Auditing Meta Ads

Sam Thompsett
May 23, 2024

Paid social account audits - some people love them, some people hate them. As somebody who audits at least one new account every week, it's a big part of my day-to-day.. Here at Genie Goals, conducting regular audits ensures that our accounts are shipshape, and I’m also consistently involved in assessing the accounts of prospective new clients. It’s an essential but ultimately time-consuming task, which is why it’s important to have a robust auditing framework - to uncover everything you need to in the shortest amount of time possible.

Whether you’re looking for an auditing framework or just want to get a better grasp of your own paid social account performance, look no further. Below are the top 10 things I look for when auditing an account.

#1 Does your account structure make sense?

Account structures of the past mirrored the marketing funnel - awareness, consideration, conversion and loyalty. That has changed with the rise and now, dominance of Advantage+ Shopping Campaigns, but it’s important to remember the fundamentals.

Reach new audiences in awareness and show them engaging content that showcases your brand USPs, then nurture them down the funnel with more product-focused ads to drive that first sale.

Often I audit accounts that have no structure at all, with multiple campaigns all trying to achieve the same thing.

Even if Advantage+ has changed the game, the differing levels of exposure to your brand remain relevant to your creative & account strategy. Produce creative relevant for every user type and use Meta’s Advantage+ solutions to target them.

#2 Is your account over-segmented?

More stringent privacy laws and the rise of AI over the past few years have driven both paid social and PPC in the same direction - towards broader targeting. Yet, it’s surprising how many accounts I audit that are still heavily segmented. The most common things I see?

  • Campaigns: Every campaign should have a clear objective. Don’t use two campaigns to target the same audience with slight differences, e.g. similar creative… just use the creative that works best.
  • Ad Sets: Audience targeting has less of a bearing on performance than ever before. In an awareness campaign only use a max of 3 ad sets. Broad and Advantage+ audiences are essential.
  • Placements: Don’t split ad sets by placement type. Test placements for a while if you like, but the best approach is to unify targeting and let Meta push the placements that work.

#3 Review your optimisation types

One of the first places I look in an account is the Ad Account Overview. Here you can spot how much budget an account is spending in the learning phase. More time spent in the learning phase = a lack of data = an inefficient account = less efficient cost per result.

A campaign/ad set needs 50 results to get out of the learning phase and this resets after every significant edit. The inability to reach that threshold is often down to the optimisation types used and a scarcity of data.

If you don’t have the volumes of data to use a sales optimisation, my best practice is always to use the next best thing, e.g. if you can’t use sales, use add to carts. If you can’t use that, use content views, then landing page views, and finally link clicks. 

#4 Pay Attention To Your Demographic Targeting

Are you spending your hard-earned budget where you expect? By that I mean, does your spend by demographic (age, gender) mirror your understanding of the brand and/or your top-performing demographics in Google Analytics?

The biggest pitfall I see is that if you optimise towards link clicks and leave the 65+  in your targeting, Meta loves to spend your money on that audience and it’s always spent just on Facebook, rather than Instagram.

Why? Because the CTRs are so high and the CPCs are low… older demographics aren’t as digital-savvy and are maybe less cynical (than me), so they’re more likely to click on an ad.

#5 High frequencies, wasted budget

Frequency is a measure of how many times a user has seen your ad. It's a cumulative metric, meaning if I look back over the last 30 days, it’s a measure of how many times a user has seen that ad over that period.

Often I see accounts where this metric is way too high. A user doesn’t need to see your ad 45 times to make a sale and that’s a waste of budget. So, what does good look like?

  • Upper funnel: Aim for 2 to 3 over the last 30 days.
  • Lower funnel: Try not to go over 10 if you can avoid it.
  • Advantage+ Shopping: In between, e.g. 4.

I’m not totally against high frequencies for all accounts and strategies, e.g. if you’re launching a new brand, I think a high frequency can be advantageous in getting your brand on the map. But, only if you’re showing a user a broad spectrum of USPs and creatives.

#6 Creative

Creative is subjective and everyone has an opinion, but objectively there are some things which never work

  • Lack of personalisation to users at different stages of the customer journey. People need to see different things about your brand based on their knowledge of it.
  • A lack of creative variety and ad formats, e.g. over-reliance on DPAs. Broad DPAs can work, but we want to build a story around your brand, not just show a picture of a shoe.
  • Not utilising primary texts, headlines and descriptions fully and not showcasing all of your brand’s USPs.

In general, I want to see that your paid social creative encompasses all aspects of your brand’s mission and strategy. 

#7 A lack of A/B tests

Everyone has an opinion of whether you should run A/B split tests or just allow Meta to optimise to what’s working best. My problem is when no A/B tests have ever been run in an account at all.

The history of the Experiments tab is something I’ll always check in an audit.

Some things need to be split-tested side by side for the test to be fair, such as optimisation types.

I lean to A/B testing most of the time, so that you can always produce reliable results. Test optimisations, ad formats, audiences test messaging variations.

#8 Broken Pixels

A bit technical and all-in-all very time-consuming, but auditing your pixel setup is imperative.

Check pixel helper and Events Manager when onboarding new clients, as well as at regular intervals for existing accounts to ensure nothing has gone amiss.

One of the most common, but niche mistakes I see relates to dynamic remarketing, where Meta is unable to match the ID that someone has viewed on the feed to the ID of the product within the feed. In this case, the products in ads that a user sees are unlikely to be the products they viewed on-site, so please check your match rate.

#9 Not applying exclusions.

New users and existing customers need to be treated differently. A common issue in accounts is a lack of a past purchaser exclusion applied to any campaigns that target new customers.

The messaging that you serve to new users, who do not know the brand, should be different from existing customers. If you’re looking to focus on new customer acquisition, you don’t want to spend your budget on existing customers who could convert via other channels anyway.

There are absolutely caveats to this, the number point one being if you don’t have past purchaser targeting set up in your account, applying exclusions will mean a valuable audience isn’t targeted via paid social.

My question to you would be why are you not targeting existing customers in their campaign? Paid social’s strength is new customer acquisition, but you can also utilise it to create brand advocates and drive lifetime value from existing customers.

Remember to do this for Advantage+ Shopping Campaigns too, although that feature is likely to go away soon.

#10 Missing or broken UTMs.

Your analysis of your account shouldn’t rely solely on Meta’s reporting suite. I want to know how qualified our traffic is once they’ve made it through to the website, so tag your ads with UTMs. 

One of my favourite Google Analytics metrics is average session duration, a measure of how long someone spends on the website and I look at this for every audience.

It’s all well and good if you have click-through rates of 5%, but if they bounce as soon as they reach the site, then where is the value?

There are many more things you could check when reviewing your accounts and I could keep going… but these are the things that regularly give me the biggest headaches. My recommendation, build these out into a simple Google Sheet and review your accounts at regular intervals to stop pesky issues hindering your performance.

Back to Resources Hub