Targeting Strategies
How to use Ours Privacy audiences well: exclusion lists to stop paying for people who already converted, retargeting windows, lookalike seeds, and suppression.
Use this page to decide what to build. The mechanics are in Building an audience; this page is about which audiences pay off. Retargeting and remarketing mean the same thing here.
Start with exclusion
The instinct is to build audiences of people to reach. The higher-return first audience is usually people to stop reaching.
A platform keeps showing an acquisition ad to someone who converted last week, because the conversion happened on your site and the platform never heard about it. An exclusion audience fixes that: build an audience of people who completed the action and attach it as an excluded audience on the campaign.
The pattern. One event condition: the conversion event, at least 1, over a window as long as somebody stays a customer. For a booked appointment that might be 90 days.
A daily sync matters most here. A CSV you uploaded last month excludes last month's customers and nobody since.
Exclude at every step
If a campaign asks people to book, exclude people who booked. If it asks them to complete an intake form, exclude people who completed it. An ad asking a current patient to become a patient is a bad experience, not just a wasted impression.
A useful audit: list your live campaigns and the action each one asks for. Every one of those is an exclusion audience you probably do not have yet.
Retargeting the right window
A retargeting audience is people who showed intent and did not finish. The thing to get right is the window. Too short and you lose people making a considered decision. Too long and you are reminding people of a choice they already made.
| Intent signal | Window |
|---|---|
| Started a form, did not submit | 7 to 14 days |
| Viewed a high-intent page | 30 days |
| Any site visit | 30 to 90 days |
| Repeat visits (three or more) | 60 days |
Always pair the intent condition with the exclusion: the intent event at least 1 and the conversion event exactly 0. Without the second condition you are retargeting your own customers.
Layer behavior with who someone is
Viewed the pricing page and is in a state you serve. Started an application and arrived from paid search. Two conditions is usually the sweet spot. Past three, audiences tend to fall below platform minimums.
Lookalike seeds
The quality of a lookalike is set by the seed.
- Seed on your best outcome, not your biggest list. Repeat customers model a better customer than everyone who ever bought.
- Respect the floor. Meta's guidance is at least 1,000 people; Google recommends at least 5,000.
- Exclude the seed from the lookalike campaign. Otherwise part of the budget re-reaches the customers you modeled on, and they convert well enough to flatter the results.
- Keep it synced. A seed frozen in a CSV models the customers you had when you uploaded it.
Suppression beyond advertising
People who asked not to be contacted, people in an active support escalation, people who already have an appointment this week. Each is an audience whose only purpose is to be excluded from something, and syncing it is cheaper than remembering.
Habits
- Name for the team, generate for the platform. A descriptive Export Name shows your strategy to everyone with access to the ad account.
- Preview before you sync. An audience of 40 people will not serve anywhere.
- Change one thing at a time, so a change in performance tells you what caused it.
- Measure. Audience Performance reports how audiences convert.
- Retire audiences. A sync runs until someone stops it.
Next Steps
- Recipes for these strategies written out as configurations.
- Event conditions for windows, counts, and the zero-count pattern.
- Audience Performance for measuring the result.
Need help? Contact support@oursprivacy.com.
How is this guide?