

88:48,

88:48if that was your flow and you could enable

88:50that. And don't forget, this has not been

88:52published.

88:53This is still exploratory analysis. And it

88:55's an easy thing to forget. We're not going

88:57from

88:57dashboard to Salesforce here. We're going

89:00from sort of detailed discovery. So an

89:02analyst has

89:03been asked to go and find this customer

89:06segment and push it to where they need to

89:09get to. As soon

89:11as that question comes back, boom, they can

89:12just go back into the chart they built

89:13without publishing

89:14anything and push it off into Salesforce.

89:17So I think that's also a really nice

89:19celebration of

89:20Tableau's heritage. It was always a data

89:22exploration tool. And it's kind of easy to

89:24miss in this demo,

89:25right? Because you're so caught up in the

89:27feature. But actually the fact they didn't

89:28publish it

89:29before doing this, I think that's super

89:30powerful. It's super important. Getting

89:32straight to the

89:33action rather than this whole governance

89:35and publishing thing. If you're empowered

89:37to do this,

89:38just go ahead, push it to the Salesforce

89:40instance and move on to the next task.

89:42Perfect.

89:42And I did all of this without leaving Table

89:47au. Now, I can't get to the screen. Oh,

89:56sorry, my bad.

89:57And now my marketing team has all the

90:00information they need to activate the

90:04segment.

90:05In a demo, when something goes wrong, and

90:07you know what's gone wrong, but the person

90:09who's

90:10experiencing it go wrong, at the moment, it

90:12doesn't do what they're expecting. It doesn

90:14't

90:14matter if they're on point. It doesn't

90:16matter if they're about to nail the demo.

90:18Nothing matters.

90:19As soon as something goes wrong, your brain

90:22just goes into freeze mode because you're

90:24like,

90:24what's going on here? You can't really

90:27think fast enough to kind of rescue Thob.

90:29So for April to

90:30sort of notice that and catch that moment

90:33and get Honto back on track, it's like,

90:36that is superhero

90:38stuff. I'd like to say not all heroes have

90:41capes, but April definitely deserves one

90:44there because

90:45it's a small thing and I'm sure Honto super

90:47appreciated it at the time. But it's so

90:51easy,

90:51so easy in that instance to just freeze for

90:54like two, three minutes, fix it and off you

90:56go.

90:58This. There we go. Like this to all my

91:08customers.

91:09The thing that's really unfair here is that

91:12Honto is not controlling the transitions

91:15between

91:15the slides from phone to laptop. Honto is

91:18not controlling that. So on top of like,

91:20things going wrong, like whoever's

91:23controlling the PowerPoint, it's just not

91:26working quickly enough.

91:27So it's not, it's one of those sort of comp

91:29ounding effects. It's kind of funny, but

91:32yeah, anyway,

91:32it looks pretty, pretty good. I did manage,

91:37and maybe this is why we didn't have a good

91:39demo.

91:39I managed to sneak my dog into the keynote

91:42presentation.

91:43All right. Now, in summary, Einstein Cop

91:52ilot, backed by the Einstein Trust Layer,

91:56will let

91:56anybody who can ask a question visually

91:58explore their data in Tableau. And with our

92:01deep integrations

92:02in the data cloud, you can get the insights

92:04you need to connect with your customers

92:07faster.

92:07And back to you, Francois. Awesome. Great

92:10job, Honto.

92:11I got to say any demo that includes a dog

92:15photo is a fantastic demo. That's the

92:19simple demo rules.

92:20You know, with the Einstein Copilot, you

92:23basically have an expert with you at all

92:25times to help you get to the answers faster

92:28. It is powerful. But we want to go further.

92:31We want to infuse data and AI everywhere.

92:34You know, today, when you think about your

92:35business

92:36applications, well, they're your business

92:38applications. If you want to get answers

92:40from

92:40those, you go somewhere else, and you have

92:43that typical swivel chair problem. You lose

92:45context,

92:46you lose the data. It's just frustrating

92:49for the users. Well, who wants a dumb CRM

92:53or a dumb

92:54application? You want smarts built in. You

92:57want insights right where you do your work.

93:01And this

93:02is what we're trying to do with our brand

93:04new intelligent applications for Salesforce

93:07. These are

93:08pre-built applications available out of the

93:11box that you can use to deliver insights

93:14right in

93:14the flow of work. So what so these are God,

93:17I want to say these are templates. But I

93:21think they're

93:22specifically Salesforce templates, right?

93:25So they're like the Tableau Exchange

93:27dashboard

93:28starters or whatever they call them. They

93:31're not called accelerators, they're called

93:34in the Tableau

93:35world. I think they're the Salesforce

93:36equivalent of Tableau accelerators, if that

93:38makes sense.

93:39Purpose built apps that hook into various

93:42parts of your CRM ready to go with all the

93:44insight built in.

93:46But with a little bit of Tableau, I think,

93:49and a little bit of Salesforce as well

93:51infused. So let's

93:52have a look. Whether you're using Service

93:55Cloud or Revenue Cloud, you have insights

93:58automatically

93:59there for you. It's easier to get going.

94:01You just deploy them and boom, you've got

94:04value

94:05automatically. So this is coming out

94:07November 23. This might be the first time

94:10you see AI and

94:12Einstein available in some sort of Tableau

94:15context. So that's super cool. But of

94:18course, it's in the CRM.

94:19It's contextual to the job you have at hand

94:23. The insights, the dashboards, the KPIs are

94:26built

94:27for the task that you have. And of course,

94:29it leverages the complete power of the

94:32platform

94:33from CRM analytics to Tableau to Data Cloud

94:36. This is the full power of Salesforce

94:40coming together

94:41in these rich applications. So to show our

94:44next applications, please welcome Shreevi

94:48Agandeshawan to the stage. Shreevi. >>

94:52Thank you, Francois. Hello, Dreyfus.

94:55As engineering leader for intelligent

94:57applications, I'm super thrilled to be

95:00showing you today what my

95:01team has built. Tailored experiences helps

95:04everyone achieve their work faster by

95:06streamlining data.

95:08With service intelligence, you get key

95:10insights to effectively run your service

95:13operations.

95:13Out-of-the-box pre-built dashboards enable

95:16you key insights for service leaders,

95:19comprehensive case views, omni-channel, and

95:23conversation mining. So let's jump in.

95:26With omni-channel view, routing cases is

95:30easier than ever. Leaders can track their

95:33volumes of work

95:35by cases and channels to reduce costs and

95:38improve service operations.

95:41Behind the scenes, all of these data

95:44sources from various channels, conversation

95:47data, omni-channel

95:49data, service data, and third party are all

95:53harmonized and powered by Data Cloud.

95:56Some of the most valuable data we have

95:59about service operations is the millions of

96:04hours

96:05agents spend on call every day with our

96:07customers. What if we use AI to unlock the

96:11data?

96:12>> So I might have been wrong about these

96:15being Tableau charts. Part of me thinks

96:19they're actually

96:20native within Salesforce. The way that page

96:22loaded made it look more like a web element

96:24than it was

96:24anything else. But I think this is one of

96:29those effects from what you call it. It

96:35used to be

96:36called Einstein Analytics. It might have

96:38been a leftover from Einstein Analytics. It

96:40sits within

96:40the Tableau world and Tableau framing of

96:43analytics, but is being deployed inside of

96:46Salesforce. So

96:47I don't quite understand what this is. It

96:49could also just be Tableau with really good

96:51animations

96:52and some caching in the background just to

96:53make it load nice and fast. But I'm not 100

96:56% sure.

96:57You could definitely build something that

96:58looks like this in Tableau. I just can't

97:02quite place

97:03what this little box is inside of Sales

97:04force. But anyway, let's keep going. We

97:06might get a clue.

97:07>> Service intelligence now brings in

97:11Einstein conversation mining,

97:13which mines all the customer interactions,

97:16integrated, identify the top reasons

97:20and key drivers and highlights it in a

97:24trusted and secure way. As you can see here

97:28,

97:28customers are having a lot of questions

97:30about their invoice.

97:32>> I think that is Tableau. Wait, wait,

97:34wait. There's a clue.

97:35>> With these insights, service leaders can

97:46identify and focus on the areas that they

97:49may

97:49not otherwise be aware of. With cases

97:53dashboard, leaders get a comprehensive view

97:57of their team's

97:58work on high level stats such as total

98:04escalated cases, CSAT by channel. As you

98:08can see here.

98:11>> There's a little bit of a spacing issue

98:12there. They're kind of being cut off. I

98:16think it might be

98:16because they're on a laptop and it's comp

98:18ressing everything and the charts don't

98:20quite fit the

98:21container that's been put around the object

98:24. A little bit of HTML tweaking that needs

98:28to go on

98:28there. But otherwise, I think these

98:30actually look quite nice. They're quite

98:32practical from

98:32a business perspective because I think they

98:34just answer the question and just get on

98:35with it.

98:36There's no flair and they're fast and they

98:38're efficient. They're going to do the work.

98:40As someone who doesn't know what tool this

98:43is, I think this is Einstein,

98:45then I think there's also this other side

98:48of Salesforce, which is the Salesforce

98:51platform

98:51itself has this huge plethora of data. If

98:54you go and build pre-built dashboards that

98:57answer

98:58typical questions that most customers

99:00answer, why build it in Tableau when you

99:02can build native web

99:03experiences that run even faster than Table

99:06au? That speaks to part of the challenge of

99:09innovation.

99:10When you're doing data exploration, you

99:11need a lot more sort of to be able to let

99:14the user choose

99:15where the questions and answers go. But if

99:17you already know and you're going to

99:19prescribe the

99:20questions and you're going to want to

99:21answer them in a specific way because that

99:23's the best

99:23way to answer it, then you can just go

99:25ahead and build a native web experience.

99:27In the monthly trend, our incoming cases

99:30have been spiking up.

99:32Normally, we would need to spend time

99:35analyzing this. But now, with one click, I

99:41can ask Einstein.

99:43Service intelligence brings me the power of

99:46generative AI, guiding me through

99:50intelligent

99:50prompts to understand the key factors that

99:54are driving my case escalation. As you can

99:58see here,

99:59it indicates that cases reason being

100:02billing as a topmost reason for the

100:05escalation.

100:08So let me explore this data more deeply.

100:12Right from here, I can explore into Tableau

100:16.

100:16So we switch into Tableau. There's the

100:19Tableau chart.

100:20Service intelligence brings the power of

100:23Salesforce Data Cloud

100:25and Tableau visual exploration analytics

100:29right for you.

100:31Now, being me, I have spotted something.

100:36There is a new icon for that data set on

100:39the top left.

100:40So that suggests a new data object. The

100:44other thing is when she clicked on it,

100:47she went straight into a pre-built

100:48dashboard in Tableau. So it's not just

100:51taking the data

100:52into Tableau and then leaving you to build

100:54this yourself. It's actually going ahead

100:56and recreating

100:57that chart you saw, but in Tableau, which

101:00is sort of interesting. Like, why isn't

101:02that just

101:03the application view of that thing? And

101:06again, it speaks to this idea that maybe

101:08that is it.

101:08They're building a native web sort of chart

101:10interface in the front. But then when you

101:13click,

101:14you go to the editable version, which is

101:16designed for data exploration inside of

101:18Tableau.

101:19That looks pretty cool. As you can see here

101:22, I have maintained the data context

101:25and connected to the same insights without

101:28any additional effort.

101:30Now, let me drill into the data more deeply

101:34. I can pull up a quick sheet in Tableau.

101:39And did you see that? I don't have to

101:41connect to the data again. I have it right

101:44here from

101:45Service Intelligence. So within a few

101:48clicks, I want to understand how my cases

101:51are geographically

101:53distributed across in the US. As you can

101:57see here, Oregon and Marin are having the

102:00higher spots.

102:01And let me get this insight back to my team

102:04. Right from Tableau, I can publish to Sales

102:10force

102:10within a few clicks. All of my insights are

102:17natively available for me. I can drill into

102:22the same filter context in the service

102:26workflow. There you have it. Service

102:31Intelligence breaks

102:32the data silos, brings analytics to every

102:36Salesforce users in their flow of work,

102:40supercharging data with AI in the world's

102:43best CRM. Back to you, Francois.

102:45[APPLAUSE]

102:53[MUSIC PLAYING]

103:03[APPLAUSE]

103:13[MUSIC PLAYING]

103:23[APPLAUSE]

104:43As we've mentioned, data is for everybody.

104:46You might even say that data is a team

104:48sport.

104:48And speaking of sports, I am a huge

104:51baseball fanatic. I played softball my

104:54entire life,

104:55as you see by my adorable cute photo here.

104:57And when growing up in Michigan is a way to

105:00stay connected back to my family when moved

105:02to California, I participated in three

105:05fantasy

105:06baseball leagues with my cousins, my high

105:08school buddies, and my college friends. Now

105:11imagine,

105:12all of that data that was available to me

105:15back then through tools and online, I wish

105:18that I had

105:18Tableau to help me. And that's why I am so

105:22excited to share this inspiring story of

105:26the Texas Rangers.

105:27They are an organization that has truly

105:30transformed their company with data both on

105:34and off the field.

105:35Let's go ahead and take a look.

105:37[VIDEO PLAYBACK]

105:40- Cool. In the past, I've gotten in trouble

105:43of playing these in my videos. So I

105:47encourage you

105:48to go to Salesforce Plus website and watch

105:50these yourself. So I'm going to skip ahead,

105:53actually to the customers just finishing

105:56talking. Because essentially, when

105:59customers agree to talk

106:00to Tableau, they don't specifically agree

106:03for people like me to critique them on the

106:06YouTube

106:06channel. I've done that in the past. And

106:08yeah, I just decided not to do that this

106:10time around.

106:11So we're going to skip ahead. And I think

106:13it's actually a really cool story about how

106:16they've used data to get through a tough

106:18patch. And then we get the customer coming

106:22on stage to

106:23talk a little bit about it. So I'm going to

106:25keep going until just after the customer

106:28finishes.

106:29I think the customer gives Emacs a hoodie,

106:31a team shirt, a team player shirt. We'll

106:35see this moment

106:35very soon. It's actually a pretty good

106:37conversation. So it's pretty awesome. But

106:40then Larissa comes

106:41on. I think they get a t-shirt here or

106:44something. So I'll actually play from here

106:46on because I know

106:47the conversation's done. And we'll go from

106:49there. - Excited. It's not the team team.

106:52It's actually

106:53the business analytics team. - Oh my god.

106:55[APPLAUSE]

106:58Thank you. I love that. - The signboard

107:00from the analytics team.

107:01This is pretty nice. - Thank you so much.

107:02[APPLAUSE]

107:05Michelle and I have become fast friends in

107:07the last month. So now I'm glad I am a

107:10Rangers fan.

107:11Now, Tableau also has fans. And that is our

107:15DataFam. So to introduce you to them,

107:17I'd love to bring up to the stage VP of

107:19community Larissa Amoroso.

107:21[APPLAUSE]

107:27I'm jealous. The Tableau community, also

107:30known as the DataFam, is a global network

107:32of more than

107:33three million people who push the

107:35boundaries of our products, champion data

107:38culture,

107:39and help people everywhere see and

107:40understand their data. But rather than just

107:44talk about

107:44the incredible DataFam, I thought I would

107:46try something a little different and show

107:49you.

107:49Here at Tableau, we're always experimenting

107:52with new and exciting ways to explore our

107:55data.

107:56- Yeah. - Like with Tableau gestures,

107:58which allows you to present any data on any

108:01computer with any webcam.

108:04Let's say you want to join the Tableau

108:09community. - So if you've not seen this

108:13before,

108:13this demo is called augmented teranomy or

108:17something like that. Anyway, it's augmented

108:21reality sort of take on Tableau with a

108:22slightly different interface. Well,

108:24completely different

108:25interface that's controlled with just

108:27gestures essentially. And it's actually

108:30been in several

108:31demos. It's won bake-offs. It's been a

108:33pretty interesting feature that Tableau

108:35have been

108:36showcasing a lot the last year as a

108:37showcase of innovation, basically. Where

108:40Tableau is thinking,

108:41it's not going to be available. There's no

108:43release date. If it was going to be

108:44released,

108:45it'll probably be next year or a year after

108:47that. There's a lot of things to sort of

108:49work out with

108:49this. And it's definitely more of a

108:51presentation tool, and I really, really

108:52like it. I like it as

108:54a concept. I'd love to know what Tableau

108:55think of this. Now you've got technology

108:57like the

108:57Apple Vision Pro available. Not that you're

109:00going to be doing this in Apple Vision Pro

109:02sat there

109:02with a headset on your face. But I do think

109:04it's kind of an interesting thing that

109:06Tableau have

109:07this innovation here with gestures and then

109:09Apple's coming out with technology which

109:11supports gestures.

109:12It feels like there's something there that

109:14's sort of interesting to explore. Anyway,

109:16it's a really

109:18cool demo. I've done a video on it. Go

109:19ahead and check it out. I'll put it up on

109:21screen. Go ahead

109:22and find that. Larissa does a very short

109:24take on it here, which is interesting. So

109:26let's take a look.

109:27Biggest community is here in the United

109:30States. But what about the rest of the

109:32world? We can

109:33clearly see the top 25 countries listed at

109:36the bottom ranked by number of community

109:39leaders.

109:40Tableau visionaries, ambassadors, and user

109:43group leaders around the world

109:47are hosting local meetups, leading

109:49visualization challenges, and answering

109:52tough data questions

109:53to help you no matter where you are on your

109:55analytics journey. That's right. We have

110:00community leaders all over the world ready

110:03to welcome you. Now I could play with that

110:06all day,

110:07but I am so excited because today we have a

110:10very special guest tuning in live from his

110:14home in the

110:14UK, Tableau visionary Tim Nwenna. Oh, God.

110:20He is. Hi there, fam. You almost need no

110:30introduction.

110:31I wanted to stop it here because this is

110:34the inception moment where I'm in the

110:37keynote

110:38whilst doing a reaction video. I still

110:40laugh at this because it's absolutely crazy

110:42, honestly.

110:44So at this point, I'm not going to continue

110:46watching myself in this keynote. The

110:50context

110:51here is that initially, it's funny. I

110:53should have known what was coming, but

110:55initially, I actually

110:56didn't put two or two together until the

110:59event. But I'd started talking to Tableau,

111:03I think, two,

111:03three months ago actually about doing

111:06something at Dreamforce. And I just thought

111:08, hey, as they have

111:10done with customers, they wanted me to talk

111:12about AI and how it could help data on this

111:15,

111:15specifically around skills and education.

111:17And actually, this is what we go on talking

111:19about.

111:19And I'll talk a bit more about my response

111:21to the questions I get asked here. But

111:23anyway,

111:23it ended with a surprise. And yeah, it's

111:26been a pretty incredible response, frankly,

111:30from the community. I'm absolutely hugely

111:33grateful and genuinely shocked at the time.

111:36And I think

111:37afterwards, it's sort of weird, like the

111:40life of a creator is completely weird.

111:42Because, you know,

111:44I signed my room here in a virtual event.

111:47And I'm actually in the keynote, we're live

111:50,

111:50nothing was pre recorded, there was

111:52incredible amounts of lag. And so a bit of

111:54inside baseball.

111:55As soon as I responded the first time to

111:58Larissa, I immediately knew how much lag

112:01there was. Because

112:02I could hear the feedback from the, would

112:06you call it from the audience from the room

112:09through her microphone. So as soon as I

112:11said the first thing, and then it came back

112:13to me two

112:14seconds, I was like, Oh, crap, I need to

112:16listen to what Larissa is saying. And as

112:18soon as I think

112:19she's about to finish, start talking

112:21immediately. So the lag is like halved

112:23almost. And that's

112:24essentially what I did for the whole thing.

112:26So every time I responded, I was actually

112:28trying

112:28to do that just to cut the lag. But these

112:30are the kind of things that go through my

112:32head here, like

112:32rather than worrying or stressing about

112:35what I'm going to say. That's sort of what

112:37was going through

112:38my mind. I was kind of stressing a whole

112:40time actually start sweating in this video.

112:42So embarrassing.

112:44I'm calling it out now because I can it's

112:45afterwards. But nonetheless, yeah, I really

112:48enjoyed this sort of Holland's beep. Anyway

112:50, I'm gonna, I'm gonna wait and maybe give a

112:52little bit

112:53more color to each of my responses in this

112:55because I think in this section, I have to

112:57give short,

112:57sharp responses. I did know what questions

112:59she was going to ask me. I didn't know what

113:01was coming at

113:02the end. But I wanted to give more context

113:04here because I think it's important. And it

113:06's actually,

113:06you know, sat here and critique the whole

113:08entire keynote. I guess I have to critique

113:11myself. So

113:12that's, let's take this. Let's take this a

113:13little bit further. I might skip

113:15a few. He is an analytic, it's immediately

113:19cringy. Let's skip ahead a little bit here.

113:25Let's double through this. Yeah. With all

113:28of his efforts, he is helping people

113:29everywhere.

113:29Okay, let's thank you so much for joining

113:32us today.

113:32I absolutely the pleasure. So there you go.

113:37That's when I noticed.

113:38How did you get from your very first this

113:42to where you are now in your career?

113:45So it started out in student analytics, I

113:49was looking at student data at the

113:51University of

113:51York where I studied and I ended up going

113:54to another opportunity to work in marketing

113:57and communications and the data that just

114:00sort of pulled me in. The problem I had

114:02though is that

114:03the stories that were coming from that data

114:05one as compelling and a lot of the social

114:08media data.

114:08So another thing I'm actually quite

114:10impressed by is we're using zoom here. Like

114:13I'm patched into

114:14conference by zoom, and zoom held up like

114:16it was solid, the sound was good, the video

114:19quality was

114:20coming through totally fine. I think zoom

114:22compresses the video down to 720p. I had

114:25like

114:25a proper camera going which could have

114:27pushed a 4k stream to this, but I think

114:291080p would have been

114:30fine anyway. And the kind of thing I'm

114:32interested after this is this actually

114:34works pretty well. And

114:35I think it's a good example of hey, if you

114:37can't get someone to the keynote, if they

114:39've got the

114:40setup, you can do a virtual setup. And that

114:42was always sort of my pitch to tablet, hey,

114:44I can't

114:44be there in person. But I think I've got a

114:46good enough setup to do this this way. And

114:48I think it

114:48worked out. But anyway, really, really cool

114:51. Now the context of this answer is about

114:53how I actually

114:54started out in analytics. In essence, I use

114:56tablet for the first time without realizing

114:59it, I was

114:59understanding data about postgraduate

115:02students, because I worked briefly as a

115:05student union

115:06president, essentially student politics

115:08here in the UK. And the super interesting

115:11there was that

115:11the university I went to University of York

115:13was one of the early adopters of tablet,

115:15they were

115:15using tablet to visualize student metrics,

115:18and they were sharing it on their website

115:20through like

115:20an embedded sort of setup. So you could

115:22have this little tablet dashboard that

115:24people could explore,

115:26and you could actually download your data

115:27from it and do various things. This is very

115:29early days of

115:30tablet. So I actually did that I downloaded

115:33the data off a chart, and then repivoted it

115:35and did

115:36some stuff in Excel to then go and try and

115:37visualize and understand what was going on

115:39with

115:39postgraduate students. Anyway, I didn't

115:41know it at the time. But that was tablet.

115:43That was basically

115:44what I was using. And it's only a few years

115:46later where I, I then came across tablet

115:48again, like two,

115:49three years later came across tablet, this

115:51time is like a professional. And I realized

115:53, hey, I've used

115:54this before I use it when I was looking at

115:56student politics, but now I'm using the

115:58authoring experience.

116:00And so that's actually sort of how it

116:01started. And I got to that experience

116:03through someone I met at

116:04university, the information lab, so Craig

116:07Bloodworth, and he encouraged me to say,

116:09hey,

116:09come come join the small companies called

116:11the information lab. You'll learn a lot

116:13more about

116:14data than working in marketing and

116:15communication. So that's what I did. I

116:17joined back then when it

116:18was a company of 11 people. And then fast

116:21forward a decade later, information labs

116:24grown, I've since

116:25moved on from the information lab, I've

116:27worked at Accenture, now work at Endpoint

116:28Digital. And it's

116:30super interesting, just to see that journey

116:33. And I think I go on to explain more about

116:36that. But in

116:36essence, it started off with sort of my

116:39passion for quantified self, which is what

116:41really sort of

116:42made things connect in my mind, it made me

116:44really understand that I was getting

116:45passionate about

116:46data, my own data in a specific way. And

116:48actually, businesses had the same passion

116:51with their own

116:52data, people in businesses really

116:53understood their businesses, as well as I

116:55understood my music data

116:56as well as I understood my running data.

116:59And so in talking to people and talking to

117:02people about

117:02their data, I started to see some

117:04challenges, things that weren't quite

117:05clicking concepts that

117:06weren't working in Tableau. And so what I

117:08did way back when, if you go to the oldest

117:10videos on this

117:11channel, you'll see the first ones about

117:13layout containers, you'll see some others

117:15about Tableau

117:1510 and design, I just thought, hey, let me

117:17make some videos just highlighting these

117:19things so that

117:20I can point people to them and see what

117:22they're like. I started and I stopped and I

117:25gave up,

117:25basically, I did it for like three months,

117:27and then I gave up, I just did the classic

117:29sort of

117:30defeated setup. And what was what was super

117:32interesting about that is that having done

117:34that,

117:36we then, you know, I just sort of carried

117:38on for another two years, went to Accenture

117:41.

117:42And at Accenture, I came across this

117:43problem again, but this time is with it was

117:45with younger

117:46data analysts, people who were just

117:47starting out as associates at Accenture,

117:50and they're just

117:51starting a career and asking me, hey, how

117:52do I use Tableau? How do I do this? How do

117:53I do this?

117:54And it was really easy to explain to them

117:56what was going on. But I just ended up most

117:58of the time

117:59just getting on a call and just showing

118:00them. And so I thought, huh, what if I just

118:03record videos

118:03instead and then send them the link that

118:05will be much faster. So that's what I

118:07started doing.

118:08And then I gave up again. And then right

118:11before COVID, there was an opportunity that

118:14came up to

118:14go to New York for a reason I won't go into

118:17. And it didn't play through because

118:20something happened

118:21in my life that changed sort of the outcome

118:22of that. So instead of going to New York,

118:23I ended up staying here in the UK. And

118:26after that, I was like really bummed out

118:29that I wasn't

118:29going to New York. So I thought, you know

118:31what, God damn it. Honestly, seriously,

118:33what can I do?

118:34Is there a way I can do what I was going to

118:36do in New York without, you know, without

118:40sort of

118:40changing something? Could I approach this

118:42concept in a big way? Can I make a bigger

118:44impact doing

118:45something else? I look back on my videos

118:46and I thought, you know what, actually, I

118:48can. I can

118:49go back to that concept of videos and start

118:51making more. So I pledged to make three

118:53videos. I pledged

118:54to make a video explaining what Tableau is.

118:56And I pledged to start making videos about

118:58what's new

118:58in Tableau. So the new in Tableau videos

119:00are what came first. Four months later, you

119:03saw the what is

119:04Tableau video that was basically planned

119:06like before that video was released. Well

119:08before.

119:09And yeah, here we are. What at 55,000 subs

119:13later, and you know, many more thousands

119:16watching every

119:17single week, every month. We're hitting

119:20milestones and that's supporting lots of

119:22people on LinkedIn,

119:23essentially taking the same concept, just

119:26scaling it up and explaining to people what

119:28Tableau is.

119:29And helping people understand how to use it

119:31and how to work with data fundamentally.

119:33And so

119:33that's what this conversation was actually

119:36sort of crunched down from. That was sort

119:37of the full

119:38context, but it was synthesized just to fit

119:40in this sort of three-minute segment in the

119:42key.

119:42That's why I thought that was useful

119:44context nonetheless. And that's sort of

119:46what we spoke

119:47about. And I'm not sure I'm now willing to

119:49listen to myself go through any of that in

119:51like a third of the time, but nonetheless,

119:54yeah, that's what it is. So let's, let's

119:56quickly,

119:57I might double speed through this and just

120:00go right past the end. So let's watch this

120:02quickly.

120:02- To the time just measured likes and these

120:05sort of basic metrics. And so I ended up

120:06finding a

120:07route into analytics. And when I started to

120:08work with data, I started to realize that

120:09businesses

120:09were super passionate about their data. But

120:11the moment it clicked was when I started

120:12looking at

120:12quantified self data. That's the kind of

120:13data Ryan was talking about data from Strav

120:14a, Last of M and

120:15so on and so forth. And it's only then I

120:16realized that the passion I had for my own

120:17data was the

120:18passion that businesses had for their data.

120:20And so I started to spot some common themes

120:22that people were struggling with as they

120:23were working with their data. And I thought

120:25I'd

120:25start to make a visual way of sort of

120:27helping them understand those problems and

120:30get past them.

120:31So that's, that's how it started. And then

120:32here we are today.

120:33- Such a story journey. Now, many people

120:37here are.

120:38- I think Larissa also understood the lag

120:40was quite big because she cut in faster

120:43than I think

120:44it would have taken for that. So I think we

120:45were both doing this thing where we were

120:47kind of,

120:47she knew what I was going to say generally

120:49speaking. So she knew when I was coming to

120:51the

120:51end of my point. So she could just actually

120:54cut in and I could do the same as well.

120:56Cause I kind of

120:56knew the general question she was going to

120:58ask me. So we kind of did some great

121:00teamwork here to kind

121:02of make it work with less lag than the

121:03Atlanta ocean actually allowed for. So that

121:06's pretty funny.

121:07- Are still, you know, just getting started

121:09on their analytics journey.

121:11And it's a little bit of a different

121:13landscape today. What has you most excited

121:16as you think

121:17about Tableau and this new AI revolution? -

121:20I think AI has this incredible opportunity

121:24to

121:25amplify people who are already

121:27exceptionally skilled, sort of raise their

121:30skills up to the

121:30ceiling, but also for people who are

121:32struggling to get into these topics or

121:34struggling to pick

121:35up these skills, it actually has an ability

121:37to lower the barriers, almost help them get

121:40into

121:40these topics. And so it can simplify the

121:42entry points for a particular topic. We saw

121:45Hunter do

121:45a nice demo there, but also I think it's

121:47got this ability to help people with their

121:50skills. It kind

121:50of brings them into a topic so they can

121:52understand what to Google. - So what's

121:54going through my mind

121:55right now is like, this is just stressful.

121:59Like I make videos all the time and it's no

122:02different to

122:03the videos except for this is live, right?

122:05And I'm not going to a script either. I'm

122:08not reading

122:08like a teleprompter. I could have, I could,

122:10I actually thought off the event, why didn

122:11't I just

122:12set this up as a teleprompter? And I wouldn

122:14't have been stressing myself out trying to

122:16make sure I

122:17thread the point I was trying to make in

122:19less sort of compressed time, rather just w

122:21affle on like I

122:21do here and like I do in most videos

122:23actually. And so you can see on my face, I

122:26'm literally

122:27stressing. If you see like sort of the sort

122:30of silver lining, it's so bad. It's

122:33honestly

122:33embarrassing, but the room wasn't even that

122:35hot. It's just one of these things where

122:37you're

122:38panicking. Cause you know, that's just like

122:40how the human body works, right? And I just

122:43suddenly

122:43just got really, really hot. And then yeah,

122:47just started panicking, I guess. So I

122:50started

122:50sweating my head off. Oh my god. And then

122:52how to build on those skills.

122:52Now with your platform, you're bringing

122:55data skills to so many people.

122:57How, um, what is something that you can

122:59share with us about teaching and giving

123:01back?

123:01It's um, it's a super interesting. That

123:05meme of the guy sweating profusely comes to

123:07my face at the

123:08moment. I think one of the things that I

123:11always tell people is that when I make

123:13videos, I learn

123:15a lot more about the topic than I would

123:16have done if I just went to learn about the

123:18topic. So

123:19teaching kind of makes you understand the

123:21topic to another level. And when you put

123:23that content out

123:24there into the community, you get sort of

123:26responses. People ask you more questions

123:27and

123:28actually those questions are the best

123:29questions to answer because they enhance

123:30your understanding.

123:31So it's this sort of nice feedback leap

123:33that just sort of happens. And what is nice

123:36about that is

123:37that, you know, I got my sort of break in

123:39data through the data farm and it's just

123:41been really

123:41special to sort of give it back as well. So

123:43hopefully that cycle continues and we keep

123:45building the data farm. Well, speaking of

123:47giving back, you've done. At this point, I

123:50am just

123:51absolutely, the sweat is fully formed. I

123:53think if we'd gone on for another two

123:55minutes, you would

123:56have seen one drip down my head. Honestly,

123:58I don't know why I'm tearing myself apart

124:00here, but it's

124:00just so funny. I don't know if you've ever

124:03been on stage in front of people and had

124:04this feeling,

124:05you know exactly what I'm talking about.

124:07But anyway, here we go.

124:08So much for Tableau and for the data fam.

124:12Before I let you go, I have one more thing

124:15I'd like to

124:15share. It's a golden hoodie and we have one

124:18in the mail on its way to you now. Tim, you

124:25truly embody

124:26what it means to be a Tableau community

124:28leader and a visionary. Thank you so much

124:32for everything

124:32that you've done. Congratulations again.

124:35And thank you so much for joining us today.

124:37Let's give them

124:38a round of applause. Thank you. Wow. What a

124:47magical moment. I wish he would have been

124:49here, but this

124:49was fun too. Back in 2022, we made a pledge

124:55to all of you to bring more foundational

124:59skills to

125:00everyone. I am so proud to say that we are

125:03on our way to enabling 10 million people by

125:062027. We hope

125:08you'll join us on this journey. And with

125:09that, I'm going to hand it back over to

125:11Ryan. Thank you.

125:13Great job. So I want to say thank you. I

125:17know we've gone a little bit over, but I

125:20also want

125:20to remind you, we are your guide here at

125:22Tableau and Salesforce on your data journey

125:25. There's a lot

125:26to learn out there. And I want to make sure

125:29that you also know. Clearly, we talk about

125:31data for

125:32everyone. We've covered the business user,

125:33we've covered the analysts and the Sales

125:35force CRM user.

125:36Honestly, I'm actually sort of blown away.

125:38Check us out. We're out here. We've got

125:40Tableau Conference also coming up in San

125:42Diego.

125:43It's pretty much the end of the conference.

125:46I'm going to try and dig into some of the

125:49content from conference itself once I can

125:51find out how to get access to what was

125:53recorded and what

125:54wasn't. I think there's lots of useful

125:56context. And once a conference is normally

125:59over, you tend

126:01to get more blogs and posts about it as

126:03well released in the Tableau blog. So as

126:05soon as that

126:05happens, I'm going to try and get access to

126:08some of that. And to give you some context

126:10behind the

126:11scenes, I've actually started to get some

126:13support and help to help make more content

126:16more quickly.

126:17One of the biggest bottlenecks in just all

126:19the videos I make is actually me being able

126:21to record,

126:22edit, rate, script, animate, draw. It's all

126:26been a one-person team all the way now.

126:29Everything you

126:30see literally goes through this desk. And

126:33what that means is some of the things I do

126:35aren't as

126:36best as they can be because if you're split

126:37100 different ways, well, you're not doing

126:39any one

126:40thing exceptionally well. So part of the

126:42effort is going into getting some of the

126:45other stuff,

126:46so the editing, the planning of content,

126:48making sure that it's all formed correctly,

126:50the checking of mistakes. I make mistakes

126:52all the time. Thankfully, you're all on it

126:55with comments,

126:55and I get emails from Tableau. I get emails

126:58from customers as well telling me where I'm

127:00going wrong. I'd love to spot more of those

127:03upfront. So I'm definitely in the process

127:06of

127:06getting help. I've actually already started

127:08to get some help. I'll talk more about that

127:09in the

127:10future once it's appropriate to do so. But

127:13that will help me stay on top of how fast

127:16this platform

127:16is moving. I fully intend to do that. One

127:19of the biggest regrets I have at the moment

127:21is not

127:21knowing more about Salesforce. So I can

127:23cover these overlaps between Tableau and

127:25Salesforce

127:26and be a conduit for people to get into

127:28that. So definitely something I'll do. Be a

127:31little bit slow

127:32initially, but we'll eventually pick up

127:33pace, and we'll find a way of making it

127:35work. San Diego is

127:37the next conference for Tableau, so April

127:4029th. It is just the Tableau conference in

127:43California,

127:44so that's going to be a really, really nice

127:46touch. 29th through May 1st, 2024. I think

127:48that makes it

127:49three days, which is super interesting. I

127:51think it's a Monday to Wednesday.

127:53And yeah, there's a couple of other

127:56experiences here on the slide related to

128:02Salesforce. So

128:03that's pretty cool. But yeah, look, that's

128:05pretty much the end of this. We'll cut it

128:06short there,

128:08and I'll take an ending there. If you're

128:10wondering when am I going to wear this hood

128:13ie,

128:13well, the hoodie has a separate story. If

128:16you want to know the full story behind the

128:19hoodie,

128:19watch my next video probably after this one

128:22, which will hopefully be with the hoodie

128:25and involve the

128:26hoodie. So let's wait and see, but I'll

128:28definitely do a separate video on that for

128:31reasons you'll

128:31find out for in the next video. Anyway,

128:33thanks for watching, and I'll catch you in

128:35the next one.

128:36[00;03;29] [END OF AUDIO]

128:37[00;03;45]

128:38[END OF AUDIO]

128:39[00;03;55]

128:40[END OF AUDIO]

128:41[00;03;59]

128:42[END OF AUDIO]

128:43[00;04;00]

128:44[END OF AUDIO]

128:45[00;04;00]

128:46[END OF AUDIO]

128:47[ Silence ]