EcommerceConversion

Ecommerce Recommendations

Suggest likely product categories, sizing journeys, gift guides, or onboarding paths.

Start with 500 free creditsDocumentation

Email gender · 1 credit per lookup · no monthly fee

Female · 94%
DressesSizingGift guide

The problem

A first-time visitor with no purchase history is a cold start: you have nothing to recommend from, so the homepage shows everyone the same generic grid and the shopper bounces before finding anything they wanted.

With Encrata

For a brand-new email with no other signal, an inferred gender gives the recommender a gentle first nudge toward likely-relevant categories, refined the moment real behaviour arrives. Keep it as a weak prior that browsing quickly overrides, so the guess helps the first minute without boxing anyone in afterward.

How it works

1

Send the address

Call the gender endpoint with just an email address; the name in the local part is enough to work from.

2

We infer the verdict

Encrata reads the name embedded in the address and returns a likely gender and country with a confidence score.

3

Use it with care

Segment, personalise or enrich on the verdict, always keeping neutral fallbacks and letting people override. One credit per lookup.

One call, structured answer

Ecommerce Recommendations runs on the email gender lookup at 1 credit per lookup. Same API key, same JSON shape as every other Encrata lookup.

  • A single male / female / unknown verdict with a 0-100 confidence score
  • Country signal included, so ambiguous names resolve by region
  • 1 credit per lookup, with free repeats for six months
  • unknown is returned honestly when a name is ambiguous, never a guess
Terminal
curl -X POST "https://app.encrata.com/api/email/gender" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"email": "jane.doe@example.com"}'
response.json
{
  "email": "jane.doe@example.com",
  "gender": "female",
  "country": "US",
  "probability": 98,
  "credits": 1
}

A weak prior for cold starts

Use the inferred gender only for the cold-start case, a brand-new email with no browsing or purchase history, as a weak prior that nudges the first set of recommendations toward likely-relevant categories. The moment real behaviour arrives, let it dominate: a couple of clicks should outweigh the inference entirely, so the guess shapes the first minute and nothing more.

A warmer first minute

This keeps the recommender useful before it has data without boxing anyone in afterward, which is the failure mode of leaning too hard on demographics. Watch the override rate, if shoppers routinely browse away from the initially-suggested categories, lower the prior's weight, the signal is there to warm up the first impression, not to define the customer.

More email gender use cases

View all
jane.doe@acme.com
+ Gender fieldFemale · 98%

Gender Enrichment

Add a probabilistic gender field to lead and customer profiles when users have not provided one

CRMEnrichment

Newsletter reach

58%
42%
Female 58%Male 42%

Audience Analytics

Understand campaign reach by inferred gender in aggregate across lists, subscribers, and event signups

MarketingAnalytics
Women's collection12,480
Men's collection9,020
2 segments built

Marketing Segmentation

Tailor content, product recommendations, or creative tests where gender relevance is legitimate

EcommerceMarketing
Start with 500 free credits
$curl https://encrata.com/api/agent/lookup -H "Authorization: Bearer YOUR_API_KEY" -d '{"e": "satya@microsoft.com"}'