Can predicted gender be used to match or filter members?

No. Matching, filtering and visibility must rely only on what members state about themselves. Predictions are statistical estimates that are wrong for some people and are limited to aggregate analytics and editable prefills.

Can the API verify that a profile is genuine?

No. A gender estimate from a name says nothing about whether a person is real or who they are. Never use it for verification, fraud checks or profile approval.

What happens with ambiguous names?

The API returns result_status unknown with a reason such as ambiguous. That is a normal, charged result, and in onboarding the right response is to leave the field blank for the member.

Gender prediction · Dating apps

Responsible gender estimates for dating app analytics

In a dating app, members must define their own gender and who they want to meet. The Finowl gender API therefore has a deliberately narrow role. Analytics teams can estimate the aggregate composition of new signups by region or campaign, and onboarding can optionally prefill the gender field with a suggestion that the member reviews, edits or clears before continuing.

A prediction reflects how a name is typically used, not how a person identifies, and it is wrong for some people. It must never be used for matching, verification, filtering, ranking or deciding who sees whom. Keep inclusive options in the field, make the prefill easy to change and never store an unconfirmed prediction as the member’s gender.

Try it in the sandbox About gender

USE CASES

Where it fits in dating apps

Aggregate signup analytics
Estimate the aggregate gender mix of new signups by region or acquisition campaign to understand marketing reach, with unknown results reported separately and no predictions attached to member profiles.
Editable onboarding prefill
Suggest a starting value in the gender field during onboarding only when confidence is high. The member must actively confirm or change it, with inclusive options and no prefill otherwise.
Campaign reach reviews
After a brand campaign, compare the estimated aggregate composition of new signups with previous periods to understand reach, informing creative and channel choices rather than any member experience.
Survey representativeness
Before publishing insights from a member survey, estimate in aggregate whether respondents look broadly similar to your member base, and caveat findings where the sample appears skewed.

HOW TO INTEGRATE

The workflow

1
For analytics, export first names and country codes into a research dataset separate from matching and safety systems.
2
Call POST /v1/gender/name with the country hint, or batch up to 50 names per request for aggregate reports.
3
For an onboarding prefill, use only identified results above a high confidence threshold and leave the field empty otherwise.
4
Store only the gender the member confirms in their profile, never the raw prediction.
5
Keep predictions out of matching, recommendations, verification and moderation entirely.

POST /v1/gender/name

curl https://api.finowl.co/v1/gender/name \
  -H "Authorization: Bearer $FINOWL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "name": "Alex Morgan",
  "country": "GB"
}'

Names like Alex may return result_status unknown with reason ambiguous. Then leave the field empty for the member to fill in. 1 credit per lookup.

GOOD PRACTICE

Before you ship

Never use predicted gender for matching, verification, filtering, ranking or safety decisions. Only the member’s own confirmed answer should shape their dating experience.
Dating data can reveal intimate details. Mention gender inference in your privacy notice, keep analytics aggregate and do not retain per-member predictions longer than needed.
A wrong prefill can feel hurtful. Use a high confidence threshold, show inclusive options and make changing or clearing the suggestion a single, obvious tap.

Read the API reference for every field and error code. Each lookup costs 1 credit (3 when it requests AI); see pricing.

A FEW GOOD QUESTIONS

Common questions