What Is the CTSG System? Creative Testing, Scaling and Graveyard for Meta Ads

Short answer: The CTSG system is a way to run a Meta ad account on three campaigns, each with one job. Creative Testing finds new winning ads by showing them only to cold audiences. Scaling gives proven winners real budget, and the Graveyard keeps ads that sell, but too expensively, running under a cost cap instead of turning them off.

Most messy accounts mix testing, scaling and rescue in one place, so nobody can tell what works. CTSG splits the jobs apart.

What does CTSG stand for?

CTSG stands for Creative Testing, Scaling and Graveyard. The CTSG system is a three campaign structure for Meta ads where each campaign has one job: find winners, fund winners, or rescue ads that sell at a cost that is too high.

It is an operating system, not just a setup, because it comes with a daily routine: review all three campaigns in order and turn what you see into tasks. See how to run a daily Meta optimization.

Why use three campaigns instead of one?

Because one campaign cannot answer three different questions at once. When testing and scaling share a campaign, the old winners eat the budget and new ads never get a fair look.

Meta's delivery system likes what already works. Put a new ad next to one that has sold for months and Meta keeps feeding the old one. The new ad gets a few dollars, makes no sales, and looks like a loser. You never learn if it was good.

The fix is to give each job its own space:

  • Creative Testing. Its one job: Find new winners. Who it shows ads to: Cold audiences only, with strict exclusions. How fast it judges: Fast, within days
  • Scaling. Its one job: Give winners real budget. Who it shows ads to: Everyone except past customers. How fast it judges: Slower, about two weeks for new arrivals
  • Graveyard. Its one job: Keep ads that sell, but too expensively. Who it shows ads to: Broad, under a cost cap. How fast it judges: Ongoing

Each campaign runs on its own clock and its own rules. That separation is the whole point.

How does the Creative Testing campaign work?

Creative Testing shows new ads only to people who have not bought and, in most cases, have not visited the site. It is where every new ad starts, and where most ads end.

Why cold audiences only? If you test ads on warm audiences (people who visited your site or already follow you), many of them were likely to buy anyway. A weak ad can look strong just because it was shown to people already halfway to the checkout. So we always exclude past customers, and recent site visitors too. Some accounts also exclude social engagers, especially once reach to new people starts to slow.

One rule here matters more than people think: every Ad Set in Creative Testing must use the same exclusions. If one Ad Set runs to cold people and another runs to warm people, comparing them tells you nothing about the creative.

How is it built? Budget is set per Ad Set, not per ad. Each Ad Set holds a small group of ads, usually three to six, so they compete for the same money. More ads per Ad Set only makes sense when the ads are close variations of each other. One ad per Ad Set is the cleanest test in theory, but it only works if you make a lot of new ads every month.

How are ads judged? Fast. Testing is meant to fail most ads quickly and cheaply. An ad that spends a meaningful amount (a multiple of your target cost per sale) with zero purchases is paused. An ad that sells is judged on what it costs to get a new customer and the return it brings, not on Meta's numbers alone. We cover that in depth in how a creative testing ad gets judged.

What counts as a winner? More than one sale, at or better than your goal, from people who are mostly seeing the ad for the first time. That last part matters. We look at daily frequency (frequency is how many times, on average, each person saw the ad). An ad that sells while its daily frequency stays close to one is reaching new people. An ad that sells only once frequency climbs is mostly selling to people who already saw other ads, which behaves more like remarketing.

How does the Scaling campaign work?

Scaling is the catch all. Proven winners from testing get copied in, and the campaign spends the bulk of the budget.

Who does Scaling show ads to? Everyone except past customers. That is the only exclusion. Site visitors and social engagers stay in the pool on purpose, because Scaling also does the job most accounts give to a separate remarketing campaign. It reaches cold people and warm people who have not bought yet, and lets Meta's budget system decide the mix.

Why not run a separate remarketing campaign? A dedicated remarketing campaign is something we investigate whenever we see one. Remarketing usually shows the best looking return in an account, and it is often the least real. Many of those buyers were going to buy anyway. Carving out fixed budget for them is usually less efficient than letting the Scaling campaign decide. If you want to know whether your remarketing spend does anything, read is your remarketing spend doing anything.

How is Scaling judged? On a slower clock. A newly promoted ad gets about two weeks before a verdict, because in the first day or two you only learn whether it gets any spend at all. Early alarms are based on multiples of the campaign's own cost per sale, so the thresholds adjust to each account's economics instead of using one dollar number for everyone.

The incumbent problem. Inside Scaling, old winners hold their spot. A new ad that crushed testing often gets little budget because the established ads keep winning the auction. That is not always a sign the new ad is bad. The fix is a runoff: pull the starved newcomers into their own Ad Set with a small, guaranteed share of spend so they get a fair test. More on that in runoffs.

What is the Graveyard campaign?

The Graveyard is a cost cap campaign for ads that have proven they can sell, just not at a cost you can live with. Instead of turning those ads off, we move them to a campaign where Meta is told the most it can pay for a sale.

The problem it solves: many ads make sales at a cost that is somewhat above goal. Pausing them throws away a creative that people clearly respond to. Leaving them in testing burns money.

The solution: a cost cap. Meta only spends on those ads when it can find sales at or under the cap. Some days it spends a lot, some days almost nothing. That is fine, because every dollar it does spend is at a price you accepted.

Two rules keep the Graveyard honest:

  1. Only ads with sales go there. An ad with zero purchases has nothing for a cost cap to rescue. It gets paused.
  2. Check that it delivers. If the Graveyard spends almost nothing, the cap may be too tight for the market. That is a signal about what sales actually cost, not a bug. Raise the cap in small steps and note where spending starts.

We go deeper in the Graveyard: why we keep losers.

How do ads move between the three campaigns?

Every ad starts in Creative Testing and ends up in one of four places. The flow looks like this:

  1. Zero sales after a fair amount of spend: paused in testing.
  2. Sales, but both cost per new customer and return are below your floor: moved to the Graveyard.
  3. Sales where one number passes and the other fails: held and checked again soon, because one number alone can mislead.
  4. More than one sale, at goal on both numbers, reaching mostly new people: promoted to Scaling.

Promotion has one more step people skip: before you promote, check whether the ad is already in Scaling. Accounts with many people touching them often have duplicate copies of the same ad spread around. Look at every Scaling campaign and the Graveyard, by exact ad name, before you add it again.

A promoted ad often keeps running in testing too. That is fine. Two copies of the same ad, both profitable, is a feature, not a conflict. Judge each copy on its own numbers. When one copy is clearly failing and the other works, pause the failing one. The testing copy usually wears out first.

Why do the three campaigns run on different clocks?

Because the cost of a wrong answer is different in each. In testing, the risk is wasting money on ads that will never work, so you judge fast. In Scaling, the risk is killing a proven winner before Meta finds its audience, so you judge slower. Swap the clocks and you either kill new winners in their first week or pour money into ads that were never going to work.

Jake's rule of thumb is that the structure is the constant and the dollars calibrate themselves. Fast in testing, about two weeks in Scaling, alarms at multiples of the account's own cost per sale. The exact numbers come from each account's goal, never from someone else's account.

How do you know if the CTSG system is working?

Look at outcomes, not settings. The checks we use:

  1. Cost per new customer, from more than one source. Meta's own reporting and a third party attribution tool often disagree. We read both, and the new customer numbers matter most.
  2. Net new reach. Is your spend reaching people who have never seen your ads, or the same people again and again? See net new reach.
  3. New versus returning delivery in Scaling. Meta can show how much of your delivery goes to new, engaged and existing audiences. A healthy Scaling campaign goes mostly to new people.
  4. A conversion lift test. This is the only way to prove Meta is causing sales instead of just being present when they happen. If you have never run one, it is the first thing we would recommend. See conversion lift tests.
  5. Creative volume. Testing only works if you feed it. A month with no new ads is a gap at any spend level.

What are the most common mistakes with this setup?

  • Testing to warm audiences. The results look great and mean nothing.
  • Different exclusions across testing Ad Sets. You cannot compare the ads.
  • No purchaser exclusion in Scaling. You pay to show ads to people who already bought.
  • Pausing an Ad Set because of its average. An Ad Set can fail as a whole while one ad inside it is working. Pause the failing ads, keep the winner.
  • Judging on Meta alone. Meta can over count and under count. Check another source before big calls.
  • Being too soft. Holding a losing Ad Set one more day, again and again, is how accounts bleed. We call this the prisoner of hope problem.

Key takeaways

  • CTSG means Creative Testing, Scaling and Graveyard: three campaigns, three jobs.
  • Creative Testing runs only to cold audiences, with identical exclusions in every Ad Set, and judges fast.
  • Scaling excludes only past customers, gives winners budget, and judges new arrivals over about two weeks.
  • The Graveyard keeps ads that sell too expensively alive under a cost cap instead of killing them.
  • Thresholds come from each account's own goal, never from a universal number.
  • Judge the system on new customers, net new reach and lift, not on how neat the setup looks.

FAQ

Is CTSG the same as an Advantage+ setup?No. CTSG is a way of dividing jobs between campaigns. You can use Advantage+ campaigns inside it, especially for Scaling. The structure is about separating testing, scaling and rescue, not about which campaign type Meta offers.

How many ads should go in a Creative Testing Ad Set?Usually three to six. Enough that they compete for the same budget, few enough that each one has a real chance to spend. Go higher only when the ads are close variations of each other.

Do I need a separate remarketing campaign with CTSG?Usually not. Scaling keeps site visitors and engagers in its audience, so it already covers warm people who have not bought. A separate remarketing campaign tends to report great numbers that are not real new sales.

What if my Graveyard campaign barely spends?That usually means your cost cap is below what sales really cost right now. Raise the cap in small steps and record where it starts to spend. The point where it delivers tells you something real about your market.

How long should a new ad stay in Creative Testing?Only as long as it takes to reach a fair verdict. Testing is meant to be fast. Read the first verdict: how early you can call a new ad for how we decide when an ad has had enough.

If you want to see how your own account stacks up against this structure, we offer a free Meta account audit. It includes a full account audit plus net new reach and cumulative reach reports, so you can see where your money actually goes. You can request one at our contact page.

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. What does CTSG stand for?

Three campaigns, three jobs: test new ads, scale winners, and keep pricey sellers alive in the Graveyard.

2. Who should Creative Testing run to?

Cold audiences show whether the creative works. Same exclusions everywhere keep the test fair.

3. Where do kill thresholds come from?

Every account has its own target CPA, so its lines come from that goal.

Ready for the full test? Take the Media Buying 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

How to Set Up a Creative Team Workflow for Meta Ads

Next
Next

How to Run a Daily Meta Ads Optimization (Without Overreacting)