How to Find Facebook Ads That Have Been Running Profitably for 90+ Days
Most spy tools point you to the same viral products that saturate in weeks. The inverse filter — 90+ running days, winning phase, low duplicate count — finds products with proven durability under real auction pressure.
How to Find Facebook Ads That Have Been Running Profitably for 90+ Days
Stop chasing viral products that die in 6 weeks. Find the ones that survive 90 days of real ad spend — and aren't flooded yet.
Why "Viral" Products Keep Killing Your Margins
Product lifecycles collapsed. Two years ago a winning product could run for 6 months before saturation hit. Now it's 6–8 weeks.
The pattern is always the same. A product goes viral on TikTok. Every spy tool surfaces it. 200 dropshippers launch the same product in the same week. CPMs spike. Margins compress.
By the time your test campaign has data, the window already closed.
Here's the math. A product with $15 COGS and a $59 selling price gives you $44 gross margin. At a $25 CPA, you clear $19 per unit. At a $42 CPA — which is what happens when 12 other stores bid on the same audience — your margin drops to $2.
The problem isn't the product. It's the signal you used to find it.
"Trending" and "viral" are popularity metrics. They tell you what everyone else is already looking at. That's not competitive intelligence — that's a crowd stampede.
You need the opposite signal. Not what's hot right now, but what's been quietly profitable for months without attracting a flood of copycats.
The 90-Day Durability Signal
An ad running for 90+ days is paying for itself. Nobody spends real money for three months on a losing campaign.
But duration alone isn't enough. A product can run for 90 days and still be saturated — if 50 other advertisers are running the same creative, the market is flooded.
The real signal combines three data points:
Running days: 90+. The ad survived three months of auction pressure. The unit economics work. The creative converts. The demand didn't evaporate after a trend cycle.
Phase: Winning. The ad isn't just alive — it's actively scaling. An ad running 90 days in "inactive" phase tells you nothing. Winning phase means the algorithm is still allocating budget to it.
Low duplicate count. This is the filter most people miss. Duplicate count tells you how many other advertisers are running the same creative. Low duplicates on a 90-day winning ad means the market validated it — but the mass market didn't flood it yet.
That's the durability sweet spot. Proven demand. Low competition.
Think about it from the advertiser's perspective. Running an ad for 90 days on Meta at $200/day costs $18,000. Nobody sustains that unless the ROAS is positive. The ad is market research you didn't pay for — and the low duplicate count means the insight isn't priced in yet.
How to Set Up the Durability Filter
Open Brandsearch Discovery and set the platform to Meta. Then apply three filters:
Running Days: 60–90+. Start at 60 if you want a wider net. For the strictest signal, set the minimum to 90. Everything below 60 days hasn't proven anything yet.
Phase: Winning. This removes dead ads, test campaigns, and scaling attempts that stalled. You're left with ads the algorithm is actively rewarding.
Sort by lowest duplicate count. Instead of sorting by reach or impressions — which surfaces the same ads everyone sees — you sort by fewest duplicates. Low duplicates plus winning phase plus 90 days running = validated product that hasn't gone mass-market.
I also add Video format. Video ads that survive 90+ days have proven both the product and the creative. Double validation.
The whole setup takes 30 seconds. Save it as a custom filter preset so you can rerun it weekly with one click.
The results look completely different from a standard search. The "trending" view shows you LED masks, portable blenders, posture correctors — the products everyone recognizes. The durability view shows products you've never heard of. Niche kitchen tools. Specialized pet accessories. B2B-adjacent consumer products with $80–$150 price points and repeat purchase potential.
These aren't sexy. They're profitable.
What to Check Before You Trust the Signal
The filter gives you a shortlist. Not every ad on it is worth pursuing. Here's what separates a real opportunity from noise.
Ad creative quality. A 90-day winner with rough production tells you the product carries itself. The demand is strong enough that even mediocre creative converts. You can outperform it with better production — real edge.
Price point. Products in the $40–$120 range have the best durability. Below $40, margins are too thin to sustain long campaigns. Above $120, the audience shrinks.
Ad copy patterns. Read the headlines. Are they pain-based ("Stop wasting money on X"), benefit-based ("Get Y in Z days"), or social-proof-based ("50,000 customers switched")? The copy angle that survives 90 days is a market-tested message. Extract the pattern for your own campaigns.
Niche specificity. The most durable products target specific audiences — not "everyone who wants to look good" but "women over 40 with joint pain who want low-impact exercise gear." The narrower the audience, the harder it is for mass-market copycats to compete.
Offer structure. Check the landing page URL on the ad card. Is it a product page, a bundle, or a dedicated landing page with a quiz? Brands running dedicated landing pages at 90+ days have dialed in their funnel. That's a signal the unit economics are tight.
Don't copy the ad. Extract the angle, the price point, and the funnel structure. Those transfer. The specific creative doesn't.
Stop reading about winners. Find them yourself.
Search 6.5M+ brands, their ads, revenue, and products — all in one place.
Try Brandsearch freeValidating the Brand Behind the Ad
Finding the ad is step one. Most people stop there. That's how you end up sourcing a product from a brand that's actually burning cash.
Click into the brand and check three things in Brandsearch Brand Analysis. Here's what it looks like on a brand like Gymshark that's been running at scale for months:
Traffic trend. Is monthly traffic growing or flat? A brand running 90-day ads with traffic up 20%+ month-over-month is genuinely scaling. A brand with the same ads and flat traffic is subsidizing losses.
Active ad count across platforms. If they're running the product on Meta and TikTok and Google simultaneously, the product works across channels. That's stronger than a single-platform winner.
Product count and revenue estimate. A monoproduct store doing $300K+/month with one product and 90-day ads? That product carries the entire business. A 500-product catalog store with the same ad? The product might be a rounding error in their revenue.
I also check if similar brands show the same pattern. Open Brandsearch Brand Library, filter by niche, sort by active ad count. If 3–4 brands are all running similar products with 60+ day ads, you're looking at a durable market — not a one-brand anomaly.
This validation takes 2–3 minutes per brand. It saves you weeks of wasted ad spend on products that looked good in a screenshot but had nothing behind them.
One more signal worth checking: the Scripts tab on tracked brands. If the brand has been running the same hooks for 60+ days, those hooks are battle-tested. Pull the top 3–5 opening lines and study the pattern. Pain point first? Curiosity gap? Outcome statement? The hook structure that survives 90 days is proven creative intelligence you can adapt for your own launch.
Why This Beats Every Other Research Method
The standard playbook — scroll TikTok, check AliExpress trending, browse Amazon bestsellers — gives you the same products everyone else sees. You're competing on speed, and speed is a losing game when 200 operators see the same trend on the same day.
The durability filter inverts the whole approach.
Instead of "what's trending right now?" you ask "what's been quietly profitable for 3 months without going viral?" Those products don't show up in TikTok compilations. They don't make AliExpress trending lists. They sit in ad libraries, running profitably, visible only to people who know how to filter for them.
Here's a real scenario. You find a posture corrector ad running for 94 days in winning phase. Duplicate count: 3. The brand behind it does $180K/month with 12 active Meta ads and traffic climbing 25% month-over-month.
That product has been validated by real auction pressure for three months — and only two other advertisers noticed.
Compare that to the viral kitchen gadget that hit TikTok last week with 2M views. By the time you source it, build a landing page, and launch your first test, 40 other stores already launched. Your CPMs are $15+ because everyone's bidding on the same audience. Your margin was dead before your first sale.
The Weekly Durability Scan (20 Minutes)
One-time searches give you one-time results. A system gives you a pipeline.
- Monday (10 min): Open Brandsearch Discovery. Apply your saved durability preset — 90+ days, winning phase, lowest duplicates. Scroll through results in your niche. Save anything with low duplicates and strong creative to a Brandsearch Swipe File folder called "Durability Picks."
- Thursday (10 min): Open your Swipe File. Click into each saved brand. Check traffic trend, revenue, and ad count in Brandsearch Brand Analysis. Kill anything with flat or declining traffic. Your shortlist should be 2–3 validated products.
20 minutes a week. After a month you have 8–12 validated products that survived real market conditions for 90+ days — each backed by traffic data, revenue estimates, and competitive context.
Compare that to scrolling "trending" feeds and hoping you're faster than 500 other operators looking at the same ads.
The scan also compounds. After 2–3 months you start seeing patterns across your saved products. Certain niches keep producing durable winners. Certain price points repeat. Certain funnel structures show up again and again. That pattern recognition tells you where to build your next brand, not just your next test campaign.
The Bottom Line
Every spy tool defaults to showing you what's hot right now. That's the wrong signal for building a durable business.
The products that make money long-term aren't the ones going viral this week. They're the ones that have been quietly profitable for three months with low competition — the ones nobody copies because nobody looks for them.
The method:
- Filter Brandsearch Discovery to 60–90+ running days, Winning phase, sort by lowest duplicates
- Validate the brand in Brandsearch Brand Analysis — traffic trend, ad count, product count
- Extract the angle from hooks and copy — don't copy the ad, extract the pattern
- Source and test one product per week from your validated shortlist
Durability is the only product research signal that doesn't expire the moment everyone else sees it.