Short answer: The Graveyard Report is a regular review of the ads in your account that failed, faded or were turned off. It finds ads that were cut by mistake and can be revived, spots patterns in how ads die, and turns those patterns into rules for the next creative brief. Your losing ads are the largest set of tests you own, and most accounts never read them.
Winners get all the attention. They get studied, copied and talked about in every meeting. Losers get paused and forgotten. That is a waste, because in most accounts the losers outnumber the winners many times over. Each one cost money, and each one answered a question. The Graveyard Report is where we go back and read the answers.
It also checks something more practical: whether ads that should have been saved actually were.
What is the Graveyard Report?
The Graveyard Report is a scheduled review of dead and dying ads that looks for three things: ads cut by mistake, ads worth reviving, and patterns that should change how you brief new creative.
It takes its name from the Graveyard campaign, the capped campaign we use to keep ads that sell but not at a profitable price. If you have not read it yet, start with the Graveyard campaign, which explains the cost cap, the rules for moving ads in, and the basic audit for missed moves.
This article goes one step further. The Graveyard campaign rescues ads with a pulse. The Graveyard Report studies the whole cemetery, including ads that never sold, to learn why they died.
Why review ads that failed?
Because a dead ad is a finished experiment. Reading it costs nothing more, and it can stop you from paying for the same lesson twice.
There are three payoffs:
- Money left on the table. Some ads get paused while they are still selling. Some die in a test that was never fair. These can often be brought back.
- Repeat mistakes. If the same kind of ad keeps failing in the same way, you are paying to relearn something you already know.
- Better briefs. Losers show the edges of what works. Knowing where an idea breaks is as useful as knowing where it works.
Daily optimization is about the next decision. It moves fast and does not look back. The Graveyard Report is the step that does look back, on a slower clock, with a wider view.
How did each ad die?
Sort every dead ad by how it died, not just that it died. The cause of death tells you whether the fault was the hook, the body, the offer, the setup or the luck of the auction.
- Never stopped anyone. What it looked like: Expensive outbound clicks, weak outbound CTR. What it usually means: The hook or first seconds failed
- Clicks but no intent. What it looked like: Decent outbound CTR, few add to carts. What it usually means: The ad attracted the wrong person, or promised something the page did not deliver
- Carts but no sales. What it looked like: Healthy add to carts, few purchases. What it usually means: Price, offer or website friction, not the creative
- Sold, but too expensive. What it looked like: Real sales above your cost goal. What it usually means: A Graveyard candidate, not a failure
- Strong start, then faded. What it looked like: A good first stretch, then a slide. What it usually means: Fatigue or a narrow audience, not a bad idea
- Never really tested. What it looked like: Very little spend before it was paused. What it usually means: No verdict at all, only a missing test
This is the most useful habit in the whole report. "It did not work" teaches nothing. "It got clicks but nobody added to cart" tells the next writer exactly where to look.
Be strict about the last row. An ad that barely spent did not fail. It was never tested. We cover why in spread vs concentrated spend. Recording it as a loss quietly removes an idea from your future that never got a chance.
Which dead ads are worth bringing back?
Ads that died for a reason that has nothing to do with the idea. If the idea never got a fair test, or got cut while it was still working, it is a revival candidate.
The main candidates we look for:
- Paused while still converting. These are the biggest misses. An ad with real sales was turned off instead of moved to the Graveyard campaign. Proven sales are sitting idle. Move it to the capped campaign.
- Starved, never tested. Ads that barely spent because a sibling took the budget, or because a promoted ad never got delivery in Scaling. Give the idea a fair test, often alongside other starved ads in a runoff.
- Faded debuts. Ads with a strong first stretch that slid below goal. The idea worked. The audience got tired. These often belong in the Graveyard campaign, or deserve a new version for a different persona.
- Near the mark. Ads just under goal that were still gaining when they were cut. These are opportunities, not losers.
- Good hook, weak middle. Ads with a strong outbound CTR but poor follow through. Keep the opening, rebuild the rest.
- Variants that were really a different test. A bundle or offer version of an ad is not the same test as the original. If it was judged as a copy, it may deserve its own verdict.
Not everything comes back. An ad that already failed under a cost cap has had its second life. We call that a second death, and it stays dead. Revival is for ideas that were cut unfairly, not for ideas that lost fair and square.
What patterns should you look for across dead ads?
Look for what the dead ads have in common that the winners do not. One loser is noise. Ten losers with the same trait are a lesson.
Group the dead ads and compare them against your winners on the same traits:
- Hook type. Do question hooks keep dying while bold claims keep working?
- Angle. Is one product truth tested again and again with no result?
- Awareness stage. Do ads that open on the product fail with cold viewers who needed the problem first? Our guide to awareness stages in ad scripts explains why that happens.
- Creator and persona. Does one type of person on camera keep losing, while another keeps winning?
- Format and length. Do long ads, statics or certain edit styles cluster in the dead pile?
- Cause of death. Does most of your cemetery die at the same step? If most ads get clicks and no carts, the problem may be the landing page, not the creative.
Two kinds of pattern deserve their own label. Winner DNA is the set of traits your winners share. Repeat offenders are traits that keep showing up in the dead pile. Both should be written down and carried from one report to the next, so a pattern that appears again is noticed, not rediscovered.
Be careful with small numbers. A trait that shows up in a couple of dead ads may be chance. A trait that shows up across many, over several months, is worth acting on. When in doubt, call it a hunch and test it on purpose rather than banning it.
How do you turn dead ads into a better brief?
Write each strong pattern as a plain rule, attach it to the next brief, and test it on purpose. A lesson that never reaches the writer has not been learned.
Here is how we move lessons from the report into production:
- Write the lesson as one sentence. "Ads that open on the product die with cold viewers. Open on the problem." Short enough to remember.
- Sort it as a do, a don't or a test. A strong, repeated pattern becomes a rule. A weak one becomes a test to run.
- Put it in the brief, not just the report. The writer, editor and creators need to see it. A Graveyard Report that only the media buyer reads changes nothing.
- Pair every don't with a do. Telling a creator what not to do leaves a hole. Tell them what to try instead.
- Check the next report. Did the new rule show up in fewer dead ads? If not, the lesson was wrong or was not used.
Dead ads are also the best source of persona ideas. An ad that failed with one type of buyer may be exactly right for another. If the cause of death was "wrong person," the fix might be persona scaling, not a new idea.
How often should you run a Graveyard Report?
On a slower clock than daily optimization. Monthly works for most accounts, with a lighter check in between if you test a lot.
The report needs enough dead ads to show patterns, and enough time for the Graveyard campaign to show what it can do. A capped campaign spends slowly by design, so reading it over a short window will under read it. We look at the Graveyard campaign over a few months and at testing over the most recent month.
Carry ads forward from one report to the next. An ad that qualified for revival last time but barely spent this time should not drop off the list just because it went quiet.
How does the Graveyard Report connect to the rest of the account?
It sits at the end of the cycle and feeds the start of the next one. Testing produces winners and losers. Scaling grows the winners. The Graveyard Report turns the losers into better tests.
It also helps explain bigger problems. If no new winners are coming through, the Graveyard Report often shows why: every test dying at the same step, ideas being cut before they were tested, or the same angle failing month after month. That makes it a key input when you work on a winner gap.
It pairs well with two other reviews. How to read a Scaling Report covers what happens to the winners. What is a good creative hit rate shows whether the share of tests that win is getting better. Read together, the three tell you whether your testing machine is learning.
Key takeaways
- The Graveyard Report reviews dead and dying ads to find mistakes, revival candidates and lessons for the next brief.
- Sort every dead ad by how it died: hook, intent, offer, cost, fatigue, or never tested. The cause matters more than the death.
- Revive ads that were cut unfairly: paused while converting, starved, faded debuts and near the mark. Leave second deaths dead.
- Look for patterns across many dead ads, and track winner traits and repeat offenders from report to report.
- Put each lesson into the next brief as a plain rule or a planned test, and check whether it worked.
- Run it monthly, read the Graveyard campaign over a longer window, and carry candidates forward.
FAQ
Is the Graveyard Report only about the Graveyard campaign? No. It checks whether the Graveyard campaign is being used properly, but it also studies ads that never sold. The goal is to learn from every dead ad, not just the ones with a pulse.
What is the difference between a dead ad and an untested ad? A dead ad got a fair amount of spend and lost. An untested ad was paused before it spent enough to say anything. Treat them very differently. Untested ads are revival candidates. Dead ads are lessons.
Should I ever relaunch an old losing ad as is? Rarely. If it lost a fair test, running it again usually gives the same result. Relaunch only when the first test was unfair, and change what caused the death when the idea itself was sound.
Who should read the Graveyard Report? Everyone who makes ads. The media buyer runs it, but the lessons belong to the writers, editors and creators who make the next round.
How many dead ads do I need before patterns mean anything? More than a handful. One or two losers with the same trait can be chance. When a trait keeps showing up over several months, act on it, or at least test it on purpose.
Want to know what your dead ads are telling you?
If your account has a long list of paused ads that nobody has looked at since, there may be both profit and lessons in that pile. Our free Meta account audit includes a full account audit plus net new reach and cumulative reach reports. Request your free audit here.
Ready for your free Meta audit?
We will review your account and run net new reach and cumulative reach reports, free.
Check what you learned
Tap an answer to see if you got it right.
1. What does the Graveyard Report review?
2. What matters more than the death itself?
3. Which ads should stay dead?
Ready for the full test? Take the Media Buying exam and get certified. Want this run on your account? Get my free audit.
Module 1: Media Buying · article 13 of 19
Next: How to Tell if Your Meta Budget Changes Actually Worked
Previous: What Is a Graveyard Campaign in Meta Ads, and Why Keep Losing Ads?
About the author
Jake Mehani, founder of Creative Analytics
Jake has spent 15 years running Meta and paid social ads for consumer brands. Meta published his RBX Active Advantage+ case study, he taught the DMAnc Meta and TikTok ads masterclass, and he spoke at Motion's 2024 Creative Strategy Summit. His senior team runs media buying, creative and measurement like an in house team.