just-tim
Video | Tableau | Data prep | Data visualisation

The Split function in Tableau.

The split function is so important Tableau often does it for you behind the scenes, so let me show you exactly how it writes that calculation.

Tim Ngwena •11 March 2021 •1 min read •9 min watch •Views on YouTube18K
Part ofTableau Functions
Watch on YouTube
  • Right-click a field and choose Transform > Split to let Tableau automatically detect the delimiter and create split fields as calculated fields
  • The split syntax is SPLIT(field, delimiter, token number) where the number targets which instance of the delimiter to extract up to
  • When a delimiter isn't found, split returns the entire original string rather than nulling it out
  • Wrapping a numeric split in INT() makes the resulting calculation more efficient when used elsewhere
  • Use the custom split option to target a delimiter, choose how many columns to extract, and break a single text field into a full hierarchy
Rollup

The split function breaks a delimited text field into separate parts, and Tableau will often write this calculation for you automatically – understanding how it works lets you build your own splits to extract any part of a string or unpack a comma-delimited field into a full hierarchy.

Using a Superstore-style dataset, Tim first lets Tableau auto-split an order ID field, then reverse-engineers the generated calculation, then writes his own splits on a comma-delimited product name field to pull out individual hierarchy levels.

The Breakdown
  1. Let Tableau auto-split a field 0:20

    Right-click a field and choose Transform > Split and Tableau analyses the string, picks a likely delimiter, and creates as many new calculated fields as it thinks the string contains – a quick way to see splitting in action without writing anything yourself.

  2. Spot and read the generated calculation 1:37

    Fields Tableau creates this way show an equals sign, marking them as calculated fields sitting alongside your own; clicking into one reveals it's just an ordinary SPLIT() formula so you can see exactly how Tableau built it.

  3. Understand the SPLIT syntax 2:00

    SPLIT(field, delimiter, token number) takes the string, the character to split on, and which instance of that delimiter to extract up to – so the same function with a different number pulls out a different chunk of the same string.

  4. Write your own split on a delimiter 2:51

    Once you know the delimiter (a comma in this case), create a calculated field and write SPLIT() yourself, targeting the first instance to pull out everything before it – this is the manual version of what Tableau's auto-split does behind the scenes.

  5. Wrap numeric splits in INT() 4:00

    When a split result is numeric, wrapping it in INT() (as Tableau's own auto-generated calculations do) makes the field more efficient to use downstream, rather than leaving it as a string that behaves like a fractional measure.

  6. Know the no-match behaviour 4:42

    If the delimiter isn't found in a given value, split doesn't return null – it returns the whole original string. Worth checking for when auditing split results across a column.

  7. Extract later segments by duplicating and incrementing the token 5:00

    To get the second, third or later piece of a delimited string, duplicate the split calculation and increase the token number; each successive split peels off the next segment, letting you rebuild a full hierarchy out of one comma-delimited field.

  8. Use custom split for full hierarchies 7:05

    Custom Split lets you pick the delimiter and how many resulting columns to generate in one step, automatically naming and creating all the split fields together – the fastest route to turning one delimited text field into a multi-level hierarchy.

Worth Knowing
  • Split returns the entire original string when it can't find the delimiter, rather than nulling the value out, so watch for that when scanning results.
  • Auto-split only works well if the delimiter is consistent across the field – running it on a field without that delimiter (like order ID with a comma) does nothing useful.
  • Some database fields pack a hierarchy into one text field using commas or spacing to save columns; splitting is the trick to unpack that back into levels for analysis.
Use It When

Reach for this when a field arrives as one delimited string – like a comma-separated category path or an ID with embedded segments – and you need to pull out individual parts or rebuild it as a proper hierarchy for drilling down.

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.