What is an audience segment?
An audience segment is a defined group of people who share one or more traits, such as interests, demographics, location, past behavior or purchase intent. In programmatic buying, a segment becomes a targeting option inside a DSP: you include it to reach those people, or exclude it to avoid them. Segments let you buy audiences instead of individual websites.
Examples include recent site visitors, people in the market for a new car, parents of young children or existing customers you want to leave out of a prospecting campaign.
How audience segments work
A segment is built by collecting signals about people, linking those signals to an identifier and grouping identifiers that match a rule. Identifiers can be cookies, mobile advertising IDs, logged-in IDs or hashed emails. When an ad opportunity comes in, the DSP checks whether the user's identifier belongs to a segment on your line item and bids accordingly.
Types of audience segments
First-party segments: built from your own first-party data, such as site visitors, app users or CRM lists.
Third-party segments: built by data providers from third-party data and offered inside DSPs, usually at an added cost per thousand impressions.
Platform segments: audiences made by the platform itself, such as Google's affinity and in-market audiences or Amazon's shopping-based audiences.
Modeled segments: a lookalike audience that finds new people who resemble a seed group, such as your best customers.
Retargeting segments: people who already engaged with your brand, used for retargeting and exclusions.
Most segments also have a membership duration, often called a lookback window. A cart abandoner list might keep users for 14 days, while a past buyer list might keep them for a year.
Why audience segments matter
Relevance: the message fits the person, which usually helps response.
Less waste: you avoid paying for people who are unlikely to care, and you can exclude current customers from acquisition campaigns.
Sequencing: you can move people from awareness to consideration with different creatives for each stage.
Clearer reporting: results by segment show which groups actually drive outcomes.
The tradeoff is scale. The narrower the segment, the smaller the reach and the higher the price you may need to pay to win impressions.
Audience segment vs contextual targeting
Audience targeting is about the person: who they are or what they did before. Contextual targeting is about the content they are viewing right now. Audience segments depend on identifiers, which are harder to use in browsers that block third-party cookies and where consent is missing. Context needs no personal data. Strong plans often combine both, using segments where IDs exist and context where they do not.
Audience segments with AdGeeks
Each platform brings different audiences. DV360 offers Google audience segments, Amazon DSP offers audiences built on Amazon shopping signals, and Adform supports first-party and partner data. AdGeeks traders build and test segments across these seats, compare results and move budget to the groups that perform.
