Broad Targeting on Facebook Ads: Budget Check and 14-Day Test

Broad targeting means you fill in only the audience controls Meta enforces (location, language, a minimum age and custom audience exclusions) and add no interest or behavior stack. Everything else you enter is a suggestion Meta can expand past.

With Advantage+ audience (Meta's automated audience setting), the audience controls are location, language, minimum age (18 to 25) and excluded custom audiences, meaning lists of people, such as past purchasers, that you keep out. Meta respects these. Age range beyond a minimum, gender, interests and lookalike audiences (groups Meta builds to resemble your customers) are suggestions. The minimum age is capped at 25; to set a higher minimum, you switch to original audiences (Meta's manual setup, where your age range applies as a limit). Jon Loomer's commentary describes anything entered in Audience Controls as a tight constraint (October 28, 2024).

Original audiences do not make interests binding either. On a conversions goal, lookalike audiences and detailed targeting expand beyond your selections anyway. An optimization event is the action you tell Meta to chase, such as a purchase.

As of January 15, 2026, Meta removed many detailed targeting interest categories, and ad sets still using them stopped delivering. Much interest stacking is therefore no longer possible, and an old interest ad set may have stopped delivering.

The practical consequence: building separate ad sets for separate interests only creates ad sets that can reach all of the same people. You pay for the split and get no separation.

If you want to keep existing customers out, put a purchasers list in exclusions. Exclusions are a control, so Meta respects them.

Your budget can feed one broad ad set if your daily spend is at least cost per purchase × 50 ÷ 7. This is our model, built on the learning phase: the period after a new ad set launches or is significantly edited, which ends at roughly 50 optimization events in 7 days. Our Meta ads learning phase guide shows the same formula.

The table applies it. Each figure is cost per purchase × 50 ÷ 7, rounded to the nearest dollar, and the four-ad-set column multiplies the unrounded figure by four.

Skip the arithmetic with our Meta ads budget and creative test calculator.

Run it backwards: at $100 a day and a $40 cost per purchase, one ad set produces 100 × 7 ÷ 40 = 17.5 purchases a week. Split four ways, each ad set gets about 4.4.

• Pass: your daily budget on the broad ad set is at or above the floor for your cost per purchase.
• Fail: it is below the floor. Do not split. Consolidate into one ad set, or optimize for a higher-funnel event such as add to cart so the ad set collects enough events to learn.

This is a planning floor, not a guarantee. It tells you whether a test can be read, not whether broad will win.

Broad targeting gives mixed results because the gain from switching is modest, and in our experience four other things often swamp it: the budget split, the optimization event, mid-test edits and undifferentiated ads.

The gain is modest. Meta reports that Advantage+ audience could get 7.2% lower cost per result on Sales and App promotion, 9.7% on Traffic, Engagement and Leads, and 14.8% on Awareness, from its analysis of A/B tests run January 2023 to August 2024. That is Meta's own analysis of its own product, not an independent study. Our inference: a single-digit gain on sales is easily buried by an underfunded or edited ad set.

The optimization event matters. Broad is worth trying when you optimize for a conversion, especially a purchase. On traffic or engagement goals, the system can find low-quality actions because it treats every click alike.

Edits restart learning. Creative changes, audience changes, optimization event changes, pausing and large budget or bid changes can restart the 7-day clock, as our learning phase guide lists.

The ads are the same for everyone. If every ad makes one general claim, Meta has nothing to sort on. The creative section below fixes this.

Keep original audiences when a rule forces a tighter audience or when your buyers are too few for a broad pool to contain. Outside those cases, restricting age and gender on purchase campaigns will likely drive up your costs.

• Minimum age above 25. Advantage+ audience caps the minimum at 25, so switch to original audiences if your buyers start at 26 or older. Legal minimums of 21 or lower work in either setup.
• Lead generation with a demographic problem. If the algorithm is filling forms from a group that never converts, an age restriction can be worth it, as that same guide notes.
• Special Ad Categories. Ads for housing, employment, credit and social issues follow separate Meta targeting rules. Read those first.
• A small buyer pool. Our recommendation: if your realistic buyers are a tiny share of the country, a broad ad set spends most of its impressions on people who will never buy, so keep a tight audience.
• A new account with no purchase history. Our recommendation: there is little to learn from, so start with a higher-funnel event and move to purchases once volume builds.

Run it as a two-arm test with identical budget, creative and optimization event, for 14 days, with no edits. This is our recommended method; our Meta Andromeda update guide also suggests a two-week split of your best audience against broad. Use our creative test capacity planner to size the test before you start.

If results fell suddenly rather than gradually before you started, the cause may sit outside targeting. The Andromeda guide covers how to check.

Write three to five ads that each address a different person, because with broad delivery the ad is what sorts viewers. That is our recommendation, based on the fact that audience inputs no longer bind delivery. Match each angle to how aware the buyer is, using customer awareness stages.

Fill each row from real customer language (see review mining for Meta ad angles), then turn each into a brief with our ad creative brief template, which is built around one test.

This is where we help. Our creative intelligence platform combines ad-performance data, competitor activity, customer signals and past ad learnings into a ranked weekly creative test plan and execution-ready creative briefs, so you know which angle to write next rather than guessing. We serve consumer brands, performance marketing teams, creative teams and agencies running Meta ads. See our pricing. To see it on your own account, book a personalized demo with our founding team.