The Scenario
You are a data analyst. The data warehouse join is scheduled for Friday. You have a Google Sheet with 600 company names entered by sales reps over the past six months — abbreviations, missing legal forms, misspellings, inconsistent use of LLC vs ООО vs ООО «».
The analytics team needs this sheet joined to the official registry dataset by canonical legal name and INN. The join is failing on 38% of rows because the sales-entered names do not match the registry strings.
The slow version of this week:
- Sort the list and manually identify obvious duplicates
- Look each ambiguous name up on the FTS registry site
- Copy the canonical name and INN
- Go back to the sheet, find the right row, paste
- Repeat for 600 entries over two days
- Deliver Friday afternoon, two hours after the data warehouse team needed the file.
The fast version is one prompt and the canonical names are in the sheet before noon.
The Easy Way: One Prompt in SheetXAI
SheetXAI is an AI agent inside your spreadsheet that reads the company name column, calls DaData's company suggestion endpoint for each row, and writes the canonical legal name and INN back.
Open the SheetXAI sidebar and type:
For each company name in column A of my sheet, use DaData to suggest the closest official Russian company match and write the canonical legal name into column B and the INN into column C. If DaData returns multiple matches, use the highest-confidence result. If no match is found, write "NOT FOUND" in column B.
SheetXAI reads column A, calls DaData's suggest-party endpoint for each row, and writes two standardized columns back. All 600 entries, resolved to official registry names and INNs, in one pass.
What You Get
Two new columns with authoritative registry data for every company in the sheet:
- Column B — canonical legal name from the FTS registry (ООО, АО, ПАО, ИП, etc. in full)
- Column C — INN (taxpayer identification number) for the matched entity
NOT FOUND rows are explicit. Entries that DaData cannot resolve with confidence are flagged rather than silently matched to the wrong company.
What If the Data Is Not Quite Ready
A sales-entered company list has patterns that complicate the lookup. SheetXAI handles them in the same prompt.
When the column mixes Russian company names and individual entrepreneur names
Some rows are legal entities (ООО, АО), others are individual entrepreneurs (ИП Surname). The lookup strategy differs.
For each entry in column A, use DaData to suggest the best match. If the result is an individual entrepreneur, write the full ИП name into column B and the ОГРНИП into column C. If it is a legal entity, write the legal name into column B and the INN into column C.
When confidence is low and you want a human review flag
Some entries are so ambiguous that even DaData's top result may be wrong. You want to flag low-confidence matches for manual review.
For each company name in column A, use DaData to find the closest registry match and write the canonical name into column B and the INN into column C. If DaData's confidence score for the match is below 0.7, write "LOW CONFIDENCE — REVIEW" in column D. Otherwise leave column D blank.
When the sheet has both company names and some INNs already
Some rows already have the INN in column C from a previous enrichment. You only want to resolve the rows where column C is blank.
Skip any row where column C already contains a value. For all other rows, use DaData to resolve the company name in column A to a canonical legal name in column B and INN in column C.
When you need the resolution plus a deduplication summary for the analytics lead
After canonicalization, some entries will resolve to the same company. The analytics lead wants to know how many unique companies are in the sheet before building the join.
For each company name in column A, use DaData to write the canonical legal name into column B and the INN into column C. Then write a summary below: total rows, number of unique INNs in column C, number of NOT FOUND results, and the top five companies by frequency of appearance in the original column A.
The pattern: the standardization and the deduplication analysis happen in the same prompt. The data warehouse join runs on Friday with a clean file.
Try It
Get the 7-day free trial of SheetXAI and open any sheet with messy company names, then ask it to resolve them to canonical registry names using DaData. The DaData integration is included in every SheetXAI plan. For related workflows, see how to enrich a supplier sheet with INN registry data or the DaData in Google Sheets overview.