Can we show different discounts based on predicted gender?

No. Predictions must not drive pricing, offers or access. Use them for aggregate analytics and light, reversible defaults such as department ordering that the shopper can change.

Why pass a country with the name?

Some names lean differently by country, such as Andrea, Jean or Kim. The optional ISO 3166-1 alpha-2 country hint lets the estimate reflect local usage.

How much does it cost to enrich past orders?

Each name lookup or batch item costs 1 credit in dataset-only mode and 3 credits when AI is requested. Batches of up to 50 items count as one request.

Gender prediction · E-commerce

Gender insights from names for e-commerce analytics

E-commerce analysts often have a name and an order history but little else. The Finowl gender API estimates how a first name is typically used, returning a confidence score and sample count, so teams can describe the audience behind a category or campaign in aggregate and plan catalogs with fewer blind spots.

Personalization should stay light and reversible. A prediction might choose which department appears first on a homepage for a new visitor who has not set preferences, but shoppers must be able to change it, and it must never affect prices, discounts, shipping options or which products someone is allowed to buy.

Try it in the sandbox About gender

USE CASES

Where it fits in e-commerce

Category audience reporting
Estimate the aggregate gender mix of buyers per category each quarter to support assortment planning, reporting unknown results separately so dashboards do not overstate certainty.
Default department ordering
For first-time visitors who have not set preferences, order department tiles using a high-confidence prediction, and remember any choice the shopper makes so their own preference always wins.
Campaign reach checks
Compare the estimated audience of an email campaign with the audience of the whole list to see whether subject lines or creative are reaching a narrower group than intended.
Size guide and content planning
Use aggregate estimates to decide which size guides, fit content or editorial stories to produce first, without tagging individual shoppers or changing what they are shown at checkout.

HOW TO INTEGRATE

The workflow

1
Collect the first name and the shipping country code your store already holds for each order.
2
Call POST /v1/gender/name with the country hint for each new customer, or batch historical records in groups of up to 50.
3
Keep results with result_status identified and a confidence threshold you choose, and bucket the rest as unknown.
4
Write the fields to your analytics warehouse rather than the customer profile used by pricing or promotions.
5
Let shoppers set their own department or style preferences, and always prefer those over the prediction.

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": "Kim Larsen",
  "country": "DK",
  "options": {
    "ai_mode": "off"
  }
}'

Dataset-only mode costs 1 credit. Kim leans differently by country, so passing the shipping country improves the estimate.

GOOD PRACTICE

Before you ship

Never vary prices, discounts, delivery options or product availability by predicted gender. Keep predictions in analytics and in easily overridden, low-stakes defaults.
Disclose gender inference in your privacy notice and keep the derived field in analytics storage with limited access, since it is personal data about each shopper.
A prediction is wrong for some people. Gendered defaults should be easy to switch, and gender-neutral layouts are often the better choice for new visitors.

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