Video | Tableau | Analytics | Data visualisation

Ask data improvements: New in Tableau 2021.1

A new centralised assistance experience for Ask Data

Part ofWhat's new in Tableau 2021.1
Watch on YouTube
  • The new Ask Data assistance window centralises online help, feedback options and a setup guide, with content targeted separately at everyday users and data owners/admins
  • Usage analytics show which fields people most often query and how the same question is asked in different ways, which is especially valuable for Tableau Online users who can't access the full Postgres database
  • Reviewing the most-queried metrics helps you build standardised data sources with pre-built aggregations so users avoid asking complex questions
  • You can switch off indexing for individual fields to improve query performance and stop Ask Data defaulting to detail levels you'd rather avoid
  • Without an indexed field, Ask Data can't recognise specific domain values (e.g. 'New York City') as context, so indexing is what gives natural-language queries their understanding

Tableau 2021.1 adds a centralised assistance window for Ask Data and the ability to switch off indexing on individual fields, giving both everyday users and data owners better tools to understand and tune natural-language querying.

Demonstrated against a Superstore-style data source with Ask Data enabled, showing what changes for people asking questions and for the admins who maintain the data source behind it.

The Breakdown
  • New assistance window appears on first use 0:09

    When you open a data source with Ask Data enabled, a new assistance window pops up automatically the first time you log in on this release; after that, you can bring it back manually via the information icon.

  • Online help and feedback for everyday users 1:00

    The window includes online help explaining how to phrase queries, since the power of Ask Data isn't obvious from the interface alone, plus a feedback button covering problems, feature requests and praise you can send back to Tableau.

  • Setup guide for data owners 1:38

    A separate section targets admins, covering setup topics like synonyms and which fields are indexed or not — worth working through if you're the one enabling Ask Data for others.

  • Usage analytics reveal how people actually query 2:07

    You can see which fields are queried most and how many different phrasings map to the same underlying visualisation, which is especially useful on Tableau Online where you don't have direct access to the full Postgres database behind the scenes.

  • Use analytics to shape your data sources 2:41

    Reviewing the most-queried metrics lets you build standardised data sources with common aggregations already built in, so users don't need to ask increasingly complex questions to get a simple answer.

  • Switch off indexing on a field 3:29

    Hovering over a field shows its indexed domain values; you can deselect a field to stop it being indexed, which removes it from Ask Data's interpretation without deleting the underlying data, and you can re-index later if needed.

  • Indexing controls detail level and performance 4:42

    Removing a field like city from the index stops questions defaulting to that level of detail and improves query performance, while users can still explicitly reference the field in their questions.

  • Only indexed fields give Ask Data value context 5:40

    Without indexing, Ask Data can build the same chart but won't recognise a specific value like 'New York City' as belonging to that field, so a question naming it won't filter correctly; re-indexing restores that recognition, which only works once the field is indexed again.

Worth Knowing
  • Re-indexing speed depends on environment — near-instant on Tableau Online but potentially slower on an on-premise Tableau Server.
  • Usage analytics may be the only way Tableau Online admins can see question-and-visualisation activity, since they lack full database access.
  • Deselecting indexing doesn't remove the field from the data source, only from Ask Data's interpretation, so users can still type it into a question manually.
Use It When

Reach for this when you're rolling out Ask Data across a team and need to see what people are actually asking, tighten up which fields drive answers, and stop queries defaulting to a level of detail you don't want exposed.

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.