Video | Tableau | Data visualisation | Tool strategy

Tableau 2020.2: High viz data quality warnings on Server & Online

Let me show you how Tableau 2020.2's high visibility data quality warnings put data issues right in front of your users, in context.

Part ofWhat's new in Tableau 2020.2Tableau Server
Watch on YouTube
  • High visibility data quality warnings are set on a published data source by an owner or administrator via the quality warning option, with up to 4,000 characters of description
  • Once saved, the warning appears as a toast notification plus a persistent indicator beside the data source, showing the message, date and author
  • When opening a view or dashboard built on that data source, users get an immediate notification that important data quality warnings affect the view
  • The warning detail sits in context within the view via the data details pane, which you can hide or reopen as needed
  • This improves transparency about issues such as stale data or problem fields, which matters in fast-changing business environments

Tableau 2020.2 lets a data source owner attach a high visibility data quality warning that then surfaces directly inside any view or dashboard built on that data source, so users see data issues in context rather than having to go hunting for them.

The demo uses a published data source linked to a Salesforce table, feeding a dashboard where a viewer has no way of knowing about underlying data issues unless told explicitly.

The Breakdown
  1. Check where warnings normally hide 0:42

    Before this feature, you'd have to manually check the data source tab or the lineage view (part of data management) to spot historical issues — most users never think to look there.

  2. Enable a quality warning on the data source 1:22

    As the owner or an administrator, tick the quality warning option, choose a warning type, and set it as a standard warning marked highly visible. You must do this at the data source level before it can appear anywhere else.

  3. Write the warning description 1:43

    Type a description explaining the issue — you get up to 4,000 characters, so you can give real detail rather than a vague note, since users will actually read this.

  4. Save and see the immediate confirmation 2:03

    Saving triggers a toast notification and adds a persistent warning indicator next to the data source, showing the message, date and author for anyone checking the source directly.

  5. Warning surfaces automatically in views 2:21

    Opening the data source's warning panel gives full context; the real payoff is that any dashboard built on that data source will inherit the warning automatically, with no extra setup per view.

  6. Users get an in-context alert on open 2:40

    When a user opens a dashboard using the flagged data source, they immediately see a notification that data quality warnings affect the view, with details available in the data details pane on the right.

  7. Dismiss or recall the warning as needed 3:41

    You can close the data details pane with the x and reopen it any time via the warning icon, so the alert doesn't have to stay on screen but is always retrievable.

Worth Knowing
  • The warning is set once on the data source, not per-dashboard, so it propagates to every view built on that source without extra configuration.
  • This relies on Tableau Server or Online — it's about how warnings surface to viewers of published content.
Use It When

Reach for this when your data can go stale, break, or have known issues (e.g. delayed refreshes or problem calculations) and you need viewers to know that in the moment they're looking at the dashboard, not buried in a separate data source page.

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.