Short answer: The prisoner of hope is the habit of keeping a losing ad or Ad Set running because it might turn around. Holding is sometimes right, but every hold on something clearly below goal should be flagged, with the case for pausing written out next to it, so the person making the final call sees both sides. That one habit turns soft calls into real decisions.
Every media buyer knows the feeling. An Ad Set had a great week. Now it is slipping. It made one sale yesterday. Maybe tomorrow is the day it comes back. So you leave it on. Then you leave it on again. A week later it has quietly eaten a real chunk of budget, and nothing came back.
Jake's way of putting it is that being soft in media buying is the prisoner of hope, and it gets accounts into trouble. This article explains how we catch it.
What is the prisoner of hope in media buying?
The prisoner of hope is a media buyer who keeps funding a losing ad or Ad Set because of what it might do, instead of what it is doing.
It is not a lack of skill. It is a very human bias. Turning something off feels final. Leaving it on feels safe, because you have not made a mistake yet. But leaving it on is also a decision, and it costs money every day.
The problem is not that buyers hold. Sometimes holding is correct. The problem is that holds tend to happen silently. Nobody writes down why. Nobody sets a date to check. And nobody makes the case for the other side.
Why is holding a losing ad so expensive?
Because a hold has no natural end. A pause is one decision. A hold repeats every day until someone changes it.
Look at how the costs stack:
- Direct waste. Every day a failing Ad Set runs, it spends at a cost per sale above your goal.
- Budget that could have gone elsewhere. That same money could have funded a new test or a proven winner.
- Bad data. A long run of weak performance blends into your averages and makes it harder to read the account.
- Habit. The second hold is easier than the first. The third is easier still.
No single hold looks dangerous. A dozen of them, running for weeks, is how good accounts drift off target.
How do you know when a hold is really hope?
When the numbers are clearly below goal on the source you trust, the test has had enough spend to judge, and nobody has changed anything recently. Then a hold is hope, not patience.
We use three questions before we call a hold "hope."
Has it spent enough to judge? A brand new test under its first checkpoint is not a prisoner of hope case. It simply has not had a fair chance yet. We cover that checkpoint in when to call a new ad.
Is it clearly below goal, not just slightly? An Ad Set a hair under goal is a normal hold. An Ad Set well below goal on your trusted attribution source over the last several days is something else. We set the line as a percentage below the goal, so tiny misses do not trigger it.
Was it changed recently? If someone cut the budget or made a change in the last few days, the change has not had time to land. Meta needs a few days to attribute sales after a change. Judging too soon means you never learn whether the change worked. In that case, mark it "too recent" and write down the date to judge it.
If all three are true, the hold gets flagged.
Which holds should always be flagged, no matter what?
Some situations are so often hope that we flag them every time, even if the basic test is not met.
- A pass that rests on one order. If an Ad Set only looks good because of a single sale, or a single unusually large order, it is not really passing. One order is luck until proven otherwise.
- The main ad is gone and a weak ad is carrying it. Many Ad Sets are driven by one strong ad. If that ad has been paused or faded, and the rest of the Ad Set is weak, the Ad Set is living on its past. You can read more about this pattern in kill thresholds.
- The same Ad Set on hold two runs in a row. One hold is a judgment call. Two in a row is a pattern.
- Extending a spread Ad Set. When spend is spread thin across many ads with no clear winner, extending the test is a soft call by nature. See spread vs concentrated spend.
When should you never flag a hold as hope?
When the hold is backed by real evidence the ad is working. Flagging those would make the flag meaningless.
Three cases never get the flag:
- Meta is under reporting. If Meta shows the Ad Set below goal but your third party attribution shows it at or above goal, that is not hope. It is a real winner Meta cannot see. We explain this in Meta under reporting.
- New tests under the first checkpoint. Too early to judge, with or without a sale.
- A deliberate extension for a low frequency test. If a test is reaching mostly new people and you extended it on purpose under your own rules, that is a plan, not a hope.
The point of the flag is to catch soft calls. A hold with strong evidence behind it is not soft.
What does a prisoner of hope flag actually say?
It states the case for pausing or cutting, with the deciding numbers and the exact action, written last so it is the final thing you read.
This is the heart of it. The flag does not override the hold. It sits under it and argues the other side. Something like:
Prisoner of hope: this Ad Set has spent past its checkpoint, its only "pass" rests on one large order, and two of its three ads have no sales. Pause those two ads. If the third has fewer than two orders by the next check, pause the Ad Set.
Now the person making the decision has both sides in front of them. They can still hold. But they are holding with their eyes open, and they know exactly what they are betting on.
We also pair every hold, pause and cut with a short "Why" and "Next":
- Why: one plain sentence with the recent numbers against goal and the one fact that settles it.
- Next: the action, plus the softer option if there is one, and the date to check again.
A hold without a "Next" date is just hope with no exit.
Should you cut the budget instead of pausing?
Often, yes. A budget cut is a good middle path for a flagged hold on a larger Ad Set.
If an Ad Set is clearly below goal but you have a real reason to keep it alive, lowering its daily budget limits the damage while you wait. We flag a modest cut on flagged holds above a small budget level. It does not change the call. It just puts a cheaper option on the table.
On very small Ad Sets, a cut does not make sense. There is not much to cut, and a tiny budget makes results even harder to read. At that level, the honest choices are keep or pause.
How many flags is too many?
If every hold gets flagged, people stop reading the flags. So we cap them.
We limit the number of prisoner of hope flags per account and rank them by recent spend, biggest first. The biggest bets get the hardest look. Smaller holds still print with their Why and Next, just without the flag. And when the cap cuts some out, we say how many, so nothing is hidden.
This keeps the flag rare enough to matter.
How do you know if your holds were right?
Score them. Go back after the change has had time to land and check whether the hold, cut or pause was the right call.
A flag only helps if you learn from it. We review budget decisions on a regular schedule and score each one as right or wrong against the account's own goals. Over time, that shows whether your team holds too long, cuts too fast, or gets it about right. We walk through how in scoring your own budget calls.
Key takeaways
- The prisoner of hope is holding a losing ad because it might turn around. It is common, human, and expensive.
- A hold is a decision that repeats every day. Treat it like one.
- Flag holds that are clearly below goal, past their first checkpoint, and not recently changed.
- Always flag passes that rest on one order, weak carriers after the main ad is gone, and repeat holds.
- Never flag real winners that Meta under reports, or new tests that have not had a fair chance.
- Every flag should state the case for pausing, with the numbers and the exact action, so the final call is made with both sides visible.
FAQ
Is it ever right to hold a losing Ad Set?Yes. A test that has not reached its first checkpoint, an Ad Set that Meta under reports, or one with a recent change that has not landed yet can all be fair holds. The goal is not to ban holds. It is to make sure every hold is a conscious decision.
How long should I wait after a budget change before judging it?Long enough for sales to attribute. Meta needs a few days after a change for conversions to show up. Judging too soon means you never learn whether the change worked.
What is the difference between a hold and a prisoner of hope?A hold has evidence behind it and a date to check again. A prisoner of hope is a hold on something clearly below goal, past its checkpoint, kept alive mostly on the chance it turns around.
Should I just pause everything below goal instead?No. That throws out real winners Meta under counts, and new tests that have not had a fair chance. The flag exists so you do not have to choose between being reckless and being soft.
Who should make the final call on a flagged hold?Whoever owns the account. The flag does not change the decision by itself. It puts the case for pausing in front of the decision maker so they choose with full information.
Want a fresh look at what you are holding?
If your account has Ad Sets that have been "about to turn around" for weeks, a second set of eyes helps. Our free Meta account audit includes a full account audit plus net new reach and cumulative reach reports. Request your free audit here.
Keep reading
More on the CTSG system:
What Is the CTSG System? Creative Testing, Scaling and Graveyard for Meta Ads
Why Is My Promoted Winner Not Spending in the Meta Scaling Campaign?
What Is a Graveyard Campaign in Meta Ads, and Why Keep Losing Ads?
Does Meta Under Report Conversions? When Not to Pause an Ad That Looks Bad
See every article on the CTSG system
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 is the prisoner of hope?
2. How should you treat a hold?
3. Which ads should NOT be flagged?
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 10 of 19
Next: What Is a Runoff Ad Set in Meta Ads and When Should You Use One?
Previous: Pausing Ads vs Cleaning Up Ads in Meta: Why They Are Two Different Jobs
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.