A Meta account can show a decent ROAS and still be quietly strangling your growth. Maybe it is retargeting past customers and claiming credit for sales that would have happened anyway. Maybe your best ad is running out of steam while the account keeps spending on it because nobody has a real creative pipeline.
That is why knowing how to audit paid social performance matters. Not to produce another tidy report full of green arrows. To find out whether Meta is actually creating incremental, profitable revenue for your Shopify brand – and where your next dollars should go.
Most agencies audit the surface. They point to CPM, CTR and ROAS, then call it strategy. Founders need a commercial answer: what is broken, what is working, what is being falsely credited, and what needs to change this week?
Start With the Numbers That Keep the Business Alive
Before opening Ads Manager, get clear on the business economics. Meta is not an island. An ad account can only be judged against the margin, cash flow and customer behaviour of the store behind it.
Start with your target customer acquisition cost. This is not a number copied from an old spreadsheet. It should reflect gross margin after product cost, fulfilment, shipping subsidies, payment fees, returns and any other costs that eat into a first-order sale. Then consider how much cash you can sensibly put at risk before a customer’s repeat purchase makes the acquisition profitable.
A brand with strong repeat purchase rates can deliberately accept a lower first-purchase ROAS than a one-off product business. That does not mean you should blindly use lifetime value to excuse bad advertising. Forecasting future customer value is useful when it is grounded in actual cohort data. It is rubbish when it becomes permission to lose money indefinitely.
Review these numbers together: blended MER, new-customer revenue, customer acquisition cost, contribution margin and cash conversion. If your blended efficiency is getting worse while platform ROAS looks fine, there is a measurement or allocation problem worth investigating.
How to Audit Paid Social Performance From the Ground Up
A proper audit works from tracking through to creative and commercial outcomes. If you start by switching campaigns off because a dashboard looks ugly, you can destroy useful learning and make the account harder to read.
Check whether the data is believable
Meta reporting is directional, not gospel. Compare reported purchases and revenue against Shopify for the same period, using consistent attribution windows. Differences are normal. Wild differences are not.
Check that the pixel and Conversions API are firing purchase events correctly, deduplicating events and passing meaningful parameters such as value, currency and content IDs. Confirm that purchase values match Shopify order values closely enough to trust revenue optimisation. Also inspect whether test orders, subscription rebills, cancelled orders or unusual discounting are contaminating the picture.
Then look at attribution. A seven-day click, one-day view setting can make Meta look more effective than it is, particularly if the account is heavy on retargeting. View-through attribution is not fake, but it needs context. If a customer saw an ad after already searching for your brand, Meta did not necessarily create the demand.
Use blended results and, where spend justifies it, controlled tests to pressure-test the platform’s claims. A simple holdout or geographic test can reveal whether increased Meta spend produces additional revenue or just claims more of it.
Separate prospecting from demand capture
The fastest way to flatter an account is to load it with warm audiences. Website visitors, Instagram engagers, email lists and past customers often convert well because they already know you. That does not mean your prospecting is healthy.
Break performance out by audience temperature. Ask how much spend is reaching genuinely new people, how efficiently those people become first-time buyers, and whether retargeting is proportionate to the traffic being generated. There is no universal budget split. A brand with high traffic from organic, email and creators can support more retargeting. A brand relying on Meta to create demand needs stronger cold acquisition.
Pay close attention to exclusions. Existing customers should usually be excluded from acquisition campaigns unless there is an intentional replenishment or cross-sell strategy. Likewise, broad campaigns can overlap with retargeting pools if exclusions are sloppy. Meta may handle some auction overlap, but relying on that is not an audience strategy.
Inspect campaign architecture for unnecessary complexity
Many underperforming accounts are not short on campaigns. They are drowning in them.
An audit should identify whether budget is fragmented across too many ad sets, tiny lookalikes, duplicated interests and endless tests that never receive enough spend to produce a signal. Complexity feels like control. Often it is just a way to hide indecision.
For most established Shopify brands, the account needs a clear structure that separates acquisition, retention where relevant, and deliberate testing. Broad targeting frequently deserves a meaningful role because Meta has more room to find buyers. Interest targeting can still work when it reflects a real customer insight or gives you a useful test, but stacking random interests is not sophistication.
Look at budget allocation over time, not just yesterday. Are the campaigns that bring in new customers receiving enough spend? Are weak ad sets being kept alive because someone is emotionally attached to them? Is the account making big daily budget changes that constantly reset stability? The right answer depends on spend level and conversion volume, but the principle is fixed: give winning ideas room to work and cut obvious waste quickly.
Creative Is Usually the Constraint
When CPM rises and ROAS falls, agencies love blaming the algorithm. Sometimes auction costs are genuinely higher. More often, the ads have stopped earning attention.
Audit creative at the ad level. Do not judge it solely by ROAS. Look at spend, thumb-stop behaviour, click-through rate, cost per landing page view, conversion rate and frequency in context. An ad with a strong click-through rate but poor conversion may be making a promise the product page cannot keep. An ad with mediocre clicks but excellent conversion can be reaching a smaller, highly qualified audience.
Review the first three seconds of video, the lead image, the offer, the proof and the objection handling. Most ecommerce ads fail because they look like ads from every other Shopify brand: polished product shots, vague lifestyle claims and a discount slapped on at the end.
The useful questions are harder. What problem does the customer recognise immediately? Why should they believe your product is better? What friction stops them buying? What evidence removes that friction? Customer reviews, demonstrations, founder stories, comparisons, before-and-afters and real usage scenarios can all work when they are built around a specific buying objection.
Creative fatigue is not simply high frequency. It is declining efficiency after an ad has had meaningful spend. A good audit checks whether new concepts are entering the account regularly or whether the brand is recycling the same angle in a different colourway. New hooks matter more than minor edits.
Follow the Click to the Checkout
Paid social cannot fix a store that leaks buyers after the click. If Meta is generating qualified traffic but the landing page converts poorly, the ad account will be blamed for a site or offer problem.
Compare conversion rate by landing page, device, new versus returning visitor, and traffic source. Check mobile speed, product-page clarity, shipping information, returns policy, reviews, size or compatibility guidance, payment options and checkout friction. Australian shoppers do not need a novel. They need confidence before they hand over their card details.
Also audit the offer itself. If competitors have stronger proof, clearer bundles, faster delivery or a more compelling entry point, better targeting will only delay the problem. Sometimes the correct recommendation is not another campaign. It is a better product page, bundle or offer.
Turn Findings Into Decisions, Not a 40-Page Deck
An audit is only valuable if it produces a short, prioritised action plan. Rank issues by likely revenue impact, confidence and effort. Fix broken tracking before debating creative. Remove obvious budget waste before launching another dozen tests. Build a creative testing plan before expecting targeting tweaks to rescue performance.
Your plan should state who owns each action, what success looks like and when the decision will be reviewed. For example, a new acquisition angle might need a defined spend threshold before it is judged. A landing-page change should be measured against a meaningful comparison period, not two days of noisy data.
The goal is not to make the account look clever. It is to create a system that finds new customers at a cost the business can afford, keeps learning from creative, and tells the truth when performance changes.
If your agency cannot explain where the revenue is coming from, what is suppressing it and what they will do next, you do not have a performance partner. You have someone managing a dashboard. The useful next move is simple: audit the account against your actual economics, then make the hard changes before another month of spend disappears into reporting theatre.