How to Audit Your Meta Ad Account and Actually Maximise ROAS in 2026

If your Meta ads seem to be performing well, decent ROAS, reasonable click through rates, but the business results don't back that up, you're almost certainly missing something. A proper Meta ads audit cuts through the surface-level numbers and shows you where spend is leaking, where tracking is broken, and where your campaign structure is working against you rather than for you.

This guide walks through every major audit area, from how you define ROAS in the first place, right through to attribution settings and landing page performance.

Key Takeaways

• Platform-reported ROAS ignores COGS, shipping, returns and overhead, a reported 2x ROAS can still mean you're losing money.
• Poor campaign structure and fragmented data confuse Meta's algorithm, wasting budget on overlapping audiences.
• Missing Conversions API (CAPI) means Meta can't see a large portion of user signals, especially post-iOS changes.
• Creative and landing page alignment matter as much as targeting, a great ad sent to a slow, mismatched page will underperform every time.
• A structured audit gives you clarity on where the real performance levers are, so you can focus spend on net-new customer acquisition and genuine profitability.

Step 1: Get Clear on What ROAS Actually Means for Your Business

Before you look at a single campaign, get this straight: the ROAS number Meta reports is not your actual return. The platform counts revenue against ad spend, nothing else. It doesn't touch cost of goods sold (COGS), shipping, returns, or any operational overhead.

To see where you actually stand, the calculation is simple: True ROAS = Total Revenue / All Costs.

As a rough benchmark, average eCommerce ROAS sits somewhere between 2x and 4x depending on product and average order value (AOV). Subscription or SaaS models typically need 5x or above because the economics are driven by lifetime value (LTV). Lead generation depends entirely on your LTV and close rate.

Here's a real illustration of why this matters. Consider a fashion brand spending $100,000 per month on Meta ads. Facebook reports a 2x ROAS, which sounds fine. But once you factor in COGS of $15 per order, shipping at $6, and an average return rate of 20%, the true ROAS drops to roughly 0.9x. That means the business is actively losing money on every sale, while the platform dashboard shows a green number.

Know your margins before you audit anything else. Everything downstream depends on it.

Step 2: Review Your Campaign Structure

Once your ROAS targets are grounded in reality, the first major area to audit is how your campaigns are actually set up.

Ask yourself these questions:

• Are your campaigns aligned with clear objectives, conversions, lead generation, awareness?
• Are you using ABO (Ad Set Budget Optimisation) or CBO (Campaign Budget Optimisation)?
• Are campaigns being duplicated too often, resetting the learning phase each time?

Meta's own recommendation is fewer campaigns with clear, distinct objectives. CBO is generally more efficient for scaling because the algorithm distributes budget towards the best-performing ad sets automatically. To exit the learning phase properly, campaigns need to gather 50 or more conversion events per week.

A common red flag: running more than 7 ad sets per campaign, combined with frequent campaign duplication. Accounts with 13 or more overlapping campaigns can burn through budget on the same audiences, causing ad fatigue without the account manager even realising it. The data gets fragmented, and the algorithm gets confused.

A cleaner structure to aim for looks like this:

Campaign Type	Purpose
Evergreen CBO	Core scaling campaign, always-on
Offer-based campaign	Flash sales, product drops, time-limited promotions
Lead generation campaign	If applicable to your business model

Test creative within isolated ad sets inside a master CBO rather than rebuilding campaigns weekly. Every time you duplicate and restart, you're throwing away accumulated performance data.

Step 3: Validate Your Pixel and Tracking Setup

Poor tracking leads directly to poor optimisation. Meta's algorithm needs accurate signal data to find the right people, if that data is unreliable, you're essentially asking it to work blind.

Work through this checklist:

• Is the Facebook Pixel installed correctly on every conversion page?
• Are key events firing: Purchase, Add to Cart, View Content, Initiate Checkout?
• Is the Meta Conversions API (CAPI) integrated to fill the gaps browser tracking leaves?

Use Meta's Pixel Helper browser extension or the Test Events Tool inside Meta Events Manager to confirm both browser-side and server-side events are firing correctly.

Browser-only tracking has become increasingly unreliable. Research by Nano Interactive found that 70% of consumers are blocking cookies online, which means a significant proportion of conversion events go unreported post-iOS changes. If there's no CAPI in place, Meta has lost visibility into a large portion of user signals.

Implementing CAPI delivers an average improvement of 8% in event match quality, better attribution across devices and sessions, and improved custom audience generation. For most accounts, this is one of the highest-impact fixes in the entire audit.

Step 4: Audit Your Audience Targeting

Who sees your ads has at least as much influence on performance as what the ads say. A weak audience setup wastes budget regardless of how good the creative is.

Check these points:

• Are you using custom audiences built from purchasers, your email list, and website visitors?
• Have you created lookalike audiences at 1%, 3%, and up to 10% based on your highest-LTV customers?
• Are your audiences overlapping with each other, causing the same people to see multiple ads from the same account?
• Are exclusions properly configured, for example, excluding past purchasers from acquisition campaigns?

A healthy audience structure usually covers three tiers:

• Cold/Warm: Lookalikes from high-LTV customers, plus interest-based audiences for broader reach.
• Hot: Retargeting people who've added to cart, viewed content, or started but not completed checkout.
• Retention: Targeting returning customers with 30-day, 90-day, and 180-day windows.

Use Meta's breakdown feature to filter performance by region, age, and device. You'll often find certain audience segments dramatically outperforming others, and shifting budget accordingly can improve efficiency without changing a single ad.

Each ad set should have a distinct audience with minimal overlap. Overlapping audiences aren't just wasteful, they actively confuse the algorithm about which ad set to optimise.

Step 5: Review Ad Creative and Copy Performance

Creative is the primary driver of click-through rate (CTR), cost per result (CPR), and ultimately ROAS. If your creative isn't connecting, no amount of structural tidying will fix performance.

Key metrics to review:

• CTR: Aim for 1% or above on cold audiences.
• Engagement rate: A useful signal for whether the creative resonates before users even click.
• Cost per result (CPR): Is the ad actually converting people, not just attracting clicks?

On format, carousels often deliver 30 to 50% lower cost per click (CPC) compared to other formats, making them worth testing if you haven't already. Short-form video in 9:16 ratio works well for Stories and Reels placements. User-generated content (UGC) consistently performs strongly, particularly in direct-to-consumer categories.

One thing that's easy to overlook: the message in your ad must match what users find on the landing page. If your carousel promotes free shipping on orders of $49 or more, but that offer is buried or absent on the landing page, expect high bounce rates and poor conversion. The promise you make in the ad has to be the first thing users see when they arrive.

For testing, Dynamic Creative Testing within a broad CBO gives the cleanest comparative data. Once you find a winning creative, transfer its Post ID to other ad sets so it keeps its accumulated social proof (comments, likes, shares) rather than starting from zero.

Kill underperformers quickly, but only once you have statistically meaningful data. Don't make decisions on thin sample sizes.

Step 6: Audit Your Landing Pages

Meta ads don't operate in isolation. You can have a technically solid campaign and still see poor ROAS if the pages you're sending traffic to are slow, confusing, or misaligned with the ad.

Run through these checks:

• Is the page designed mobile-first? More than 90% of Meta traffic is mobile.
• Is page speed under 2.5 seconds? Use Google PageSpeed Insights to check.
• Is the call-to-action visible above the fold without scrolling?
• Does the landing page headline mirror the ad copy and offer?

Common problems that drag down conversion include slow mobile load times (which raise bounce rate and suppress conversions), poor CTA contrast or placement, and message dissonance, where the ad promises one thing and the page either doesn't mention it or contradicts it.

One practical example: a CPG brand reduced bounce rate by 23% simply by improving mobile page speed and making a Buy 1 Get 1 Free offer sticky as users scrolled. That kind of improvement ådoesn't require new ads or new audiences, just fixing the page itself.

To improve dollars per session (DPS), consider adding urgency triggers, one-click offers, embedded upsells, or dedicated landing pages per campaign or offer. You've already paid to get users to that page, don't waste that spend with sloppy UX.

Step 7: Check Your Attribution Model and Reporting

Meta's default attribution setting post-iOS is a 7-day click and 1-day view model. This is worth understanding because it affects which conversions get credited to which campaigns, and therefore how ROAS figures compare across different time windows and ad sets.

Attribution mismatches are a common source of confusion during audits, particularly when Meta's reported figures diverge significantly from what your CRM, Shopify dashboard, or GA4 shows. Always cross-reference platform data against at least one independent source. If the numbers are materially different, you need to understand why before making any scaling decisions.

Consider what attribution window actually makes sense for your buying cycle. A high-consideration product with a longer research phase might see conversions that fall outside Meta's default window entirely, making your campaigns look worse than they are.

What a Good Meta Audit Actually Delivers

A thorough Meta ads audit isn't a one time exercise, it's a reset that gives you accurate data to make better decisions going forward. The goal isn't to find a single magic fix. It's to systematically close the gaps between what the platform tells you and what's actually happening in your business.

For most accounts, the biggest wins come from a combination of fixing tracking (especially adding CAPI), simplifying campaign structure, aligning creative with landing pages, and building proper audience exclusions. None of these are technically complex, but all of them require you to look at the account with fresh eyes rather than just checking the dashboard ROAS number.

If you're running paid advertising and want to make sure your campaigns are actually building the business rather than just generating platform metrics, our paid advertising service is built around exactly this kind of structured, profit-focused approach.

Stop optimising for numbers that don't reflect real profit. A proper audit tells you what's actually working, and that's the only place to start.