How to Audit a Meta Ads Account Before Hiring

Most business owners meet a Meta ads agency before they have ever opened Business Settings. The pitch happens, the proposal arrives, and the decision gets made on the strength of the slides. Then the account gets handed over, and three months later nobody can say whether the new team improved anything — because nobody wrote down where things started.

An audit fixes that, and it is not a technical exercise. You are not trying to become a media buyer in an afternoon. You are trying to answer one commercial question: is the money underperforming because the account is broken, or because the offer is? Those two answers lead to completely different hires.

The context makes this worth the effort. The Department of Statistics Malaysia recorded 95.3 per cent of establishments with internet access in 2024, up from 94.0 per cent the year before. Almost every competitor you have is now reachable in the same feed, bidding against you. An account with sloppy foundations does not fail loudly — it just quietly pays more than it needs to, month after month.

This guide walks the audit in the order a consultant would run it, then turns the findings into a hiring decision. The video below covers the mechanics inside Ads Manager first.

PART 1 · DIAGNOSE

What a Meta Ads Audit Is Actually For

IN BRIEFThe audit exists to give you a baseline and a shortlist question. It tells you what your account is currently worth, which problems are cheap to fix, and which need a specialist. Bring those findings to every Meta Ads management conversation and the pitches change immediately.

There are two versions of a Meta ads audit and they are not the same product. An agency’s audit is a sales document — it finds problems it can fix and prices the fix. Your audit is a purchasing document. Its job is to make you a harder customer to bluff.

The difference shows in what each one measures. An agency audit tends to catalogue settings. A buyer’s audit answers three questions:

• What did last month actually cost per result? Not impressions, not reach — the number of enquiries, bookings or sales, and the ringgit divided by that number.
• Was that number measured honestly? If the tracking is broken, every other figure in the account is a guess dressed up as data.
• What is the cheapest thing that would move it? Often the answer is not a new agency. Sometimes it is one conversion event set correctly.

This is also the moment to be honest about what you were buying before. If the account has only ever run boosted posts, you do not have an underperforming ad account — you have no ad account structure at all, and the audit will be short. That distinction is covered in boosting posts versus running proper Meta ads, and it changes what a fair proposal looks like.

PART 2 · DIAGNOSE

Who Actually Owns Your Meta Ad Account?

IN BRIEFCheck ownership first, because everything else is worthless if you cannot take the assets with you. Your own business portfolio should own the ad account, the Page and the pixel dataset, with agencies attached as partners. The Ads Manager walkthrough covers where these settings live.

This is the single finding that has cost Malaysian SMEs the most money over the years, and it never appears in a performance report. An agency creates the ad account inside its own business portfolio, runs the campaigns, and everything works fine — until the relationship ends. Then the ad history, the custom audiences and the pixel data stay behind, because they were never yours.

How to check who really owns your Meta ad account

Five minutes in Business Settings settles it. Work through these in order and write down what you find.

1. Open your business portfolio settings. In Meta Business Suite, go to Business settings. If you have never seen this screen and your agency has, that is already an answer.
2. Check the ad account owner. Under Accounts, open Ad accounts. The account should sit inside your business, not appear as one shared with you.
3. Check the Page and the dataset. Do the same for Pages and for Data sources, where the pixel dataset lives. Ownership can differ asset by asset.
4. Review the Partners list. Meta documents how partner access to business assets is granted, and every agency, freelancer and ex-staff member with access should be listed there.
5. Confirm your own permission level. Meta’s guide to ad account permissions explains the difference between admin and advertiser access. You want admin on your own account.

If ownership sits with someone else, fix that before you shortlist anyone. Asking for a transfer while the current agency still wants the account is a very different conversation from asking after you have given notice.

BENCHMARK BRIEFING 1 OF 4

What Does a Healthy Meta Ads Account Look Like?

IN BRIEFHealthy means five things line up: you own the assets, events are matching well, server-side events are running, the campaign optimises for a real business outcome, and the attribution window is stated. The table below turns each into something you can check yourself before any decision about running Meta ads in-house or hiring out.

PART 3 · DIAGNOSE

Is Your Tracking Telling You the Truth?

IN BRIEFCheck tracking before performance, because bad signals make good campaigns look bad. Open Events Manager and read the event match quality rating on your main conversion event, then confirm whether server-side events are running. The Meta Pixel and Conversions API setup guide covers the repair.

Meta scores how well the customer information your site sends can be matched to a real account. Its documentation describes event match quality as a rating from Poor through OK, Good and Great, and better matching means more conversions get attributed and the delivery system learns faster. A Poor rating on your purchase event is not a reporting inconvenience. It is the algorithm training on a blurred picture of who buys from you.

Two follow-up checks matter as much:

• Server-side events. Browser-only tracking loses events to ad blockers, privacy settings and browser restrictions. Meta’s Conversions API documentation exists to close that gap, with deduplication so the same event is not counted twice.
• Where conversions are counted. Enquiries that land in WhatsApp or an instant form are measured differently from ones that land on your website. If most of your leads arrive by chat, the comparison in WhatsApp ads versus lead forms explains what you gain and lose either way.

Write the findings down as plain sentences: purchase event rated Poor, no server-side events, leads counted in three places. That sentence is worth more in a hiring meeting than any dashboard screenshot.

BENCHMARK BRIEFING 2 OF 4

Which Problems Show Up at Which Spend Level?

IN BRIEFAudit findings are not random — they cluster by budget. Small accounts fail on foundations; larger ones fail on structure and fatigue. The illustrative model below shows which faults to expect at your spend level so you can tell a real diagnosis from a generic one, alongside the metrics that actually matter.

PART 4 · DESIGN

Reading the Account Without Being a Media Buyer

IN BRIEFYou do not need to judge bid strategy. You need to see whether the money went to one clear job, whether anything was retargeted, and whether the account has been left alone long enough to learn. That is also the frame for how to split a Meta ads budget.

Set the reporting date range to the last 90 days and look at four things only.

1. Where the money went. Sort campaigns by amount spent. If the top campaign is an engagement or traffic campaign, the account was buying attention, not customers.
2. How many ad sets are live. A small budget spread across eight ad sets gives none of them enough data to optimise. Fewer, better-funded ad sets almost always beat many thin ones.
3. Whether warm audiences exist. Look for a campaign or ad set aimed at website visitors, video viewers or past enquirers. If there is none, the account has been paying full price for every conversation, twice.
4. How often things were changed. Daily budget edits and constant restarts keep resetting the learning process. A calm account outperforms a fiddled one at the same spend.

None of these require expertise. They require ten minutes and a willingness to write down an uncomfortable answer.

BENCHMARK BRIEFING 3 OF 4

What Is Each Fix Actually Worth Per Month?

IN BRIEFRanking fixes by recoverable ringgit stops the audit becoming a wish list. On an illustrative RM 5,000 monthly budget, correcting the optimisation event and closing tracking gaps recover the most, which is why a good Meta ads agency shortlist starts there rather than with creative.

PART 5 · DESIGN

Account Problem, or Offer Problem?

IN BRIEFIf cost per click is normal but cost per enquiry is terrible, the ads worked and the offer or the landing experience did not. That is not a media-buying problem and hiring a better buyer will not fix it. Ad creative that stops the scroll only helps once the offer stands up.

This is the finding most audits skip, because it points away from the thing being sold. Walk the path yourself on a phone: tap the ad, wait for the page, find the price, and try to make the enquiry. Time it. If that journey takes more than a few seconds longer than you expect, the money is leaking after the click, not before it.

PART 6 · DESIGN

What the Findings Say About Who to Hire

IN BRIEFMatch the hire to the findings, not to the budget. Foundation faults suit a one-off specialist; structural and creative faults need ongoing management. The trade-offs are set out in Meta ads agency versus freelancer.

Whichever route you pick, use the findings as the brief. Ask each candidate what they would do about your two worst findings, in what order, and what evidence would tell you it worked by week six.

BENCHMARK BRIEFING 4 OF 4

How Long Does Recovery Take After an Audit?

IN BRIEFExpect little movement in month one and most of the gain by month three. Signal quality has to rebuild before delivery improves, which is why any promise of an immediate turnaround deserves a follow-up question about how the warm audience gets rebuilt.

PART 7 · DEPLOY

Turning the Audit Into a Hiring Brief

IN BRIEFTurn the findings into one page: current cost per result, the five checks with a pass or fail, your two priority fixes, and the number you want moved by when. Send the same page to everyone you shortlist. The same discipline works when shortlisting an SEO agency in Malaysia.

A shared brief does something a request for proposals never does: it makes the answers comparable. When every candidate is responding to the same findings, differences in their replies are differences in thinking, not differences in what they assumed.

Keep two of those findings out of the brief and ask about them in the meeting instead. A candidate who spots the ownership problem or the fatigued creative without being told has actually opened the account.

THE VERDICT

So Should You Audit Before You Hire?

Yes — and the version that matters takes an afternoon, not a fortnight. Ownership, tracking, optimisation event, structure, and the walk from ad to enquiry. Five checks, written down as plain sentences, with a date beside them.

The value is not in the findings themselves. It is that you stop buying marketing on trust and start buying it against a baseline. Agencies that are good at this welcome the scrutiny, because a documented starting point is also the only fair way for them to prove they improved something. At IZI Digital Marketing that diagnosis is where every engagement starts, before anyone talks about budget.

If the audit shows a broken account, hire for repair. If it shows a healthy account and a weak offer, fix the offer first and keep your money. Either way, you now know which conversation you are having.

FAQ

Frequently Asked Questions

1. How long does it take to audit a Meta Ads account?

Two to three hours for the buyer’s version described here. It depends on how much history the account has — an account with three years of campaigns takes longer to read than one with three months. A full technical audit by a specialist typically runs one to two weeks.

2. Can I audit the account if the agency owns it?

Only partly, and that itself is a finding. Ask for admin access in writing before you start; a reasonable partner grants it. If access is refused or delayed, treat the ownership problem as your first priority rather than as a paperwork detail to sort out later.

3. What is the single most important thing to check?

Whether the campaigns optimise for a real business outcome. It depends on your funnel — a chat-led business and a checkout-led business count outcomes differently. But an account optimising for page views or engagement is buying the wrong thing, no matter how good the reporting looks.

4. Should I pay an agency to audit the account before hiring them?

Usually no, because a free audit is a sales document and a paid one from a candidate still is. Run the buyer’s checks yourself first. If you want an independent technical review afterwards, commission it from someone who is not bidding for the retainer.

5. What if the audit says the account is fine?

Then the problem sits in the offer, the pricing or the follow-up, and hiring a new media buyer will not move it. Check how quickly enquiries get answered and whether the landing page states a price. Fixing response time is usually cheaper than any change to the ads.