Categorizing expenses means judgment. "SQ *BLUE BOTTLE" is coffee, so it is Food. A formula does not know that unless you tell it, keyword by keyword.
This is the kind of task Google Sheets was never built for. It is reading, not math.
The Old Way (Nested IF or a Lookup Table)
Most people build a keyword table and VLOOKUP against it, or stack IF statements until the formula scrolls off the screen.
=IFS(REGEXMATCH(A1,"UBER|LYFT"),"Travel", REGEXMATCH(A1,"BOTTLE|CAFE|RESTAURANT"),"Food", REGEXMATCH(A1,"AWS|ADOBE|SLACK"),"Software", TRUE,"Other")Every new merchant means a new keyword. Miss one and it falls into "Other." A restaurant you have never seen before gets no label. You end up maintaining a keyword dictionary forever, and it is always one transaction behind.
The FITS Way (Let the AI Read the Merchant)
With FITS, you name the categories once and let the AI decide where each row belongs.
=FITS("Categorize this expense as Food, Travel, Software, or Other: " & A1)It knows Blue Bottle is coffee and Adobe is software without a keyword list. New merchant next month? It still gets it right, no formula edit.
=FITS("Label this transaction with a spending category and a confidence note: " & A1)Because it reads the merchant the way you would, unfamiliar names still land in the right bucket.
When to Use Each
A lookup table is fine when you have a short, fixed list of known merchants. Reach for =FITS() when new vendors show up constantly or the descriptions are cryptic. The same "describe the categories" trick works to classify any text into categories or run sentiment analysis on feedback. Categorizing is one job on a long list. Read the full guide to automating Google Sheets tasks you used to need regex for.
Stop Maintaining Keyword Tables
FITS puts plain-English AI formulas inside Google Sheets. Name your categories. Let it sort the rows. Free tier included.