How to diagnose if you have a lazy Meta Ad Account
· Paid Social

Phil here, Director of Paid Social. I spoke to two brands this week with the same problem.
The first: an activewear brand, twenty years in market, top-rated products, customers who say the brand changed their life. Return on ad spend over 4. And nobody new knows they exist.
The second: an edibles brand, massive wholesale presence, strong organic sales. Half of the last six months of ad spend going to existing customers. They knew something was off. They just didn't know how to see it.
Different categories. Different sizes. Different teams. Same root cause.
Meta had gotten lazy on both of them.
What a lazy account looks like
Meta's algorithm is built to find conversions. It's very good at this. The problem is that "finding conversions" and "finding new customers" are not the same job.
When your account has a history of existing customers converting - and most established brands do - Meta learns that these are your buyers. It knows where to find them. It goes back to them. Repeat customers convert at higher rates and lower CPA than cold audiences. The algorithm rewards itself for finding them, and then finds more of them.
Platform ROAS looks healthy. Sometimes it looks great. Because existing customers do convert well, and Meta takes credit for every conversion it can touch.
What isn't visible in the blended number: the percentage of your spend going to people who were going to buy anyway. The new customer acquisition rate quietly flattening. The top of the funnel narrowing while the bottom appears fine.
This is a lazy account. Not broken — optimising for the wrong thing.
Why strong brands are most at risk
This is the counterintuitive part.
The brands most likely to have this problem are the ones doing the most things right. Great product. Strong retention. Loyal customers. High repeat purchase rate. The things that make a brand genuinely good are the same things that give Meta a warm, efficient pool to recycle through indefinitely.
If you have almost no repeat customers, Meta has to go and find new people because there's no shortcut. The account is forced to do acquisition work.
If you have a 60 or 70% repeat customer rate - which is genuinely good for a DTC brand - Meta sees all those familiar buyers and defaults to them. Why work hard to find strangers when the regulars keep showing up?
One of the brands I spoke to this week had spent the first twenty years building exactly the kind of customer loyalty every brand wants. And that loyalty was the thing making it hard to scale on Meta. Not bad ads. Not wrong strategy. Not poor creative. The algorithm was just too comfortable.
The attribution cover-up
The reason this goes undetected for so long is that platform ROAS stays strong throughout.
Existing customers buy efficiently. The CPA on returning customers is significantly lower than on cold audiences. When Meta concentrates spend on them, the numbers that most teams watch stay acceptable. Sometimes they look excellent.
But those sales would largely have happened anyway. Existing customers who love a product don't need an ad to remember to buy again. The ad is a closer - it shows up at the moment of purchase intent that already existed and takes credit. It is not an introducer. It didn't cause a new customer to discover the brand.
I posted about a brand recently that thought they'd scaled spend 33% year on year. They had. But new audience spend only grew 17%. The rest went to existing customers - Zuck quietly increased spend on that segment from $5k to $60k over the period. Nobody on the team made that decision. The algorithm did.
The gap between what Meta claims it drove and what was actually incremental widens the stronger your organic and retention base is. In one of the accounts I looked at this week, there's a reasonable chance that the majority of what Meta attributed to itself over six months was spend on people who would have converted through direct, email, or word of mouth.
We won't know for certain until we test it. But the audit data makes a strong case that the lazy period was long and expensive.
Audience Segments as the diagnostic
The only way to see this without running a holdout test is Audience Segments.
Pull the breakdown in Meta Ads Manager. Set it to show New Audience, Engaged Audience, and Existing Customer spend as a percentage of total each week. Watch how those percentages change over time.
For the edibles brand, when we started tracking this, the new audience percentage had been declining for weeks. Existing customer spend was running at $2,300 in a single week. Then $1,700. Then $1,200. Then, once we turned off the campaign structure that was feeding it, it dropped to $572. This week it's at $240.
The New Audience percentage went from the low forties to 59% in the space of a week.
The account data for the first few months of this process wasn't visible because segments weren't defined early enough. But from the point we started tracking, you could watch the account shift in real time. Week by week. The existing customer spend falling. The new audience percentage climbing. The account, to use Rad's word, healing.
That data only exists because segments were defined before it mattered. If you set them up now, you start building the diagnostic history from today.
What fixing it looks like
There is a transition period. It is not instant and it is not smooth.
When you cut off the easy audiences - stronger exclusions, turning off campaigns that were defaulting to warm pools, forcing the account cold - the CPA goes up before it comes down. The algorithm has to find new people it has never targeted before. Early conversion rates are lower because cold audiences convert at lower rates than existing customers. This is correct and expected.
During this period there are what I'd call martyrs. Ads that ran during the lazy phase that took spend in an environment where the account was defaulting to warm audiences. Some of those ads were probably good. They just never got a fair test. Launching them again once the account is in better shape might tell a different story.
The scaling strategy changes too. When an account is in transition, CBO is not the right tool. CBO assumes you know the account will go after the right audiences and optimise efficiently. During healing, you want more control - ABO, where you can see what's happening at the ad set level and make decisions based on it. Once the new audience percentage is holding at a healthy level and new customer CPA is stable, then you graduate back to CBO with confidence.
For the edibles brand, we're now a few weeks into that transition. Spend is up 11%, new customer revenue is up 20%, new customer CPA is flat. Early data, short window, one sale period in there. But it's pointing in the right direction for the first time in a while.
The incrementality question
Both of these brands asked a version of the same question: what happens if we turn the ads off?
It's an uncomfortable question. The fear is that you turn them off, sales drop, and you can't get the performance back. There's also the learning phase anxiety - the assumption that pausing means starting over.
Here is what I tell brands in this situation: if your current account is too fragile to handle a two-week pause, it is not resilient enough to scale. If you can't turn it off and turn it back on, how are you going to triple your spend? The fragility at the current level becomes catastrophic at 5x the budget.
And if you pause for two weeks and sales don't drop - if they hold flat or barely move - that tells you something important. It tells you the ads weren't the reason people were buying. They were the reason Meta was being paid. Those are different things.
The activewear brand is considering this test. For the edibles brand, we already have a version of the answer in the data - spend came down significantly from a previous period and the organic sales base held. That's the signal. The ads were closing, not introducing.
The check worth running today
Pull your Audience Segments breakdown in Ads Manager. If it's not set up, set it up now. The data is not backdated - you start building history from the day you define the segments.
Once you have it: look at what percentage of your spend is going to New Audiences versus Engaged and Existing. Look at how that percentage has changed over the last eight to twelve weeks. If prospecting was 80% six months ago and it's 65% now, the drift is underway. The account is getting lazier.
Then ask: what's my new customer CPA separate from my blended CPA? Not the number your platform reports. The number in your third-party attribution tool, filtered to first-time buyers only. How does that compare to the blended number? The gap between them is the lazy tax — the premium you're paying to convince your own existing customers to do something they were going to do anyway.
A strong brand and a lazy account are not mutually exclusive. They often go together. The brand loyalty is real. The ROAS is partly fiction.
Read more about CPMr:https://taikundigital.com/phil-kiel/cpmr-calculation
Book a call with me.