How to Find Ad Angles by Studying What Competitors Stopped Running

The abandoned-cluster method for competitor ad creative analysis — find angles your competitors gave up on too early.

Most competitor ad creative analysis looks the same. Open an ad library. Filter to winning ads. Study what's scaling. Copy the pattern.

That's half the picture.

You're only looking at what survived. You never see what got killed. The killed ads tell you just as much about a market as the winners do.

Think about what you're actually doing when you filter to "winning" ads running 25+ days. You're seeing the 10-15% of creatives that passed the test. The other 85-90% — the formats that got pulled after 3 days, the angles that ran for 2 weeks then disappeared — you never see them.

A brand that ran 12 UGC-testimonial creatives for 6 weeks and then pulled all of them made a decision. That decision is data you can use.

A brand that tested a before-and-after format for 3 days across 5 products and killed every version? That's a different kind of data — faster, louder, clearer.

Both signals are invisible if you only study active ads. You see the survivors. You miss the graveyard. And the graveyard is where the real competitive gaps hide.

Most operators only study active ads. The killed ads reveal the real opportunities.

What Abandoned Ad Clusters Actually Tell You

Not every killed ad means a failed angle. People assume "stopped running" equals "didn't work." That's wrong about half the time.

There are two distinct signals in abandoned creative clusters.

Signal 1: Proven failure. A brand tests an angle across 3-5 creative variations. Every variation gets pulled within 3-7 days. Two other brands in the same niche tried the same format the same month and killed it just as fast.

That's a genuine failure signal. The angle doesn't convert in this niche right now. Skip it. You just saved $2,000-$5,000 in wasted ad spend learning a lesson someone else already paid for.

Signal 2: Underexecuted opportunity. A single brand ran a creative angle for 4-6 weeks, got moderate traction, then stopped. No other brand in the niche picked it up.

That's not failure. That's a brand that moved budget elsewhere — maybe they hit a cash ceiling, maybe their creative team pivoted to a new launch, maybe the execution was wrong but the angle was right.

You can test that angle with better execution and own it.

The difference between these two signals comes down to two numbers: lifespan and frequency across brands.

Here's the quick read:

Under 7 days + 2 or more brands abandoned the same format: proven bad angle. The niche rejected it. Skip.
7-14 days + single brand abandoned: inconclusive. Not enough data to act on. Note it and watch for more signals.
14-42 days + single brand abandoned: underexecuted opportunity. The economics were close to working. Study the creative, find the execution flaw, and test your own version.
42+ days then killed: this was working and stopped. Could be seasonal, could be creative fatigue, could be a strategy pivot. Check if they replaced it with something similar — if so, the angle is still alive.

Most people bucket every abandoned ad as "failed." The lifespan tells you exactly how close it was to working.

How to Read Abandonment Patterns (The Framework)

Here's the system I use to turn competitor ad creative analysis into a map of what angles are open. It takes about 20 minutes per brand.

Step 1: Pick 5-8 competitors in your niche. Not just the biggest. Include 2-3 mid-size brands doing $200K-$500K/month. They test more aggressively because they're still looking for their winning formula.

One-product stores with 5 ads don't give you enough signal. You need brands that launch 10-20 new creatives per month — that's where abandonment patterns become visible.

Start by searching your niche in Brandsearch Discovery. Filter to brands with 50+ active ads and 100K+ monthly traffic. Those are your research targets.

Step 2: Open their Creative Tests tab. In Brandsearch Brand Analysis, the Creative Tests tab groups every ad a brand has run into clusters — batches of similar creatives tested together. Each cluster shows how long it ran and whether it's still active.

You'll see two things side by side: the surviving clusters (ads still running or that ran 30+ days) and the abandoned clusters (ads that got pulled early). That side-by-side is where the intelligence lives.

Focus on clusters that went inactive in the last 60 days. Anything older is too stale to act on — the market may have shifted.

Gymshark Brand Analysis showing ad history — the starting point for mapping abandoned creative clusters

Step 3: Tag each dead cluster. I use a simple spreadsheet with four columns: brand, angle description, lifespan, volume. An angle that ran 3 days gets tagged "fast kill." An angle that ran 4+ weeks gets tagged "slow fade."

Volume matters too. One ad killed fast is a test that didn't pass the first gate. Eight ads killed after 12 days is a committed bet that failed.

Here's what a real log looks like after mapping one brand:

|-------|--------|----------|----------|-----|

Two fast kills, two slow fades. The slow fades go on your research queue.

Step 4: Cross-reference across brands. This is the step most people skip — and it's the one that separates real competitor ad creative analysis from browsing.

Pull up the same tab for your other 4-7 competitors. Look for the same angle. Did anyone else test it? How long did it survive?

If 3+ brands killed the same angle fast, mark it red. Skip it.

If only one brand tried it and killed it slowly, mark it green. That's your opportunity zone.

Stop reading about winners. Find them yourself.

Search 6.5M+ brands, their ads, revenue, and products — all in one place.

Try Brandsearch free

Turning Dead Angles Into Your Next Creative Test

You've got your green-tagged angles. Now you need to figure out why the original brand stopped — and what you'd do differently.

Go back to the dead cluster and study the actual creatives. Look at three things:

The hook. Did the ad open strong or bury the value prop? A good angle with a weak hook dies fast — not because the angle was wrong, but because nobody watched past 1.5 seconds.

The format. Was it UGC when it should have been a product demo? Was it a talking head when the angle needed visual proof? Format mismatch kills good angles constantly.

The offer. An ad can have the right hook, right format, and still fail because the landing page had no social proof, the price was too high, or the CTA was wrong. The angle is never just the ad — it's the combination of message, format, and funnel.

Here's a concrete example. Say you're in the supplement niche. A competitor tested a "morning routine" format — 15-second video of someone adding their product to coffee. Ran for 5 weeks, 6 ad variants, then killed.

You check the creatives. The hook was generic ("Start your day right"). The production looked like a stock video. The CTA was buried at the end.

The angle — morning routine integration — is proven to work in adjacent categories like skincare and fitness apparel. The execution was the problem. You reshoot the same concept with a real customer, a specific hook ("I replaced my $7 latte with this"), and a CTA in the first 3 seconds.

That's not copying a failed ad. That's fixing one.

Here's another scenario. A competitor tested a "what's in the box" unboxing format across 8 creatives over 5 weeks. They killed it. Every ad started with 3 seconds of talking to camera before the box appeared. The hook buried the interesting part.

I tested the same angle but opened with the box being ripped open. First frame: hands on product. No talking head intro. That version ran for 11 weeks.

The angle wasn't broken. The execution was.

I track competitors in Spectre to catch these patterns over time. When a tracked brand kills a batch, I see it within days. When they relaunch a new version of a killed angle 4 weeks later with different creative, that's a strong buy signal — it means they believed in the angle enough to retry.

Over 4-6 weeks of tracking, the abandonment patterns become obvious. You start seeing which formats get tested and killed repeatedly (pain-point openers, for example, cycle in and out of supplement niches every quarter) and which formats get tested once and never come back.

The Angles Nobody's Testing (And Why That Matters)

There's a third category beyond "proven failure" and "underexecuted opportunity."

It's the angle nobody in your niche has tried at all.

Cross-referencing abandoned clusters across 5-8 competitors gives you a map of what's been tested. Anything NOT on that map is untested territory.

Check adjacent niches. If the "myth-busting" hook format is crushing it in skincare but no supplement brand has tried it, that's an information gap. Most brands only study their direct competitors.

Open Brandsearch Discovery, filter to a related niche, set Phase to Winning, and look at the formats and angles scaling there. Then bring them back to your category.

Example: comparison-style ads ("Brand X vs us") are standard in SaaS and DTC supplements. But almost no fashion or home goods brand uses them. If you're selling candles or kitchen tools and you see comparison formats crushing it in the supplement space, that's an angle worth borrowing.

The absence of a test in your niche is also a signal. Nobody's validated it — but nobody's invalidated it either. That's where the biggest creative advantages live.

This is how you get ahead of every other brand running the same competitor ad creative analysis. They're all studying the same 5 competitors and copying the same winners. You're studying what those competitors abandoned, what they never tried, and what's working in the niche next door.

Discovery showing winning video ads filtered by format — the starting point for finding angles from adjacent niches

The Weekly Abandonment Audit (20 Minutes)

One-time research gets stale. Competitors launch and kill ads every week. A weekly cadence keeps your angle map current.

Tuesday (10 min): Open Brandsearch Brand Analysis for your top 5 competitors. Check the Creative Tests tab for each. Flag any new dead clusters from the past 7 days. Add them to your spreadsheet — brand, angle, lifespan.

Thursday (5 min): Cross-reference the week's dead clusters. Any angle killed by 2+ brands is a red flag. Any angle killed by one brand after 3+ weeks is a green flag.

Friday (5 min): Pick one green-flagged angle. Study the original creatives. Write a brief for your creative team: the angle, why the original failed, what to change.

After a month you'll have 4 tested angles — each one backed by competitive intelligence, not gut feeling. Two or three won't work. One will. That's how creative testing works when it's driven by data instead of brainstorming.

That's more competitive intelligence than most brands get from spending $50K/month on their own creative testing. You're reading your competitors' test results for free.

The Bottom Line

Every competitor ad creative analysis guide tells you to copy winners. That's table stakes. The operators who consistently find fresh angles are the ones studying what competitors stopped running.

The method:

Study abandoned clusters, not just surviving ads — Brandsearch Creative Tests tab
Tag each dead cluster: fast kill (proven failure) or slow fade (possible opportunity)
Cross-reference across brands — if 3+ killed the same angle, skip it
Study the execution of slow-fade angles — find the hook, format, or offer mistake
Check adjacent niches in Brandsearch Discovery for untested angles

Don't just follow winners. Mine the graveyard.