Why Is My Promoted Winner Not Spending in the Meta Scaling Campaign?

Short answer: When you move a winning ad from creative testing into a Scaling Campaign, Meta's budget system usually keeps feeding the ads that already sell there, so the new ad gets little or no spend. Some winners are also starved because they sold mostly to people who had already seen your ads, which makes them behave like remarketing ads. The fix is to check every promoted ad's share of spend on a schedule, and give starved winners a protected place to compete.

We call these ads "dead winners." They did everything right in testing. Then they got promoted and nothing happened. Nobody turned them off, nobody judged them, and nobody noticed. This article explains why that happens, how we find them, and what we do about each one.

What is a dead winner in a Meta Scaling Campaign?

A dead winner is an ad that proved it could sell profitably in creative testing, was promoted into a Scaling Campaign, and then received too little spend to ever be tested there.

It is not a failed ad. A failed ad got a real chance and lost. A dead winner never got the chance. That difference matters, because the two need opposite actions. You kill a failed ad. You rescue a dead winner.

In our setup, every account runs three campaigns we call the CTSG: Creative Testing, Scaling and Graveyard. (You can read the whole model in the CTSG system.) Creative Testing finds winners on cold audiences. Scaling is where winners go to spend real money. The Graveyard holds ads that sell, but not at a profitable price. The handoff between the first two is where dead winners are born.

Why do promoted ads get no spend in a Scaling Campaign?

There are two main reasons: the incumbents hold their spot, and the new ad's buyers may not be the kind of people the Scaling Campaign is built to find.

Problem one: incumbents hold their perch

Most Scaling Campaigns run on Campaign Budget Optimization, or CBO. With CBO, you set one budget at the campaign level and Meta decides how to split it across the Ad Sets and ads inside. Meta leans heavily on history. An ad that has sold steadily for weeks has a long track record. A newly promoted ad has none inside this campaign.

So Meta keeps doing what has worked. The established winners keep most of the budget, and the newcomer gets crumbs. This is rational behavior from the algorithm's point of view. It just happens to mean your new winner may never get enough delivery to show what it can do.

Problem two: the ad sold like a remarketing ad

The second reason is about who bought from the ad during testing. We track daily frequency in creative testing. Frequency is the average number of times each person saw an ad. On a single day, a frequency very close to one means the ad is mostly reaching people for the first time. A higher daily frequency means the ad is reaching people who have already seen your ads.

When an ad's sales in testing came on high frequency days, it was selling to people who already knew you. That ad sells like a remarketing ad. And a prospecting Scaling Campaign, built to find new buyers, often will not give it much spend.

Low frequency is a ticket in, not a guarantee. An ad that sold to new people has a better chance of getting budget in Scaling, but it still has to win the auction against incumbents.

One important habit: always judge frequency by single days, not by a multi day blend. Over a week, frequency always drifts higher because the same people stack up across days. A weekly number will make nearly every ad look like it is hitting warm audiences. The daily view tells the truth.

How do you measure whether a promoted ad is getting its share?

Compare the ad's recent spend to the average spend per ad in the same campaign. If it is well below average, it is not getting its share.

Here is the simple version of how we think about it. Take the campaign's spend over the last week. Divide by the number of ads that actually spent. That gives you "normal spend" for one ad. Then compare your promoted ad to that number.

  • At or above normal, or among the top spenders in the campaign: the ad is getting spend.
  • Below normal: low spend.
  • Far below normal and in the bottom half: very low spend.
  • Almost nothing, or zero impressions: no spend.

If the ad lives in more than one Scaling Campaign, judge it by its worst read. An ad that spends in one campaign and starves in another still has a problem worth fixing.

Give new promotions a few full days before reading them. On day one, all you learn is whether Meta started delivering at all.

How do you sort promoted ads so you know what to do?

Sort every promoted ad from the last month into a small set of groups. Each group gets its own action, so nothing is left floating.

  • Getting spend. What it means: Promoted and receiving its share. What to do: Judge it like any scaling ad
  • Winning in testing, not getting spend in Scaling. What it means: Still profitable in testing, starved in Scaling. What to do: Rescue it (see below)
  • Not getting spend, testing Ad Set paused. What it means: Was a winner in testing, now has nowhere to spend. What to do: Rescue it if its testing numbers were at goal
  • Not getting spend, failing in testing too. What it means: Below goal in both places. What to do: No action, let it go
  • Too new. What it means: Only a few days in Scaling. What to do: Note the date to judge it

The core rule is that every promoted ad prints in exactly one place. If you see the same ad in three lists, nobody knows what to do with it.

How do you rescue a winner that is starving in Scaling?

The action depends on that ad's daily frequency in testing. Low frequency winners get a protected Ad Set inside Scaling. High frequency winners go to retargeting.

Low frequency winners: give them a runoff

If the ad sold to mostly new people and is still winning in testing, it deserves a fair fight. We break starved winners out into their own Ad Set inside the Scaling Campaign. Jake calls this the Starvation Runoff, or a Runoff Ad Set in Scaling.

The trick is that CBO campaigns still let you set a minimum and a maximum daily spend per Ad Set, even though the budget is shared. So the newcomer Ad Set gets a small guaranteed floor and a cap.

  • The floor makes sure every promoted winner gets a real test.
  • The cap protects the incumbents' budget.

Keep the floor small relative to the whole campaign. Spend minimums limit how freely Meta can optimize, so the guardrail should open a door, not steer the campaign. An ad that proves itself in the runoff moves into the main Ad Set. An ad that fails its verdict there is cut like any other.

When you see two or more starved winners at the same time in one account, that is the signal to build the runoff. You can read more about runoffs in the runoff Ad Set article.

Also check that the ad is in every Scaling Campaign it belongs in. Sometimes the "starvation" is just a missing copy.

High frequency winners: send them to retargeting

If the ad sold mostly to people who had already seen your ads, a runoff in prospecting is the wrong fix. The ad sells like a remarketing ad, so put it where remarketing happens. Add it to your retargeting campaign if it is not already there.

How long should a promoted ad get before you judge it in Scaling?

Give newly promoted ads about two weeks, not the fast clock you use in testing. In the first day or two you only learn whether the ad gets any spend at all.

Creative testing runs fast. Scaling runs slow. Do not mix the two clocks. A slow start in Scaling is often just Meta learning. A week of spend with no sales does not always mean failure.

That does not mean you wait two weeks with your eyes closed. We use early warning and early kill points that are set as multiples of a benchmark CPA. CPA, or cost per acquisition, is what you pay for one sale.

The benchmark is the lower of two numbers: the CPA of the top spending ad in the Scaling Campaign, and the campaign's average CPA, both over a recent period. Taking the lower one guards against a tired incumbent making the bar too easy.

  • At a lower multiple of the benchmark spent with zero sales, watch closely. A good ad can still show nothing at that point by chance.
  • At a higher multiple with zero sales, the odds that a good ad shows nothing drop low enough to cut it.

When you are unsure in between, look upstream. Is the ad getting adds to cart at a healthy cost compared to the campaign? Then the funnel is filling and sales may be coming. Healthy outbound clicks but no adds to cart means traffic without intent. Expensive outbound clicks and weak outbound CTR (the share of people who click out to your site) means the ad is not stopping anyone, and you can cut with confidence.

What if an ad hits the end of its window but barely spent?

Do not record it as a fail. An ad that reaches the end of its window without a minimum amount of delivery was never tested.

The verdict has several parts: time, delivery, money and reach. If the calendar ran out but the delivery never happened, flag it as "never tested" and decide whether to restart it or release it. Logging a starved ad as a loser teaches you the wrong lesson about the creative.

Should you pause the testing copy once an ad is in Scaling?

No, not just because it was promoted. Judge the testing copy on its own numbers.

When an ad is promoted, the original usually keeps running in creative testing. Raising its budget there is vertical scaling. The copy in Scaling is horizontal scaling. Running more than one live version of the same ad is fine. We call this Best Version Running: compare the versions and keep the best ones live.

If both copies are converting at goal, leave both on. Two profitable copies is a feature. If the testing copy is failing on every source you trust, pause it, and note that the ad lives on in Scaling. If the Scaling copy is not delivering, the testing copy stays regardless, because it may be the only version actually selling.

Key takeaways

  • A promoted ad that gets no spend in Scaling was never tested. Treat it as a dead winner to rescue, not a loser to kill.
  • CBO favors incumbents with history, so newcomers often starve.
  • Daily frequency in testing tells you where a winner will find budget: low frequency winners go to a Runoff Ad Set in Scaling, high frequency winners go to retargeting.
  • Measure each promoted ad's spend against the average per ad in its campaign, and sort every promotion into one clear group.
  • Judge promoted ads on a slower clock than testing, with early warning and early kill points set as multiples of the account's own benchmark CPA.

FAQ

Why does Meta keep spending on my old ads instead of the new winner?With Campaign Budget Optimization, Meta favors ads with a proven record inside that campaign. Your new winner has no history there yet. Without a protected Ad Set or a spend floor, it may never get enough delivery to build one.

What is a Runoff Ad Set in Scaling?It is a separate Ad Set inside your Scaling Campaign that holds newly promoted winners, with a small minimum and a maximum daily spend. The minimum guarantees each winner a real test. The maximum protects the budget of the ads already working.

Does high frequency mean my winning ad is bad?No. It means the ad sold mostly to people who had already seen your ads. It still earned its promotion. It is just more likely to find budget in a retargeting campaign than in a prospecting Scaling Campaign.

How do I know if a starved ad failed or was never tested?Check whether it reached a minimum amount of delivery, set against your benchmark CPA. If the window ran out before it spent that much, it was never tested. Flag it to restart or release, and do not count it against the creative.

Want a second set of eyes on your Scaling Campaign?

If you suspect winners are dying quietly after promotion, we can look. Our free Meta account audit includes a full account audit plus net new reach and cumulative reach reports, so you can see where your spend is actually going. Request your free audit here.

Keep reading

More on the CTSG system:

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.

Get your free audit

Check what you learned

Tap an answer to see if you got it right.

1. Why do newcomers often starve in Scaling?

Older ads have history, so Meta keeps feeding them.

2. Where should a low frequency winner go if it starves?

Low frequency winners reach new people, so give them a fair shot in Scaling.

3. How should promoted ads be judged?

Scaling needs more time, with warning and kill points tied to the account's benchmark.

Ready for the full test? Take the Scaling exam and get certified. Want this run on your account? Get my free audit.

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.

Jake on LinkedIn  |  Case studies  |  Free Meta audit

Previous
Previous

When Should You Move a Winning Ad From Creative Testing to Scaling in Meta?

Next
Next

How to Read a Scaling Report for Meta Ads