The First 30 Minutes in a New Meta Ads Account

Why Most Founders Audit Backwards

Every time I get access to a new Meta Ads account, I want to do the same thing you do: look at the campaigns. See what's running, check the ROAS numbers, understand what's been tested. It's the natural instinct. The campaigns are the product. That's where the money is going.

It's also the wrong place to start.

The campaigns are a symptom. They reflect the decisions that were made - about structure, about tracking, about budget - and if those decisions were wrong, the campaign-level numbers are misleading at best and actively deceptive at worst. I've walked into accounts where the reported ROAS was 5.4x and the business was barely breaking even, because the attribution settings were inflating every number. I've seen accounts where every campaign showed green but 60% of ad sets had never exited learning phase and were never going to.

The campaigns tell you what happened. The five checks below tell you why it happened and whether the numbers mean what they appear to mean.

Here's the sequence. Each check takes five to ten minutes. Done in order, they give you a complete picture before you change a single setting.

Check 1 - Pixel Health and Signal Quality

Start in Events Manager, not Ads Manager. This is the most important check on the list because every other part of the account runs on the quality of data the pixel sends back. A broken or degraded pixel means the algorithm is making optimization decisions on incomplete or inaccurate information. Your ROAS targets, your bidding, your audience building - all of it is downstream of pixel quality.

Go to Events Manager, select your pixel, and look at three things:

• Event Match Quality (EMQ) score for Purchase. This is a 0-10 rating of how well Meta can match the people who triggered your Purchase event to real Facebook profiles. Anything below 6 is a significant problem. A score of 4 means roughly 40% of your conversions are disappearing into a signal void - the algorithm never gets credit for them, which means it never learns who converted and can't find more people like them.
• Event deduplication status. If you're using both browser pixel and Conversions API (CAPI), you need deduplication configured so the same purchase doesn't fire twice. Over-reporting inflates conversion counts and makes your bidding too aggressive for your actual volume.
• Event count trend over the last 28 days. Is the number of Purchase events roughly consistent with what your Shopify or payment processor reports? A significant gap - more than 15-20% - means your pixel is missing events. That's signal loss the algorithm is working around.

Most founder accounts have an EMQ score somewhere between 4 and 7. The most common fixes are adding hashed customer data parameters (email, phone, first and last name) to the Purchase event and implementing server-side tracking via CAPI. The full diagnostic and repair guide is in our piece on Meta pixel signal quality and Event Match Quality - if your score is below 7, that post should be your immediate next read.

Check 2 - Campaign Structure and Learning Phase

Now open Ads Manager. Filter the ad sets view to show active ad sets and look for the Learning Phase indicator in the Delivery column. Then ask a simple question: how many active ad sets are stuck in learning?

In most founder accounts I audit, the answer is somewhere between a third and two-thirds of all active ad sets. This isn't just a minor inefficiency - an ad set that never exits learning phase is an ad set that never fully optimizes. Meta needs 50 optimization events per ad set per week to exit. If your budget and conversion volume can't generate that, you're permanently running a suboptimal algorithm.

The reason accounts end up this way is almost always the same: over-fragmentation. A founder tests an interest target, then a lookalike, then a broad audience, then a retargeting segment, then a different prospecting angle - each in its own ad set, each with its own slice of the budget. Six months later, there are 14 active ad sets each getting $30 per day and none of them generating 50 weekly purchases.

The structural fix is consolidation. Here's how to read the severity:

The full mechanics of learning phase - including the CBO vs ABO decision and how to consolidate without destroying historical data - are covered in our post on Meta's learning phase and account structure. The short version: fewer ad sets with more budget each is almost always better than more ad sets with less budget each, until you're spending enough to fund both.

Check 3 - Attribution Window Settings

This check has ended more than a few conversations about whether an account was actually performing. Meta's default attribution window is 7-day click plus 1-day view. Most founders have never looked at it and have no idea what it means for their reported numbers.

Here's what it means: every purchase that happens within 7 days of someone clicking your ad, or within 1 day of someone merely seeing your ad without clicking, gets attributed to that campaign. The view-through portion is the aggressive one. Customers who would have purchased organically through direct traffic, email, or search get credited to Meta if they happened to see your ad in the last 24 hours.

In Ads Manager, click the Columns dropdown and select "Compare Attribution Windows." This shows you the same campaign's performance under different windows side by side. Compare 7-day click plus 1-day view against 7-day click only. In most accounts, the gap is 15-40%. That gap is how much view-through attribution is inflating your numbers.

I'm not saying view-through attribution is worthless - Meta's ads do create awareness that contributes to purchases that happen through other channels. But you need to know the gap exists before you can make good decisions about budget and ROAS targets. Our full breakdown of how Meta's attribution window works covers the specific settings and how to choose the right comparison window for your business model.

Every Meta account I audit is either using a setting nobody told them about, or they chose a setting without understanding what it claims.

While you're in attribution settings, also check whether you're using campaign-level attribution or ad set-level. Campaign-level is almost always better because it lets Meta's delivery system allocate spend to the ads that are actually converting, rather than forcing budget toward each ad set independently.

Check 4 - Creative Health

Now look at the actual ads. There are two things to check at this stage: frequency and relevance diagnostics.

For frequency, add the Frequency column to your ad-level view and sort by it. Any single creative with a 7-day frequency above 3 in a cold prospecting ad set is approaching saturation territory. Above 5 is a red flag. The symptom shows up as a rising CPM alongside a falling CTR - the platform is paying more to reach an audience that's increasingly tuned out.

For relevance diagnostics, go to Columns and add Quality Ranking, Engagement Rate Ranking, and Conversion Rate Ranking. These three scores are how Meta evaluates your ads against competing ads targeting the same audience. A low Quality Ranking means your ad is being penalized in the auction. A low Conversion Rate Ranking means Meta has learned that people who click your ad don't buy at the rate it expected - which signals a mismatch between the ad's promise and the landing page's delivery.

The pattern these three scores create together tells you where the breakdown is. Below Average in all three usually means the creative itself is weak. Below Average in Conversion Rate Ranking only usually means the post-click experience is failing - your ad relevance diagnostics are intact but something is breaking after the click. This matters because the fix is different in each case.

Check the frequency data alongside the creative age. A creative that's been running for six weeks on a meaningful budget has almost certainly saturated a portion of its audience. If your newest active creative was uploaded more than four weeks ago, creative refresh should be on the immediate action list. The mechanics of creative fatigue and when to rotate are worth reviewing before you make any changes here.

Check 5 - Where Budget Is Actually Going

The final check is the one that consistently surprises founders the most. You know your total spend. You probably have a rough sense of which campaigns are getting what. But do you know where within those campaigns the money is actually landing?

Open the Breakdowns menu and run three cuts: Placement, Age/Gender, and Device. You're looking for budget concentration in places that aren't driving results.

The most common finding is Audience Network eating 15-25% of spend. Audience Network is Meta's off-platform inventory - apps, websites, games. It runs cheap CPMs but in most direct-response accounts, it converts at a fraction of the rate of Instagram or Facebook feed. If you're not watching it, you're funding it by default.

The second common finding is mobile-web conversions being dramatically outperformed by app conversions on the same campaign, or a specific age range that accounts for the majority of spend but a minority of revenue. These are the kinds of structural inefficiencies that don't show up in top-line ROAS but are quietly diluting every number you're looking at.

Our breakdown of how to use Meta's Breakdowns report walks through exactly how to read this data and what actions to take based on what you find. The short version: when you see a placement, age group, or device that has high spend and low conversion rate relative to the rest of the account, you have two choices - exclude it from delivery or create a separate ad set targeting just that segment with creative calibrated for it.

What Comes After the Audit

After running these five checks, you'll typically have a list of two to four problems. Not fifteen. If you're finding fifteen problems, you're probably conflating root causes with symptoms. Fragmented ad sets stuck in learning phase is one root cause. The six bad numbers that flow from it are symptoms.

Fix in order of impact. Pixel health first, because it affects everything else. Campaign structure second, because fragmentation corrupts your creative data and makes it impossible to read what's actually working. Attribution settings third - you need to know what you're measuring before you make any optimization decisions. Creative health and budget allocation are tactical and can be addressed in parallel once the foundation is sound.

The most common mistake founders make when they find problems in an audit is trying to fix everything at once. They restructure campaigns, refresh creative, change attribution settings, and reallocate budget in the same week. Then something improves and they have no idea which change drove it. Or something breaks and they don't know what caused it.

Sequence your changes. Make one structural intervention, let it run for a full week, measure the delta, then move to the next. This is slower than the "blow it all up" instinct but it's the only way to build actual knowledge about your account rather than just hoping the new configuration works better than the old one.

One thing worth saying plainly: most of what's wrong in underperforming Meta accounts is not the targeting or the creative. It's the structural and tracking infrastructure underneath both of those things. Bid strategy, attribution, pixel health, campaign consolidation, audience exclusions - these are less interesting to talk about than hooks and audiences, but they're where most of the unlocked performance actually lives.