AdRoll Ads Targeting: Reaching the Audience That Converts

14 October 2026 6 min read Arda Acar

Written by:

Advertising

View profile

Ask AI about this page

10 views 6 min read

AdRoll ads targeting starts with the AdRoll pixel, which builds audience segments from site visitors, and extends to customer lists, lookalike audiences and contextual targeting across display, social and native placements. The audiences that convert best are usually narrow segments with clear intent, such as cart abandoners or pricing page visitors, combined with sensible frequency caps and exclusions of recent buyers.

How does AdRoll build its audiences?

Once the AdRoll pixel is on every page, it records visits and events. In the Audiences section, you create segments based on URL rules, events and time windows. Those segments can then be used across AdRoll’s channels, which include display across ad exchanges and social placements on networks like Facebook and Instagram where connected.

AdRoll Ads Targeting: Reaching the Audience That Converts — overview

Clean data is the foundation. If the pixel fires twice, misses pages or fails to record purchases, every segment built on it will be wrong. Test the pixel with the AdRoll pixel checker or your browser’s network tab before building audiences.

Which retargeting segments convert best?

Not every visitor is equally valuable. Building segments by intent lets you match message and budget to how close someone is to buying.

  • Cart and checkout abandoners: highest intent, short window such as 7 to 14 days.
  • Product page viewers: medium intent, show the viewed product or category.
  • Pricing or demo page visitors (B2B): high intent, focus on proof and next step.
  • Blog readers: low commercial intent, offer content or a newsletter rather than a hard sell.
  • Past customers: cross-sell, replenishment or loyalty offers, excluded from acquisition ads.

Choosing time windows

Duration should reflect your sales cycle. A fashion store might see most retargeting value within a week or two, while a B2B software company with a long evaluation period may keep visitors in a segment for 60 or 90 days. Start with the defaults, then compare conversion rates by recency once you have data.

AdRoll Ads Targeting: Reaching the Audience That Converts — in practice
How do lookalike and contextual targeting fit in?

Retargeting only reaches people who already know you. To find new buyers, AdRoll offers lookalike audiences built from your existing segments or customer lists, plus contextual targeting that places ads on pages related to chosen topics or keywords.

Lookalikes work best when the seed audience is your best customers, not all visitors. A seed of repeat buyers or high-value customers gives the model a clearer picture than a list of everyone who landed on the home page. Contextual targeting suits brands in privacy-sensitive markets or categories, since it relies on page content rather than personal data.

Customer lists and CRM segments

Uploading hashed customer emails lets you target or exclude people based on what they did offline or in your CRM, such as lapsed customers, trial users who did not upgrade or high-value accounts. Keep these lists fresh; an export from six months ago will include people who have since bought, churned or unsubscribed. Many teams automate the sync through AdRoll’s integrations with platforms like Shopify, HubSpot or Salesforce so lists update without manual uploads.

Exclusions are targeting too

The fastest way to waste budget is to keep showing ads to people who already converted or who will never buy. Set exclusions deliberately.

  • Exclude recent purchasers from acquisition and abandonment campaigns.
  • Exclude visitors to careers, support login or investor pages.
  • Exclude very short sessions, for example visitors who bounced within seconds, if your plan supports it.
  • Exclude existing customers from first-order discount offers.

For example, an online pet food store selling subscriptions should exclude active subscribers from all discount retargeting, and run a separate campaign for lapsed subscribers instead.

Check segment sizes before launch. A segment needs enough people in it for the campaign to deliver; AdRoll shows estimated audience size, and very small segments often fail to spend at all. Combine closely related segments when they are too small to run on their own.

What frequency cap should you use?

Seeing the same ad many times a day does not make people more likely to buy; it makes them annoyed. AdRoll lets you set frequency caps at the campaign level. Start conservatively, such as a few impressions per user per day, and watch performance. If conversion rate falls as frequency rises, lower the cap or refresh creatives.

Frequency also interacts with segment size. A small segment with a large budget will hit the same people repeatedly. If spend cannot be delivered without high frequency, the segment is too small for that budget.

Budget allocation across segments

Segments with the highest intent are usually small, so they cannot absorb large budgets without raising frequency. A sensible split puts a modest, steady budget behind abandoners and high-intent visitors, a larger share behind lookalike and contextual prospecting, and a small test budget behind experimental segments. Revisit the split monthly as segment sizes and costs change, rather than setting it once and forgetting it.

Creative and messaging by segment

Targeting and creative work together. A cart abandoner should see the product they left, perhaps with free shipping or a guarantee. A blog reader should see a helpful resource. Sending one generic brand banner to every segment wastes the effort spent building them.

Use dynamic product ads for e-commerce when your product feed is connected, and prepare static creatives in the common sizes AdRoll recommends for display. Test one change at a time: a new headline, a new offer, or a new visual, not all three at once.

How do you test and measure targeting?

Measure each segment separately. Look at view-through and click-through conversions as distinct numbers, because view-through attribution can make retargeting look more effective than it is. Where possible, run a holdout test: exclude a random share of a segment from ads and compare their conversion rate with the exposed group. The difference is the true incremental effect.

Track cost per acquisition and ROAS by segment, and compare them with the same metrics from your analytics platform. If AdRoll reports strong results but site revenue does not move, attribution may be claiming conversions that would have happened anyway.

Retargeting depends on cookies and identifiers, so consent rules apply. In regions covered by GDPR or similar laws, make sure your consent banner controls when the AdRoll pixel fires, and that visitors who decline are not added to segments. AdRoll provides consent settings and integrates with common consent management platforms. Segments will be smaller as a result, which is another reason to lean on contextual targeting and first-party customer lists for prospecting.

Where to start this week

Check that the AdRoll pixel records page views and purchases correctly. Then split your single all-visitors audience into three segments: cart abandoners, product viewers and past customers. Give each its own message and frequency cap, exclude recent buyers from the first two and review cost per acquisition by segment after two weeks.

Keep reading: Adroll Ads · Advertising

Share