Video |

Tableau | Data visualisation | AI & ML

Ask data date range enhancements

You just typed and talked to Tableau to build a chart — how cool is that?

Tim Ngwena •15 December 2020 •1 min read •5 min watch •1K

Watch on YouTube

Ask Data lets you type a plain-language question into Tableau and returns a table or chart in response
Build questions in stages so you can confirm Ask Data interprets each step correctly before moving on
The new date range slider lets you tweak start and end dates directly in the view after building a chart
You can swap chart types, dimensions and orientation after the fact, and either click options or type them in
Investing time in metadata such as synonyms, meanings and data roles makes Ask Data work well on your own datasets, which you can then embed for others to query

0:00Hey, it's Tim here and in today's video I'm

0:02here to talk to you about Ask Data.

0:03Now, the feature I'm about to talk today, I

0:06'm actually quite impressed.

0:07I haven't used Ask Data in a while myself.

0:09I know I do these videos showing you all

0:10the features,

0:11but there's something different about Ask

0:13Data now.

0:14And I think it's actually quite interesting

0:15.

0:15Let me just show you how it works.

0:17If you're not aware, Ask Data is a new

0:18feature where you can essentially just

0:20type a question into Tableau and it gives

0:22you the answer.

0:23To show you that, I'm here on Tableau

0:24Online.

0:25I'm going to go to the Explore tab here,

0:26go to Samples and there's of course a

0:28sample Superstore dataset that you can use.

0:31Now, what's required here is when you

0:33upload a data source,

0:34Tableau will start analysing it.

0:36And this is a fairly sanitised dataset.

0:39Of course, it's a Superstore.

0:40Tableau have done some good work to make

0:42this nice and optimised.

0:43But when you upload your own dataset,

0:45you have to invest a little bit of time

0:47making this dataset sort of work in this

0:49platform.

0:50So things like synonyms, meanings, data

0:53roles all have to be applied to Ask Data.

0:56Check out my channel for features on Ask

0:59Data.

0:59I've covered pretty much everything since

1:012020.3.

1:03So let's go ahead and check out how this

1:05works.

1:06So the first thing I'm going to do is just

1:08type in a question.

1:08I want to see the profit by category.

1:11OK, I can't type so I have to make sure

1:14that I use the correct category.

1:18Here we go. And it will come up with a

1:20suggestion.

1:20I'll click the suggestion so that it cleans

1:22up my typos, as it were.

1:24And now it'll open up a new window and it

1:25'll actually go ahead and load that, which

1:27is great.

1:28This is kind of what I wanted, but it's not

1:30quite the question I had.

1:32What I'd recommend you do with Ask Data is

1:34you ask your question in stages

1:35because then you know it's actually getting

1:37the question correct each time.

1:39So the next thing we're going to do is

1:41actually make sure that we see this by

1:43month.

1:43I want to see this as a time series,

1:45essentially.

1:46So by month, let's just type that.

1:49And when we do that, it obviously breaks

1:50this down.

1:51Now, the thing I'd like to do is actually

1:53then choose a date range.

1:55OK, so I want to say between March 2015 and

2:00December 2019.

2:03OK, so that's my little window here.

2:05I've done a typo here, so let's go hit

2:07December there.

2:08Let's fix it. March.

2:10I'm going to give it the benefit of the

2:11doubt here and just enter.

2:13Now in an organization, you'd have typos.

2:15And so here we have it. We have this nice

2:17sort of timeline.

2:18Nice new feature here is you can actually

2:20tweak this date range a little bit better.

2:22So you can actually go in here just by

2:23clicking that and actually tweak it.

2:25So let's say you got the year wrong and you

2:26wanted to go to 2019.

2:28Hit accept and it changes the view. This is

2:30really, really cool.

2:31And of course, you can change this to

2:33whatever chart type you want.

2:35Just go over here on the right hand side

2:36and hit stacked bar.

2:38And you can see that it comes up actually

2:40is a nice option.

2:41You can actually type it in here.

2:42So let's just remove that.

2:44It goes back to the line chart and just

2:46type it ourselves as stacked,

2:48stacked bar chart.

2:52And you can see that comes up there.

2:54And of course, you basically typed and

2:56talked to Tableau.

2:56How cool is that?

2:58So that's a great little addition there.

2:59And of course, you can change any of these

3:01parameters once you've built this.

3:02And you can go back to any particular

3:04question and change the particular thing.

3:06Maybe you didn't want it by category.

3:08Maybe you wanted it by country instead.

3:10Hit accept. Boom. You've just got a new

3:12chart right there.

3:13And then, of course, this is all the United

3:14States.

3:15So country made no sense.

3:17We should have chosen state or something

3:18instead.

3:19Hit accept. And instead you get a very,

3:22very noisy chart.

3:23And of course, it's not colored it

3:24correctly because we've changed things.

3:25And then in the middle of the way, we're

3:27asking questions.

3:27But this just gives you a nice little

3:29glimpse of how this is working

3:31and how it's dynamic and how it really

3:33actually makes it easy

3:34to build a new visualization.

3:36If you want to pivot this, you can, of

3:37course, here

3:38you've got the option at the top just to

3:39change the orientation,

3:41which is really nice.

3:42You can then grab this, embed this

3:44somewhere else

3:45and let someone else ask their own

3:47questions.

3:48So you can actually take this

3:49ask data experience on this data source and

3:52put it somewhere

3:52in your organization where people can

3:54actually type these questions

3:55and ask them themselves.

3:57If you've invested a little bit of time

3:59into the metadata,

4:00that's going to be really, really good.

4:01So the actual feature here was this sort of

4:03relative date range slider.

4:05This is new. It makes it much easier to

4:07pick dates.

4:07But just check out how much this feature

4:09has come along

4:10in the time that Tableau has been building

4:12it.

4:12This is supposedly one of the new areas of

4:15analytics

4:15where essentially you're just going to be

4:17talking to a computer

4:18and it will be doing this for you.

4:19And there's actually, you know, I see a

4:21world where this is actually realistic,

4:23where you simply type something into a text

4:25line

4:25and you get a response back with a table or

4:28bar chart

4:28or pretty much any of these chart types

4:30that Tableau have already added.

4:31So that's pretty much the feature in a

4:32nutshell.

4:33If you've enjoyed this video, you know what

4:35to do.

4:35If you haven't hit the dislike button twice

4:37, I'll catch you in the next video.

Rollup

Tableau's Ask Data lets you type plain-language questions and get back a chart, and a new date-range slider makes tweaking time periods far easier than before. It's a fast way to build and refine visualisations without touching the usual drag-and-drop interface.

Demoed on Tableau Online using the sample Superstore dataset, which comes pre-optimised for Ask Data. Working with your own data requires extra setup first.

The Breakdown

Start with a simple question 1:02

Type a basic query such as profit by a given dimension and accept Tableau's cleaned-up suggestion if your phrasing has typos. This gives you a starting chart to refine rather than trying to get the perfect question first time.
Build the question in stages 1:20

Add complexity incrementally — for example, asking to break the data down by month after the first result loads — so you can confirm at each step that Ask Data has interpreted you correctly before adding more.
Specify a date range in your question 1:41

You can name a start and end period directly in the text query, and Ask Data will filter the view to that window.
Fine-tune dates with the new slider 2:20

Once the chart exists, click into the date range control on the view itself to adjust the start or end date without retyping the whole question — useful if you got a date wrong or want to explore nearby ranges.
Change dimensions on an existing chart 3:01

Go back to any part of your built question and swap out a dimension for another; Tableau rebuilds the chart accordingly. Be aware this can produce a noisy or oddly coloured result if the new dimension doesn't suit the chart type, so you may need to keep iterating.
Pivot orientation with one click 3:40

Use the orientation option at the top of the view to flip the chart layout without rebuilding the question.
Invest in metadata before sharing widely 3:41

For your own datasets, set up synonyms, meanings and data roles so Ask Data interprets questions correctly; once done, you can embed the Ask Data experience elsewhere so others can query it themselves.

Worth Knowing

Ask Data only works well out of the box on well-prepared datasets like the sample Superstore — your own data needs synonyms, meanings and data roles configured first.
Typos in your question can still get misread even after Tableau's autocorrect, so check the interpretation at each stage.
Changing a dimension after the chart is built can produce a badly coloured or overly noisy result if it doesn't suit the existing chart type.

Use It When

Reach for this when you want to build or tweak a chart quickly by typing rather than dragging fields, or when you want to let non-technical colleagues query an embedded data source in plain language.

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.