
How to Audit a Meta Ads Account (Step by Step)
Audit a Meta Ads account in 8 steps: Event Match Quality, Conversions API, creative variety and budget. Meta reported +8% ads quality on selected segments.
By Abhilash LR
A Meta Ads audit works best as eight checks in a fixed order: measurement first, then creative supply, then structure and budget. Andromeda, the retrieval engine Meta described in December 2024, raises the stakes. Meta's engineers reported a +8% ads quality improvement on selected segments from it, but the engine can only work with the signal and creative you supply. This guide covers where to click, what a pass looks like, and which mistakes to watch for. If you also need the reporting basics, start with our guide to reading Ads Manager.
Key Takeaways
- Check Event Match Quality first. Weak match data feeds a noisier signal into every automated feature.
- Confirm Conversions API is sending events. Being switched on is not enough.
- Count distinct creatives. Meta reported a +8% ads quality gain on selected segments from Andromeda, and variety gives it more to retrieve.
Step 1: Check Event Match Quality in Events Manager
Event Match Quality (EMQ) is a 0-10 score in Events Manager that shows how confidently Meta can match an event to a real person using the customer parameters you send. Check it for each event type separately. A single account-wide impression hides weak spots, and a low score weakens every optimization built on that signal.
Open Events Manager, choose your data source, and review purchase, lead and add-to-cart events one at a time. A pass looks like scores that sit close together across events. A fail looks like one event trailing the rest. That event is your first fix. The usual cause is a missing parameter, such as a hashed phone number or external ID, or a Conversions API connection that is configured but not firing. The common mistake is checking only your headline event and assuming the others match.
Step 2: Is Conversions API Actually Sending Events?
"Connected" and "sending" are different states. Conversions API (CAPI) passes events from your server to Meta, which matters where in-app browsers and cookie restrictions weaken browser-only Pixel tracking. Verify recent server-side event volume in Events Manager diagnostics instead of trusting a green checkmark on the integration.
In Events Manager, open the diagnostics for your data source and look at recent server events. A pass shows server-side events arriving alongside browser events for the same actions. A fail shows a live integration with little or no server traffic. Attribution is where we most often find broken setups when we review accounts. A business that closes sales offline, through a store visit, a call or a demo, has a harder question to answer: how do you attribute those offline events to your online spend? When nobody answers it, campaigns have no clear view of real performance, and budget leaks into spend nobody can tie to a result. Step 2 shapes every decision Meta makes afterward, so check it first. The common mistake is treating the setup wizard's completion as proof of delivery. Our conversion tracking guide walks through the setup itself.
Check this step with extra care if your customers mostly browse inside social apps, because browser-only tracking has the least to work with there.
Step 3: How Many Genuinely Distinct Creatives Are Running?
Count distinct creatives, not total ads. Five ads that share one layout and differ only in headline give Andromeda's retrieval stage little to separate. Five ads built around different hooks or formats give it real choices. Compare that count with your spend level before judging whether supply is adequate.
Pull a list of active ads and group them by concept: the hook, the format and the visual approach. Each group counts once. A pass is a spread of concepts that matches your budget. A fail is a high-spend account running two or three near-identical ads. The common mistake is duplicating a winner with minor edits and calling it testing. Our creative testing guide covers how to build real variation.
Step 4: What Does Your Format Mix Look Like?
Static, video and Reels-native creative behave differently by placement. An account with no video is effectively absent from placements that favor it, however well its static ads perform elsewhere. Auditing format mix shows which placements you are paying to enter and which ones you cannot compete in.
Open your active ads and tag each by format. Then compare that mix with the placements in your spend breakdown. A pass is a mix that covers the formats your placements reward. A fail is a single format carrying everything, which can leave whole placements unserved. The common mistake is resizing one static image for every placement and treating that as format coverage. Square and vertical crops of the same picture are still one concept.
Step 5: Where Is Learning-Phase Churn Hiding?
Repeated edits can reset learning phase progress, which inflates cost per result in a way that looks like an audience or creative problem. The cause is avoidable. Checking campaign edit history shows whether your own changes, rather than the market, are driving unstable delivery.
Open the edit history for each campaign and look at ad sets where budget, audience or creative changed several times in a short window. A pass is a history of spaced, deliberate changes. A fail is a cluster of edits followed by a cost spike. The common mistake is reacting to a bad day by changing three things at once, which makes the cause impossible to isolate. If delivery tires out, our guide to ad fatigue on Meta helps separate fatigue from churn.
Step 6: Do Your Ad Sets Overlap in the Same Auction?
Audience overlap puts your own ad sets into competition in the same auction. You pay for that internal contest, and it splits delivery across ad sets that were meant to reach different people. Reviewing targeting side by side is the quickest way to find it.
List every ad set within a campaign, then across campaigns if you run several, and compare their audience definitions. A pass is ad sets that each target clearly different people. A fail is two or three ad sets built on the same interests or lookalike sources. The common mistake is splitting one audience into many ad sets by age band or placement in the hope of more control. Fewer, broader ad sets often give the system more room to work.
Step 7: Review Advantage+ Campaign Setup
Advantage+ campaigns automate delivery, but automation depends on the creative and audience signal you give it. Switching one on and leaving it untouched starves the system. Check what each campaign is being fed, because automation cannot compensate for thin inputs.
For Advantage+ Shopping or App campaigns, review the creative inside them and the conversion event they optimize toward. A pass is a campaign with varied creative, a clean event signal from steps 1 and 2, and a refresh schedule. A fail is a campaign launched months ago with the same two ads. The common mistake is blaming Advantage+ for weak results when the inputs, not the automation, set the ceiling. Our piece on scaling budgets shows how those inputs interact with spend.
Step 8: Does Budget Allocation Still Match Current Performance?
Andromeda changed retrieval-stage behavior, so placements and formats that once carried your account may no longer do so. Compare where budget sits with where results now appear. This is the final check because it relies on clean data from the earlier seven.
Pull a recent placement breakdown in Ads Manager and set it beside your budget split. A pass is spend concentrated where results are showing up. A fail is spend parked on a placement or campaign that carried the account last year. The common mistake is trusting the platform's attributed numbers without questioning them. Our guide to attribution and measurement explains why a breakdown can mislead, and it pairs well with the red flags in our PPC audit checklist.
| Step | What it catches |
|---|---|
| 1. EMQ per event | Weak match data undermining every optimization layer |
| 2. CAPI actually sending | Under-counted conversions where browser tracking is weak |
| 3. Distinct creative count | Retrieval-stage candidate-pool starvation |
| 4. Format mix | Placements your account is invisible in |
| 5. Learning-phase churn | Self-inflicted cost inflation from frequent edits |
| 6. Audience overlap | Ad sets competing against each other |
| 7. Advantage+ inputs | Automation running on too little signal |
| 8. Budget vs. current performance | Spend concentrated where results no longer show up |
What to Do With the Findings
Rank every finding by impact against effort, then fix high-impact, low-effort items first. Measurement problems from steps 1 and 2 usually go to the top, since later diagnoses depend on trustworthy data. Slow, expensive fixes such as new creative production come next. Write each finding down with an owner and a date.
Here is a simple way to score them. Give each finding an impact score and an effort score from 1 to 5, then subtract effort from impact. Work down the list from the highest result.
Illustrative example, not a client result. Suppose an audit surfaces three findings. A CAPI gap scores impact 5 and effort 2, so 5 minus 2 gives 3. Learning-phase churn scores impact 3 and effort 1, giving 2. A creative shortage scores impact 4 and effort 4, giving 0. The order is CAPI first, churn second, creative third. The creative work still matters, but it needs a plan rather than a quick fix.
Frequently Asked Questions
How often should I audit my Meta Ads account?
Audit on a regular schedule and after any major change, such as a new pixel setup or a large budget shift. A cadence you can sustain beats an ambitious one you skip. Our guide on how often to audit your ads covers timing, and the principles carry across platforms.
How is this different from auditing a Google Ads account?
The mechanics are platform-specific. Andromeda's retrieval logic has nothing to do with Google's bidding targets. The discipline is shared, though: check measurement quality first, because automated features on both platforms inherit whatever is wrong with the data underneath them.
Do I need a certain number of active ads to pass?
There is no official number, and any universal count is a guess, since Meta does not publish one. The real test is variety relative to spend. A high-spend account leaning on two or three near-identical creatives has a candidate-pool problem whatever a generic benchmark claims.
What is the single highest-leverage fix in this list?
Step 2 is the fix we would verify first. Confirming Conversions API is sending events is easy to mistake for "done" when it is not. The gap silently affects every optimization decision Meta makes afterward, so it is worth verifying before anything else.
What if my account fails several checks at once?
That is common and no reason to start over. Fix measurement first, steps 1 and 2, because everything else is easier to diagnose once the data is trustworthy. Then work through the rest using the impact-versus-effort ranking above.
Should I hire someone to run the audit?
You can run these eight steps yourself, and our guide to running Meta ads yourself shows what that involves. If you would rather bring in outside help, read the questions to ask before hiring an ad auditor first.
Get Your Account Audited
Loupe by COACT is a one-time Meta and Google Ads audit. It is a 133-point review delivered with a recorded walkthrough and prioritised fixes, and there is no retainer. Pricing is US$199, US$299 or US$399. If you want a structured review rather than self-diagnosing, see Loupe, or contact COACT with questions about which option fits your account.
Abhilash LR is the founder of COACT, a growth-focused agency operating across India, Southeast Asia and Singapore. He has spent a decade working with startups and scaleups, where he kept seeing the same problems: businesses leaving growth on the table because of the wrong partners, systems, strategy and incomplete data loops. COACT is built on the principles he works by: performance-driven, founder-obsessed, and focused on what actually moves the needle. COACT also sells Loupe, a fixed-price ad account audit. Questions or corrections: contact us. Learn more about COACT.
Published at www.coact.biz2026-09-30 · Abhilash LRWant results like these for your brand?
COACT runs performance-based growth across SEA & India. Start with a free 15-day pilot.
Book a strategy callKeep reading
Audit Findings: What to Do With Them, Step by Step
Turn audit findings into fixes in five steps, tracking first. Get an impact and effort table and learn how to judge a fix after Google's 7-day learning window.
Ad Account Audit: Do It In-House or Hire Someone?
Audit your ad account yourself or hire out? Use a 3-question test and a hybrid workflow, or a US$199 fixed-fee audit as a starting point if you need help.