Bulk gender detection from CSV and Excel
Drop in a file with a name column and get it back with gender, probability and the evidence behind each answer. No code, no API key, no waiting on an engineer.
100 free credits every day. No card.
The limits are real numbers, not marketing ceilings
One million rows and 100 MB per file, CSV or XLSX. A genuine one-million-row CSV processes in about 8.6 seconds of compute; the XLSX equivalent in about 34, because the format costs more to parse.
Duplicates are collapsed before lookup rather than after billing. Real customer lists repeat first names heavily — a 200,000-row file typically holds around 16,000 unique ones — so this is usually an order-of-magnitude difference on the invoice rather than a rounding one.
What comes back
Your original columns, untouched, plus six appended:
| Column | What it holds |
|---|---|
| gender | male, female, or empty when there is no evidence |
| probability | 0–100. How dominant that gender is in the observed data |
| total_names | How many observations the answer rests on |
| country | The country the answer was resolved against |
| source | Where the answer came from |
| matched_as | The entry actually matched, when it differs from your input |
Add a country column before you upload
It is the highest-leverage thing you can do to a name file, and it costs nothing. ISO codes (TR, DE, US) or plain country names both work. Without one, a country-blind lookup returns the globally dominant reading, which is wrong in a predictable, clustered way rather than a random one.
The second thing worth checking is that each cell holds a name rather than a record. Ayşe Yılmaz is fine — the first name is extracted. Yılmaz, Ayşe or Mrs A. Yılmaz will lower your match rate for no reason.
Cost is shown before it is charged
Columns are detected automatically and shown for confirmation, along with the row count and the exact credit cost. Nothing is charged until you start the job, and large files keep processing after you close the tab.
Spreadsheet inputs are treated as hostile by default: a cell beginning with = or @ is a formula, not a name, and is rejected rather than normalised into one. That check exists because a formula string once resolved to a real-looking answer with 95% confidence.
Questions
Which file formats work?
CSV and XLSX, up to 1,000,000 rows and 100 MB.
Are failed lookups billed?
A lookup consumes a credit per resolved name, and duplicates within a file are resolved once. Purchased credits never expire.
Can I do this over the API instead?
Yes — /api/bulk accepts a JSON array of names in a single request, and the credit is charged per resolved name rather than per request.
What happens to my uploaded file?
You can purge the stored input and output from the job page at any time; retention terms are in the data processing agreement.
Related pages
Append a gender column to a customer list, CRM export or lead database, with a confidence field you can filter on and no monthly credit expiry.
Predict gender from a full name. How the given name is extracted from titles, initials, particles and surname-first orders, and how to audit that it worked.
What is actually in the name-to-gender dataset: seven counted national registries, WGND 2.0 for the rest, versioned snapshots and a per-answer source field.
A walkthrough for analysts: upload a name list, pick the right columns, read the confidence fields, and decide what to do with the rows that come back empty.
A reproducible benchmark across 4,610 names, 25 countries and nine writing systems. Two scripts return almost nothing, and we publish those numbers too.
Check it against your own list
Every number on this page is reproducible with a free key. If your data breaks it, that is the more interesting result.