Organize information

Standardize category labels in Excel

Map known label variants to an approved vocabulary while exposing unknown values

6 min setupTested with:ChatGPTReviewed: 2026-08-28

Add your context

Your text stays in this browser. AILesson Prompts does not send it to a model or server.

Approved categoriesObserved variantsMapping policy

Your prompt

Unfilled fields remain visible as placeholders, so you can still copy and edit the prompt

Design an auditable Excel mapping for category labels.

Approved categories and definitions:
[vocabulary]

Observed raw values and counts:
[observed]

Mapping policy:
[policy]

Do not merge labels only because they look similar. Normalize a comparison key separately from the approved output. Return a mapping table with RawVariant, NormalizedKey, CanonicalLabel, Confidence, Reason, and ReviewerNeeded; formulas to apply it; and a report of unmapped or ambiguous values. Preserve raw values and counts so the effect of every mapping can be reconciled.

Private by defaultPrompt assembly happens locally in your browser. Avoid placing confidential information into any AI service unless your organization allows it.

From input to outcome

A worked example

See how concrete context turns this recipe into a usable result

Actual input

Approved categories: New = not yet contacted; In progress = contact or work underway; Won = completed successfully; Lost = closed unsuccessfully.
Observed variants: new 42; New 18; NEW-LEAD 6; contacted 15; in progress 31; working 4; closed 12; won 20; lost 9; blank 3.
Mapping policy: Case, outer spaces, and hyphen versus space may be normalized. 'Contacted' and 'working' can map to In progress after review. 'Closed' is ambiguous between Won and Lost. Blank must remain Missing, not New.

Example output

Build a Mapping table rather than using nested SUBSTITUTE. A comparison key can be =LOWER(TRIM(SUBSTITUTE(A2,"-"," "))), but the mapping table remains authoritative.

RawVariant	NormalizedKey	CanonicalLabel	Confidence	Reason	ReviewerNeeded
new / New	new	New	High	case-only variation	No
NEW-LEAD	new lead	New	Medium	approved synonym must be confirmed	Yes
contacted / working	contacted / working	In progress	Medium	policy-approved workflow synonyms	Yes
closed	closed	—	None	outcome is unknown	Yes
blank	—	Missing	High	absence is not New	No

Apply with =IF(A2="","Missing",XLOOKUP(LOWER(TRIM(SUBSTITUTE(A2,"-"," "))),Mapping[NormalizedKey],Mapping[CanonicalLabel],"Review")). Do not add closed until its 12 rows are resolved from other evidence. Summarize counts by raw and canonical value; the mapped total plus Review must equal 160. Store mapping version and approval date so future imports can be reproduced.

Why this works

1
A mapping table turns silent replacements into explicit, reviewable decisions.
2
Unmapped-value reporting prevents new labels from disappearing into a fallback category.