Lookalike Audiences vs Saved Audiences: Modeled Population or Manual Definition
In short: Both end up attached to an ad set, but one is an opaque model and the other is a transparent recipe. A Lookalike Audience is a single modeled object - you pick a seed and a percentage, and Meta decides who is inside with no visibility into the individuals. A Saved Audience is a stored bundle of locations, ages, detailed targeting, and included or excluded audiences - fully editable, and it can wrap a lookalike inside it alongside other criteria. Saved audience criteria recalculate live every time it is used; a lookalike is frozen until you deliberately rebuild it. Use a lookalike when you want Meta's model to find resemblance for you, and a saved audience when you need full control over a reusable set of criteria - often both together, with the lookalike as one ingredient.
By the AdFlint research team · Fact-checked against current Google and Meta platform behavior · Last reviewed July 2026
Lookalike Audiences
Meta audiences modelled on a seed list, matching users who resemble your source on the signals Meta holds, sized as a percentage of a chosen country.
You pick a source - purchasers, high-value customers, engaged visitors - and a percentage; smaller percentages mean closer resemblance and less reach, larger ones the reverse. A value-based source lets Meta weight by revenue. Their edge has narrowed as broad targeting and Advantage+ audience improved. The persistent error is seeding from a large low-quality list, which models the wrong person very precisely.
Full definitionSaved Audiences
A reusable Meta targeting definition - locations, ages, detailed targeting, and included or excluded custom audiences - stored so ad sets can apply it without rebuilding it.
It is a saved configuration, not a fixed list of people; membership is recalculated at delivery every time it is used. Teams create them for consistency across many ad sets and for exclusion sets applied everywhere. The recurring problem is drift: a saved audience edited months ago silently propagates into new campaigns, and under Advantage+ audience its interest criteria act only as suggestions.
Full definitionSide by side.
The differences that actually change what happens in your account.
| Lookalike Audiences | Saved Audiences | |
|---|---|---|
| Foundation | A seed audience plus a percentage, modeled by Meta. | Manually assembled criteria - locations, ages, detailed targeting, include or exclude audiences. |
| Who defines the population | Meta's model, based on resemblance to the seed. | You, by picking demographics and interests - though under Advantage+ audience these become suggestions. |
| Needs a Custom Audience to exist | Yes, always - a lookalike cannot exist without a seed. | No - can be built from pure demographics and interests with zero data sources. |
| Recalculation | Fixed at creation until you deliberately rebuild it against updated seed data. | Recalculated live every time the ad set using it delivers. |
| Can it include or exclude other audiences | No - a lookalike is itself a single finished audience object. | Yes - it is a container that can include or exclude a lookalike or any Custom Audience. |
| Debuggability | Opaque - you cannot see who is inside, only the percentage and seed you chose. | Transparent - every criterion is visible and editable at any time. |
| Behavior under Advantage+ audience | Becomes an optional suggestion signal Meta can use or move beyond. | Its interest criteria also become suggestions, while any included Custom Audience or lookalike stays a firmer rule. |
What actually separates them.
A lookalike is itself a finished audience object; a saved audience is a container that can wrap a lookalike alongside demographics, locations, and other Custom Audiences.
Lookalike membership is opaque and modeled - you cannot inspect who is inside beyond percentage and seed - while saved audience criteria are fully visible and editable at any time.
A saved audience recalculates live at every delivery from its stored criteria, while a lookalike is a static model that only changes when you deliberately rebuild it.
A saved audience can exist with zero Custom Audiences, built purely from demographics and detailed targeting, while a lookalike cannot exist without a seed.
Editing a saved audience changes every ad set referencing it immediately, while updating a lookalike - regenerating it against fresh seed data - requires manually swapping it into each ad set that used the old version.
Which one should you use?
Use Lookalike Audiences when
- You want Meta to find people who resemble your best customers, not people who merely fit demographics you guessed at.
- You have a strong, value-weighted seed and want algorithmic expansion rather than manual criteria.
- You are prospecting cold and want a single audience object you can drop straight into an ad set.
- You want to compare tight versus loose resemblance percentages as a lever, rather than tightening interests.
Use Saved Audiences when
- You want a reusable bundle of locations, ages, and detailed targeting applied consistently across many ad sets.
- You need to layer an exclusion set - customers, recent converters - on top of demographic or interest targeting.
- You want full visibility into and control over every criterion, not an opaque modeled population.
- You are standardizing targeting logic across a team so one edit updates it everywhere it is used.
Common questions.
Can I put a lookalike inside a saved audience?
Yes, this is a common pattern - the saved audience uses the lookalike as its base included audience and then layers locations, ages, or detailed targeting on top of it.
Why can't I see who is in my lookalike audience?
Meta does not expose individual membership of a lookalike for privacy and modeling reasons. You can see the seed, the percentage, and the resulting reach estimate, but never a list of the actual people inside it.
Do I need a saved audience if I'm just using a lookalike?
Not strictly - a lookalike can be targeted directly in an ad set with no saved audience involved. Wrapping it in a saved audience only becomes useful when you want to add reusable demographic criteria or an exclusion audience alongside it.
Does editing my saved audience affect my lookalike?
No. They are separate objects - a saved audience simply references the lookalike as one of its included audiences, so editing the saved audience's other criteria does not alter the lookalike itself.
Or stop choosing between them.
AdFlint picks the setting, writes the ads, and keeps optimizing inside the Google and Meta accounts you already own.
Related comparisons
- Advantage+ Audience vs Broad Targeting
- Broad Targeting vs Detailed Targeting
- Broad Targeting vs Interest Targeting
- Broad Targeting vs Custom Audiences
- Broad Targeting vs Lookalike Audiences
- Broad Targeting vs Website Custom Audiences
- Broad Targeting vs Customer List Custom Audiences
- Broad Targeting vs Engagement Custom Audiences