The short answer
Broad targeting works when the ad itself filters who stops. Winners on our board do it mostly with the awareness level: problem-aware ads, which name the problem up front, have the longest median run (125 days) and the highest median estimated spend ($254K). In niches with a specific problem, such as joints or nerve care, about two thirds of winners open that way.
The short answer
Broad targeting works when the creative does the filtering. If the first seconds of your ad make it obvious who it is for, Meta's delivery system can find those people without interest targeting. If the ad could be for anyone, broad sends it to everyone and learns slowly.
This is less a strategy choice than it used to be. Meta's own help page on Advantage+ audience says the system takes your audience selections and can expand beyond them to find people likely to convert. The hard limits you can still set are things like minimum age, location, language and custom audience exclusions. Inside those limits, the ad picks the audience.
What we can see, and what we can't
The Meta Ad Library does not show an ad's targeting. So we can't tell you which of the 2,443 winners on our board run fully broad. What we can see is how they filter inside the creative, which is exactly the part that makes broad work.
The main filter is the awareness level the ad is written for: does it assume the viewer has never thought about the problem (unaware), knows the problem (problem-aware), knows the kinds of solutions (solution-aware), or knows the product?
| Awareness level of the ad | Ads | Median days live | Median estimated spend | Share reaching top tier |
|---|---|---|---|---|
| Unaware | 825 | 106 | $232K | 8.6% |
| Problem-aware | 491 | 125 | $254K | 10.2% |
| Solution-aware | 496 | 111 | $209K | 11.5% |
| Product-aware | 126 | 112 | $231K | 7.9% |
| Most aware | 205 | 117 | $230K | 7.8% |
Top tier is the top 10% of the board by Ad Radar's winner score (live 21+ days). Problem-aware ads have the longest median run and the highest median estimated spend of any level. They are the clearest filter: the first line names something only the right person has. A broad audience sees it, and only the people with that problem stop.
Unaware ads are the largest group, but they run shorter. They cast a wider net and depend more on the system to find buyers among the people who stop for a story.
How the filter changes by niche
The right filter depends on how specific the problem is. Here is the awareness mix of winners in a few niches (ads with an awareness tag):
| Niche | Ads | Unaware | Problem- or solution-aware |
|---|---|---|---|
| Nerve | 47 | 28% | 72% |
| Hormones | 70 | 27% | 66% |
| Joints | 189 | 31% | 65% |
| Foot | 102 | 36% | 61% |
| Pet | 248 | 33% | 44% |
| Food and nutrition | 147 | 40% | 38% |
| Cleaning | 147 | 51% | 31% |
| Digestion | 185 | 51% | 37% |
When the buyer has a specific, nameable problem, about two thirds of winners name it up front. That is broad targeting with a built-in filter. When the product is something nearly everyone uses (cleaning products, food), half of winners go unaware and rely on a story or a surprise to pull people in, because almost anyone is a possible buyer.
Lead generation is the harder case
Broad works best when the event you optimize for is a purchase, because only real buyers complete it. Lead campaigns optimize for a form fill, which is cheap and easy, so broad delivery will happily find people who fill forms without buying anything.
The board reflects how much harder that is. Only 121 of the 2,443 winners (5%) use a lead call to action, from 53 brands. Their median estimated spend is $168K, against $225K for ads asking for a purchase, and 8.3% reach the top tier against 9.7% for purchase ads.
If you run lead campaigns broad, the creative has to filter harder than for a product:
- Say who the offer is not for, or name the qualifying detail (location, budget, situation) in the first lines.
- Put the commitment in the ad: price ranges, time required, what happens after the form.
- Send quality signals back to Meta. Optimizing for a later event, such as a qualified lead or a booked call, teaches delivery what a good lead looks like.
Three ads built to run broad
I was cleaning pee every single day until I found what rescue shelters use to stop indoor peeing.
- Est. spend
- $2M
- Days live
- 482
- Format
- Video, 93s
Only dog owners with this exact problem keep watching. No interest targeting could find them as precisely as the first sentence does.
Cabinet door broken?
- Est. spend
- $1.6M
- Days live
- 341
- Format
- Video, 55s
- Variants
- 2
Three words that filter a mass audience down to people holding a broken hinge. There is no interest for that.
I don't like oatmeal, it's just, I can't with the texture.
- Est. spend
- $1.1M
- Days live
- 118
- Format
- Video, 47s
- Variants
- 2
Filters by objection instead of by problem. It finds the people who think the category isn't for them.
When broad works, and when it doesn't
| Situation | Broad? | Why |
|---|---|---|
| Steady purchase data, several distinct concepts live | Yes | The system has signal and options |
| Product with a nameable problem, ads name it | Yes | The creative filters |
| Mass-market product, unaware story ads | Usually | Almost anyone is a buyer |
| New pixel, few purchases a week | With help | Add audience suggestions or customer lists as a starting point |
| Very small B2B or professional audience | With help | The creative alone may not find them fast enough |
| Every ad is generic brand footage | No | Nothing in the ad tells delivery who it is for |
The last row is the common failure. Broad doesn't fail because the audience is broad. It fails because the ad is.
A checklist before you go broad
- Check the first line of every ad. Can you tell who it is for without seeing the product? If not, the ad has no filter.
- Match the filter to your niche. Specific problem: name it, problem-aware. Everyday product: a story or a cost comparison can work unaware. See problem-aware ads and unaware audience ads.
- Run more than one filter. One problem-aware concept and one unaware concept reach two groups inside the same broad audience. Our guide to creative as targeting shows how winners phrase the call-out.
- Keep the hard limits. Location, minimum age, language and exclusions of existing customers still matter. Broad means no interest guessing, not no rules.
- Fix your signal first. Broad learns from conversions. If events are missing or duplicated, see Conversions API and signal quality.
To see which filter winners in your niche use, filter Ad Radar's board by niche and awareness level. Each ad carries the awareness tag, so the mix in the table above can be pulled for any niche you sell in.
Figures marked as estimated spend come from Ad Radar's model of engagement on public Meta Ad Library ads. They are estimates, labeled as such, and are best used to rank ads against each other.