Predicted gender is a profile attribute that estimates whether a shopper is likely female or likely male, based on their first name. It is a prediction, not a fact a shopper has told you, so treat it as a helpful signal rather than a certainty.
How Predicted gender works
Instant looks at the first name on a shopper’s profile and compares it against global census data to estimate the most likely gender associated with that name. Each shopper is placed into one of three groups:
- Likely female
- Likely male
- Uncertain
A shopper is marked Uncertain when the first name is missing, not recognised, or ambiguous. Unisex names (for example Alex, Sam, or Jordan) and names that do not appear in the census data will fall into this group. Predicted gender is worked out automatically, so you do not need to set it up or collect anything from your shoppers.
To segment by Predicted gender:
- Go to Shoppers in your Instant AI dashboard
- Open the Advanced Filter
- Set the first dropdown to Profile Attributes
- Select Predicted gender in the conditions field
- Choose In to include a group, or Not In to exclude one
- Select one or more values: Likely female, Likely male, or Uncertain
- Click Apply
The shopper count updates so you can see how many shoppers match before you save the rule.
Ways to use it
Predicted gender is most useful when part of your range is aimed at a particular audience and you want the right shoppers to see the right products. Common uses include:
- Sending a gendered product range (for example menswear or womenswear) to the group most likely to be interested
- Tailoring the imagery, wording, or featured products in a campaign
- Combining it with other attributes, such as location or past purchases, to narrow a segment further
Tip: Because Predicted gender is an estimate, it works best for softening or steering content rather than hard targeting. Where you can, include products or messaging that work for everyone, so a shopper who is predicted incorrectly still sees something relevant.
Note: The Uncertain group can be a large share of your list, because it includes every shopper with a missing, unisex, or unrecognised first name. If you send a campaign only to Likely female or Likely male, you leave these shoppers out entirely. Consider including Uncertain in gender-specific sends, or sending them a neutral version.
FAQ
How accurate is Predicted gender? It is an approximation based on first name and census data, so it will not be right for every shopper. Names vary by region and culture, many names are unisex, and some shoppers use nicknames or initials. Use it as a directional signal, not a guarantee.
Does this use gender a shopper has given me? No. Predicted gender is estimated from the first name only. It does not read a gender field a shopper has filled in themselves. If you collect self-reported gender, that information is more reliable and you can segment on it separately.
Why are so many shoppers marked Uncertain? A shopper is Uncertain whenever the first name is missing, not recognised, or ambiguous. If a large part of your list has no first name captured, or has unisex names, you will see a bigger Uncertain group.
Does Predicted gender update over time? It reflects the first name currently on the profile. If a shopper’s first name is added or changed, the prediction is based on the latest name.
Can I use it to permanently label a shopper’s gender? No, and you should not treat it that way. It is a marketing estimate to help you tailor content, not a fixed attribute of the person. Keep this in mind for any sensitive or personal messaging.