Finowl
Name to gender
Analyze a first or full name and receive an estimate with confidence, or an explicit unknown result.
- First and full-name queries
- Optional country context
- Latin and non-Latin names
POST/v1/gender/name
FINOWL / GENDER
Gender predictions
A little more context. Room for nuance.
Start with a first or full name, email, or username. Get an estimate with confidence, one input at a time or in batches of up to 50. Keep unknown results visible when no prediction is available.
THREE STARTING POINTS
Meet your data
where it is.
Choose the input you already have. Interpret the prediction alongside its confidence and preserve unknown results.
Email to gender
Extract name signals from common email formats and enrich customer, lead, and analytics records.
- Personal and business formats
- Useful for CRM and lead enrichment
- Confidence in every response
POST/v1/gender/email
Username to gender
Find names, initials, and separators inside usernames and return a prediction when the signal allows.
- Social and community datasets
- Multiple username patterns
- Clear confidence output
POST/v1/gender/username
INPUT. PREDICTION. CONTEXT.
Different inputs. Familiar JSON.
Explore individual and batch requests, including what an unknown result looks like.
curl https://api.finowl.co/v1/gender/name \ -H "Authorization: Bearer $FINOWL_API_KEY" \ -H "Content-Type: application/json" \ --data '{ "name": "Alice" }'{ "data": { "gender": "female", "result_status": "identified" }}BUILT FOR THOUGHTFUL USE
People are more
than a prediction.
Keep unknowns visible.
An unknown result is useful information. Do not turn it into a guess or confuse it with a request error.
Let people self-identify.
A prediction is not verified identity. Information people choose to share should take precedence.
Use context, not certainty.
Treat the output as a limited signal, not a basis for consequential decisions about a person.
BATCH REQUESTS
Up to 50 inputs in one request.
Send names, emails, and usernames together. Each item gets its own result, so one invalid input does not hide the predictions for the rest.
- 1 to 50 items per request
- Mix names, emails, and usernames
- Optional IDs to match results to your records
POST/v1/gender/batch
- index
- The position of the item in your request.
- id
- Your own identifier, when you provide one.
- data
- The prediction, with unknown results preserved.
- error
- A per-item error code when that input could not be processed.
A FEW GOOD QUESTIONS
A signal, with context.
Add context. Keep the nuance.
Try the Gender API with the input you have, and build around the result your application receives.