Video | Tableau | Data visualisation | Analytics

The Total function in Tableau & partitions

Find out why the total function shouldn't be considered as a mathematical total.

Part ofTableau Functions
Watch on YouTube
  • TOTAL computes a single value across the whole partition and then repeats that value in every cell, so it isn't the same as SUM.
  • A partition defines the direction Tableau looks when calculating; table down looks at the entire table while pane down looks at each individual pane.
  • Adding more pills to the rows shelf shrinks the definition of a pane, which changes what TOTAL returns.
  • TOTAL(AVG([Sales])) returns the average of every underlying record in the partition, not the average of the visible aggregated values, which is a common source of confusion.
  • Think about what the function actually does rather than its name, since TOTAL works with any expression including averages.

The TOTAL function doesn't do a mathematical total - it computes a single value across a defined partition and repeats that value in every cell, so to use it properly you need to understand what a partition actually is.

Tim builds a simple table (subcategory rows, sales as text) and wraps TOTAL(SUM([Sales])) around it to compare against the plain SUM already on the view.

The Breakdown
  1. Build a simple aggregated table 0:14

    Start with a basic view - a dimension on rows and a measure summarised as text - so you have a plain aggregate (like SUM) to compare TOTAL against.

  2. Wrap TOTAL around an expression 0:54

    TOTAL takes an expression, such as SUM([Sales]), and returns one value for the whole partition, then repeats that same value in every row - it isn't adding up what's already on the screen the way SUM does.

  3. A partition is the direction TOTAL looks 2:13

    By default the compute-using setting is table down, so TOTAL looks at the entire table, calculates one number, and stamps it into every cell - this is why the value stays identical even when you rearrange dimensions that don't change the partition.

  4. Switch compute-using to pane down 3:26

    Changing the pill's compute-using setting to pane down redefines the partition as each individual pane rather than the whole table, and TOTAL immediately recalculates to match just that pane - this is a required setting change, not a tip, if you want per-pane totals instead of a single grand total.

  5. Adding pills to rows shrinks the pane 5:19

    Once you have a hierarchy on rows, each combination of pills defines a pane; adding another pill (for example a date field) makes panes smaller and more specific, which changes what pane down totals actually cover. Use the summary pane (worksheet options) to verify these totals against a manual selection.

  6. TOTAL(AVG()) still scans every record 7:28

    Swapping the inner expression to AVG([Sales]) doesn't average the aggregated values you can see - it goes back to the underlying row-level data for the whole partition and averages every individual record, which usually gives a different number to what you'd expect from averaging the visible subtotals.

  7. Excluding a category changes the record-level average 9:22

    Because TOTAL(AVG()) recalculates from raw records, excluding a low-performing category from the partition shifts the average noticeably - a useful way to prove to yourself that it isn't working off the visible aggregates.

  8. Think about behaviour, not the name 10:21

    Treat TOTAL as "apply this expression across whatever partition is defined" rather than as a synonym for summing - that mental model works whether the inner expression is SUM, AVG or anything else.

Worth Knowing
  • TOTAL(AVG()) is a common source of confusion because it averages every underlying record in the partition, not the visible aggregated values you see on screen.
  • Compute-using options beyond table down and pane down (pane across, pane down then across, etc.) get confusing fast - Tim flags this as a separate topic he plans to cover on its own.
  • Total and partitions intersect with table calculation direction and performance optimisation, which is covered by wider Tableau community discussion rather than in this video.
Use It When

Reach for TOTAL when you need a single reference value - a grand total or a pane-level total - repeated alongside your regular aggregates for comparison, but only once you're clear on which partition (table vs pane) you want it computed against.

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.