Video |

Tableau | Data prep | Industry trends

Tableau Prep Showcased at Conference - | Announced at Tableau Conference 2023

Tableau Prep is quietly turning into a secret power tool, inching closer to Alteryx with every release.

Tim Ngwena •19 May 2023 •1 min read •5 min watch •1.1K

Part ofTableau Conference 2023 Tableau Prep Builder tutorial Jan-2020

Watch on YouTube

Tableau GPT can write calculations inside Prep, such as extracting an email from a JSON dataset, and the feature works across the wider platform including Desktop.
Prep now supports spatial joins on two spatial datasets, defining overlap or outer regions natively rather than relying on Python or third-party tools.
You can choose which rows contain your headers, solving the long-standing annoyance of headers not sitting on row one of CSVs and flat files.
A smart suggestion feature for removing duplicates was shown, though Tim cautions to test it rigorously since duplicates sometimes exist for legitimate reasons.
Tableau Exchange connectors, such as Amazon S3, can be added to Prep today for free, with a growing list of vendor-built connectors.

0:00Hey, it's Tim here. In this video I want to

0:02cover everything announced at Tableau

0:03Conference for Tableau Prep.

0:05Let's get started.

0:06Okay, so Tableau Prep didn't have as many

0:07blockbuster features as the rest of the

0:09platform, but nonetheless there were a

0:12handful of things to touch on and I think

0:13they're important.

0:14So let's take a look at them in this

0:15particular video.

0:17Now, what I will also say is that a lot of

0:19Tableau Prep capabilities do tend to be

0:21platform wide.

0:22So if you take something like Tableau GPT,

0:24that was platform wide and we saw one

0:26example of how it can work inside of Table

0:28au Prep.

0:29We also have things like Tableau Pulse,

0:30although that is an ability to build

0:32metrics, I wouldn't be surprised if Tableau

0:35Prep has a unique capability to push

0:37certain types of metadata to things like

0:40Tableau Metrics as well and Tableau Pulse.

0:42So that would be another interesting area

0:44to wait and see what specific detail

0:46actually refers to Tableau Prep as we get

0:48our quarterly release cadence.

0:51Nonetheless, the first thing I want to talk

0:53about is obviously Tableau GPT inside of

0:55Tableau Prep.

0:56The example we saw on Devs On Stage was

0:58essentially Matt Miller demo from him where

1:01he essentially asked Tableau to write a

1:03calculation to extract the email from a

1:06JSON dataset inside of Tableau Prep and

1:10that was fantastic.

1:11That is just a simple example and you can

1:13imagine how that calculation feature was

1:15demoed in Tableau Prep but actually works

1:18right across the platform, something like

1:20Tableau Desktop, no reason why that wouldn

1:22't have it as well.

1:24The other area that was featured in the

1:26video was the ability for Tableau Prep to

1:28do joins on spatial data.

1:30Now if you use a tool like Alteryx, this is

1:31a native capability, essentially being able

1:34to take two spatial datasets and being able

1:36to define a join by specifying whether they

1:39overlap, whether you want the outer regions

1:41that don't overlap and doing that in a

1:43spatial way.

1:44Now that is a very sort of unique dataset

1:46and typically people have had to do this

1:48outside of Tableau Prep or they've used

1:50third party tools through the Python

1:52capability that you can run inside of Table

1:54au Prep.

1:55That was announced as well so that was good

1:57to see.

1:58I think Tableau Prep is turning into this

2:00secret power tool because the features keep

2:03coming release by release and if you've

2:05been keeping track of all these features,

2:08it's getting it closer and closer and

2:10closer towards matching some of the basic

2:12capabilities of something like Alteryx

2:15which has been around for years.

2:17I don't know about you but the Alteryx

2:19interface for me is just long overdue a

2:21refresh and a change and Tableau Prep has

2:23really taken a fresh approach to how to

2:25improve that kind of interface but also

2:27make it fun and interesting and applicable

2:30to the Tableau platform.

2:32So that's always a personal sort of thing I

2:34like to champion about Tableau Prep.

2:36It might not have the same power but it's a

2:38damn sight nicer to use even though at some

2:40points it does have quite a few annoying

2:42bugs.

2:43Anyway, let's keep moving.

2:44One of the nicest features was the ability

2:46to choose the rows on which your headers

2:48are located.

2:49Sometimes when you connect to a data source

2:51, the headers are not on row one of that

2:53data source and with Tableau Prep it's been

2:55super annoying because you've had to go and

2:57specify and then rename a bunch of fields

3:00and that's always been some sort of messy

3:01way of doing it.

3:02Maybe bulk pivoting and then unpivoting.

3:04There's so many different approaches to the

3:06same problem depending on your file.

3:08What they showed was a capability that can

3:10automatically do that or specifically an

3:12option where you can go and specify your

3:14specific rows.

3:16Something like named ranges in Excel does

3:18this sort of if you're kind of working with

3:20those but if you're working with CSV and

3:22flat files you don't get that and so this

3:24is a super nice feature to use.

3:26The last two features I'd like to talk

3:27about are the ability to remove duplicates.

3:30That's a very sort of difficult thing to do

3:32correctly and Tableau had a smart

3:34suggestion feature that could do that.

3:37Again you kind of have to wait till you see

3:39this in practice because removing dupl

3:41icates sometimes sounds easy to do but there

3:44are also instances where duplicates exist

3:46for a reason.

3:47Let's say you go to a shop and you buy two

3:49things at exactly the same time the

3:51transaction might look like you've got dupl

3:53icates in there but unless you've got some

3:56sort of unique identifier or some other

3:58sort of information that can tell you

4:00whether it's a duplicate or not.

4:02That's always a tricky thing to do so it's

4:03one of the features that I like to not

4:05trust until I've rigorously tested it and

4:07seen it for myself.

4:09The final thing is I think something that

4:11is growing a lot of momentum which is Table

4:13au Exchange, specifically Tableau Exchange

4:15connectors inside of Tableau Prep.

4:18This enables features and connectors like

4:20Amazon S3 inside of Tableau Prep so that is

4:23a super nice feature.

4:25If you want to check out these connectors I

4:26'll put them up on screen but there is a

4:28growing list more and more vendors are

4:30building connectors for their technologies

4:32inside of the Tableau platform.

4:34So something to check out and something to

4:36be super sort of alert of because you might

4:39just find in a year from now we have twice

4:41the number of connectors and these

4:43connectors can just be used today for free.

4:45They can be easily added to Tableau or

4:46Tableau Prep and it's now great to have

4:48them directly inside of the product.

4:51That's pretty much it for the features, I

4:52've gone through my list of features and I

4:54think that's pretty much everything.

4:56If you've got any questions, any notes

4:57about Tableau Prep drop them in the

4:59comments below I'd be happy to answer them.

5:01Apart from that, thanks for watching and I

5:03'll catch you in the next video.

5:05[BLANK_AUDIO]

Rollup

A round-up of what Tableau Prep got at Tableau Conference 2023: fewer blockbuster features than the wider platform, but several practical ones that push it closer to a tool like Alteryx.

Tim reacts to conference announcements rather than building anything himself, walking through each new Prep capability as it was demoed on stage.

The Breakdown

Prep gets platform-wide features, not just its own 0:06

Many capabilities announced under Prep, like Tableau GPT, actually apply across the whole platform rather than being Prep-exclusive, so features you see demoed in Prep may well show up in Desktop too.
Tableau GPT can write calculations for you 0:51

In the demo, GPT was asked to write a calculation to extract an email address from a JSON dataset inside Prep. Treat this as an example of a general capability: you can describe what you want and have a calculation generated rather than writing it by hand.
Spatial joins now live natively in Prep 1:20

You can join two spatial datasets and define the join by whether they overlap or by their outer, non-overlapping regions, without going outside Prep or relying on the Python integration or third-party tools as you previously had to.
Choosing which row holds your headers 2:40

When headers aren't on row one of a CSV or flat file, you can now specify which row they're actually on instead of manually renaming fields or resorting to pivot/unpivot workarounds. Useful whenever you inherit messy flat files with header rows buried further down.
Smart suggestion for removing duplicates 3:26

Prep can suggest duplicate rows to remove, but treat this as a starting point, not a final answer: duplicate-looking rows can be legitimate (e.g. two identical purchases at the same time), so test the suggestion rigorously against your own data before trusting it.
Tableau Exchange connectors inside Prep 4:09

Vendor-built connectors, such as one for Amazon S3, can now be added into Prep directly, and the list is growing. Worth checking the Exchange periodically since new connectors can be added for free and used immediately.

Worth Knowing

Don't trust the duplicate-removal suggestion blindly — verify it against cases where duplicates are legitimate before relying on it in production.
Tim expects Prep's calculation and metadata features (e.g. anything Tableau GPT related) to extend to Desktop and other parts of the platform, not stay Prep-specific.
He notes Prep still has a number of annoying bugs despite its nicer interface compared with Alteryx.

Use It When

Reach for this update if you're evaluating whether Prep can replace tasks you currently do in Alteryx or manually — spatial joins, awkward header rows and connector access via Exchange are the concrete wins here.

How this Rollup was made provenance & method

A Rollup is drafted by AI from the video's transcript, then reviewed and edited by Tim. Everything used to produce this one is listed below — the model, the exact prompt, and the source video — so the process is transparent and reproducible.

Transcription: On-device — NVIDIA Parakeet v3 for recent videos, OpenAI Whisper large-v3 for earlier ones. The transcript never leaves the machine or gets published.
Drafting: Claude Sonnet 5 in the cloud, from that transcript.
Prompt: The exact Rollup prompt (v2) — the full system prompt, unedited.
Source video: Watch on YouTube
Drafted: 5 July 2026 at 09:38
Reviewed & edited: 5 July 2026 at 09:41 · by Tim Ngwena

Model + prompt + video is everything you'd need to recreate a Rollup like this yourself. The one thing we don't share is the transcript.

Rights. The video and its transcript are the property of TN Media Ltd. Unauthorised use or download is prohibited. © TN Media Ltd.