A winning product is one that people want, that is easy to show in an ad, and that leaves enough margin to pay for customer acquisition. The fastest way to find candidates is to watch what other stores are already advertising, and more importantly, what they keep advertising.
What makes a product a winner
- Solves a clear problem or creates a strong desire
- Easy to demonstrate in a few seconds of video
- Not easily found in local stores
- Sells for at least three times its landed cost
- Light enough to ship cheaply
- Appeals to a broad audience or a passionate niche
Signal 1: running time
Stores turn off unprofitable ads quickly. If several stores have advertised the same product for weeks, the economics probably work. Filter ad searches to ads running 30 days or more and look for products that appear again and again.
Signal 2: number of advertisers
One store running a product could be luck. Five stores running it is demand. Search the product name or its main benefit and count how many different advertisers are pushing it. Many advertisers with recent start dates suggests a product is trending, many with old start dates suggests a durable evergreen product.
| Pattern | What it suggests |
|---|---|
| Few advertisers, recent ads | Early trend or a test, risky but potentially big |
| Many advertisers, recent ads | Hot trend, act fast with a strong angle |
| Many advertisers, long-running ads | Proven evergreen, compete on brand and creative |
| Few advertisers, very old ads | Niche winner owned by one brand |
Signal 3: creative volume
Brands invest creative effort in products that sell. When one store runs ten or more different videos for one product, it is usually scaling. Open those ads and note which hooks appear most often, they show you what the market responds to.
Signal 4: store behaviour
Click through to the store. Look at whether the product is the hero of the homepage, whether it has bundles, and how many reviews it has. Store DNA shows the catalogue size, recent product launches and tech stack, which helps you understand whether the store is a one-product brand or a general store testing lots of items.
A weekly product research routine
- Pick two or three niches you understand.
- Browse new and long-running ads in each niche with video previews on.
- Shortlist products that appear across several advertisers.
- Check margins with a profit calculator using real supplier costs.
- Look for an angle competitors are not using.
- Test with three to five creatives before committing inventory.
Validating before you scale
Ad research reduces risk but does not remove it. Validate with a small test budget, watch link click-through rate and cost per add to cart in the first days, and only scale when the numbers clear your break-even ROAS. Remember that the stores you studied have their own brand trust, so your offer and creative need to stand on their own.
Mistakes to avoid
- Jumping on a product only because one viral ad exists
- Ignoring shipping weight and delivery times
- Competing only on price against established stores
- Reusing a competitor's exact creative instead of making your own
- Scaling before the unit economics are clear
Summary
Winning products leave a trail in the ad library: long running times, many advertisers and lots of creative variations. Combine those signals with solid margins and a fresh angle, then validate with a disciplined test before scaling.
Search millions of real Facebook and Instagram ads with video and image previews. Free to start, no card needed.
Frequently asked questions
What is a winning product?
A product with strong demand, easy demonstration in ads and enough margin to acquire customers profitably.
How long should an ad run before I treat it as a signal?
Thirty days is a useful threshold. Ads running for several months are even stronger signals.
Is it too late if many stores already sell a product?
Not necessarily. Evergreen products can support many stores. Win with a better angle, offer or brand.
Where can I see how many stores advertise a product?
Search the product in an ad library and group the results by advertiser. Media Ads Library shows advertiser profiles for this.
How much should I spend to test a product?
Enough to get meaningful data across a few creatives, often a few hundred dollars. Set a stop-loss based on your break-even cost per purchase.