Marketing & GrowthFree
Competitor Ad Analysis Ranked by Longevity
Competitor ads sorted by how long each has run, then followed to the landing page behind it, because the survivors are the tested ones.
Counting a competitor's live ads tells you what they are trying. It does not tell you what worked. Across four hearing-aid competitors, 312 ads were live on one day and 18 of them had been running past 180 days. Those 18 are the only rows anybody paid to keep. The price angle was 30.8% of everything live and 1.0% of what lasted. Bluetooth and features was the second-most-run angle in the set, with zero ads past 180 days. Counting the set and ranking it by longevity produce opposite briefs.
Then every survivor gets followed to the page it lands on, because two ads carrying the same visible angle routinely sit on top of different businesses. One competitor ran a hearing-test ad for 284 days onto a six-question screener with no price anywhere. Another ran the same angle for 47 days onto a product grid with prices from $1,299 above the fold. A creative-only teardown files both under the same heading.
Built for performance marketers writing next quarter's concepts, product marketers checking whether their positioning is contested, and agencies pitching a new account cold. Pair it with the creative read by attribute once your own ads are running, and the competitor teardown for the product and pricing side. When the page you land traffic on is the constraint, that is a funnel and landing-page review. An open angle becomes real copy in the page copy review.
Why the ads they stopped are the harder half
Meta's own documentation is the reason this works. For an ordinary commercial ad the Ad Library shows only the date on which the ad started to run, which is all longevity needs. The same table says those ads get no archival at all, so a competitor's ad vanishes the day they switch it off. Everything you can see is by definition still working, and everything they tested and rejected is gone.
There is one way back in. Ads delivering to the EU or the UK are archived for a year after their last impression, so switching the country selector recovers what stopped. On this set that turned 166 live ads into 1,247, of which 1,081 had been switched off. Ranking angles by median days before they were killed put them in exactly the order the live survival rate did. Two separate observations, same answer, which is more confidence than a creative teardown usually earns.
Google is the harder library. Its launch post lists what the Ads Transparency Center gives you: the last date an ad ran, and the format. No start date, so longevity there needs either two passes weeks apart or an EEA region, where disclosure rules add one. Then follow the survivors to their pages. Fifteen of the eighteen land somewhere that shows no price before it asks for a contact detail, while 91 of the 96 live price-angle ads open with one.
How it works
Name the set
The competitors worth watching, and the markets you care about them running in.
Pull and date
Every live ad captured with its start date, so days running is a real column.
Recover the dead
The region switch that surfaces stopped ads, and the median run length per angle.
Read the pages
Each survivor's landing page opened, so the offer behind the angle gets written down.
What you get
- Every competitor ad ranked by days continuously running, which is the only free performance signal.
- Survival rate per angle, so the mix that lasts separates from the mix that ships.
- The kill list recovered from the archive that keeps stopped ads, not just live ones.
- Each surviving ad followed to its landing page, with the offer and price placement recorded.
- An angle-by-speaker grid showing which cells nobody in the set occupies, and why not.
- A Sheet of their live ads by angle and longevity, plus a Doc of three concepts.
Common questions
Why is how long an ad runs a better signal than how it looks?
Because somebody kept paying for it. Nobody funds a losing ad for six months, so a 284-day run is a competitor telling you their result without meaning to. On this set 18 of 312 live ads had run past 180 days. The other 294 are still a question mark, including all 74 feature-led ones.
Can't a long-running ad just be one nobody bothered to turn off?
Sometimes, which is why the kill list is checked against it. Ranking angles by median days before they were stopped produced the same order as the live survival rate, on two datasets that do not share records. Neglect would not agree with itself twice. A single long run can be an accident; a pattern across 18 is not.
How do you see ads a competitor already stopped?
By changing the country. Commercial ads are not archived when they stop, so the default view is survivors only. Ads that delivered to the EU or the UK stay for a year after their last impression, and on this set that recovered 1,081 stopped ads against 166 live ones. The archive is European, so treat it as a European kill list.
Does this work for Google search ads?
Partly, and the difference matters. Google's transparency tool publishes a last-shown date and a format, not a start date, so days running is not sitting there to be read. Two passes a month apart gives you a floor on it. Selecting a European region gives you a first-shown date, because disclosure rules there require more.
Why read the landing pages if you already have the creative?
Because the same angle sits on top of different businesses. Two competitors ran a hearing-test ad. One landed on a screener with no price and ran 284 days; the other landed on a product grid from $1,299 and ran 47. Filed by creative they are one row. Filed by what happens next they are opposite strategies.
How do you find an angle nobody is running?
By building a grid rather than a list. Seven angles against three speaker positions gives 21 cells; thirteen were occupied and three are empty because the claim would be regulated. That leaves five open cells, 23.8% of the message space, with five advertisers in the market. A gap you can name beats a gap you sensed.
What does this say about my own ads?
It gets held against them. Ferndale's 22 live ads sat in three angles whose survival across the competitive set was 1.0%, 0.0% and 5.9%, and none in the two highest at 18.4% and 29.4%. Its longest continuous run was 34 days, which is short of learning anything. Both facts came out of the same pass.
Competitor Ad Analysis Ranked by Longevity
Fill in the form and your workspace opens with the work already underway.