When it’s time to audit a new client’s ad account, the history to consider consists of everything from the account setup to the results. In between, one key element is what they’ve done on the ad side.
This is part of a series of posts I’ve written about account audits. If you want to take this seriously, there’s much more to consider than the performance alone.
Here’s what’s been covered so far:
As has been the case in prior posts of this series, the audit isn’t all about marking whether something is “good” or “bad,” “right” or “wrong.” While there are certainly some strategies that classify as red flags, this is more about documenting the habits and general approach that’s been taken so far.
Maybe whatever they’ve done is unconventional, but it gets results. By documenting everything, you will have a list of areas to consider if performance isn’t ideal.
And now let’s take a closer look at the ads…
Ad Volume
When it comes to ad volume, there are two primary considerations:
1. How many ads are active at one time per ad set?
Generally, I’ll see anywhere from one to six active ads per ad set. Meta says that there’s no benefit of including more than six (this would exclude Advantage+ Shopping Campaigns, but those are getting replaced by Advantage+ Sales, which will follow the standard rules).
“Normal” would be anywhere from two to six ads per ad set, depending on the budget assigned. Only one ad would be risky, as it gives the algorithm a single choice to work with. More than six might be overkill, but budget and results would matter.
2. How many ads are active at one time overall?
The number of total ads isn’t necessarily a huge deal, but it does help get a sense of the approach. For example, you could have a situation where an account has 10 ad sets with 10 total ads (one ad per ad set). On the other hand, they may have one ad set with 10 ads.
Both would be abnormal and something to note. But how and why they create the ad sets they do will also matter (addressed in the post about ad sets).
Ad Rotation and Manipulation
Here are a few questions to consider…
1. How often do they publish new ads?
Are they on a schedule of publishing a new ad every two weeks, or do they let performance be their guide?
Some advertisers will certainly over manage this, forcing new creative and killing what was otherwise working. The size of the audience and budget will certainly matter, but targeting a country the size of the United States should be enough to sustain larger budgets without requiring a constant rotation. Creative fatigue is rarely the risk we think it is, especially if we embrace broader targeting.
2. For how long does an average ad remain active?
This can depend on the promotional calendar since some offers will be seasonal. Short promotions are certainly more challenging, especially since it requires a constant rotation of new creative and can be difficult to maintain a high performance.
If no such promotional calendar exists, I would like to see history of effective ads allowed to run for weeks and even months at a time until they are no longer useful.
3. Do they manually stop ads or let the algorithm sort it out?
There was a time when manually stopping ads made a lot more sense. But these days, it’s usually a way to stay busy and show that you’re doing something, rather than doing anything particularly impactful.
Advertisers will often look at five ads and take the results at face value. They’ll see what appears to be Meta’s random distribution based on results. But a whole lot goes into the results that you see. Different ads are more likely to get different people to act and from different placements. “Bad” overall results won’t mean that if you stop the ad you’ll get better results.
Meta describes this as The Breakdown Effect, and shares an example related to distribution by placement. But a similar argument can be made for how the algorithm distributes different ads.
I generally recommend a hands-off approach unless the ad set itself isn’t performing. It may then make sense to pause the ad that is getting the most budget to see what would happen if other ads were featured more.
Approach to New Ads
Should you create a separate campaign or ad set just for testing, or should you simply add new ads to an active ad set?
This tends to be a controversial topic in advertising circles, and there are various factors to consider regarding what makes sense. Let’s consider…
1. Creating a separate testing campaign or ad set.
This could theoretically make sense for high budget clients who are planning ahead. Essentially, you have the main active ad set running your primary ads. You create a separate campaign or ad set strictly for testing purposes so that you can find which versions to use in the future to limit changes later.
I would not do this for mid to low-budget clients, however. You’re stealing budget from the main ads just for testing purposes. You can also create Auction Overlap by splitting them up this way, and that won’t help performance.
I also would be sure that if this is happening that they’re optimizing for the action they want. There’s a questionable “guru” strategy of optimizing for link clicks or landing page views to find the versions that get the best CTR, then moving the winners into the active ad set. That’s nonsense. A high CTR when optimizing for clicks will not guarantee conversion success.
2. Introducing new ads as needed.
The vast majority of advertisers can take a far more simplistic approach to introducing new ads. Compared to the separate ad set approach, it sounds too easy. But the benefits of separating them when we’re dealing with lower budgets and smaller sample sizes simply do not outweigh any potential risks.
If an ad set is working with this approach, there’s no reason to introduce new ads. If performance is not what you want, introduce new ads and turn off those that have run the course. This will reset learning. Monitor performance to determine if that addition was productive. If not, rinse and repeat.
Make note of whether they’ve historically taken approach 1 or 2. If the results are there, you can potentially ignore it. If not, this is on your list of potential changes to make.
Manual vs. Advantage+ Catalog Ads
Advantage+ Catalog Ads will only apply to e-commerce brands who have a catalog of products that they want to promote. Everyone else will rely on manually created ads.
If this client has used Advantage+ Catalog Ads, what has been their approach?
1. Only Advantage+ Catalog Ads.
In theory, there’s nothing wrong with this approach, assuming they’re getting good results. Smaller budgets will also have fewer options. If you have a $100 daily budget, I’d rather see you put it all into one ad set with Advantage+ Catalog Ads than cut it in half by also running manual ads.
Simply make note of it.
2. Mixture of Advantage+ Catalog Ads and manual ads.
This often provides some benefits, since you can lean partially on automation and partially highlight specific promotions. I’d favor this for higher budgets, as explained above.
For either case, make sure that the catalog is set up properly: No errors, proper sized creative, and accurate and complete text.
Ad Format
When this client creates manual ads, which formats do they tend to use?
You won’t be surprised to read me say there isn’t a right or wrong here. But, it certainly helps to provide some variety.
First, it’s a matter of utilizing all placements available. Some placements only utilize static images while some will only display videos.
Second, there are benefits and drawbacks to each format. Link ads with a static image tend to do a better job of attracting a click whereas videos generally do better at grabbing attention and educating. This isn’t universal, of course, but it’s generally the case.
Some advertisers have great success with carousels, though that is typically for e-commerce brands. I’ve found that others struggle to get the same results from carousels as from single image and video.
Flexible Ad Format is only available for certain objectives (you can replace this with Dynamic Creative for the other objectives), and I’ve actually found it to be quite useful. If you demand control and the ability to isolate “winning” creative combinations, you’ll hate it. But I’ve otherwise seen that Flexible tends to outperform single image or video.
The main thing is to make note of what is used, and hopefully there’s some variation. If performance suffers, this is an area to potentially address.
Copy Evaluation
There is plenty about ad copy that is subjective, but there are a few things to pay especially close attention to…
1. Do they provide multiple text variations?
You can provide up to five primary text, headlines, and description options. I strongly recommend that you provide at least three primary text variations (so does Meta), and I would consider that as well for at least the headline. The description is less important since it doesn’t always appear.
Is longer or shorter text better? Which benefit should you highlight? By providing text variations, you can try different things within a single ad.
This is one of the many ways that you can avoid creative fatigue, regardless of the number of individual ads that you create. It also allows you to try different text without needing to create new ads. One type of person may respond better to one variation, and another person to another variation.
You can view results by text variation with Breakdowns, but I wouldn’t obsess over this information. You shouldn’t use text variations for split testing purposes. This is about giving the algorithm options and preventing creative fatigue.
2. How is the text formatted?
Do they use all caps? A bunch of emojis? Long paragraphs or does each sentence get its own paragraph? These things can matter.
3. Are proper spelling and grammar used?
Sure, typos can attract engagement from people who have the need to correct you. But is that the type of engagement that you want?
The type of brand and expectations matter, of course.
4. Is the voice appropriate for the brand and audience?
Is the voice casual or professional? Would it be more appealing to Gen Z or retirees? The way your ad communicates needs to be consistent with the brand and your intended audience.
5. Is there a clear call-to-action?
Is it clear what you want your audience to do? Is it engaging? Does the copy invite action at the beginning and end?
Don’t assume that they will know what you expect them to do or why they should do it.
6. Do they use AI-generated text?
Meta provides the option of utilizing AI-generated text suggestions. I guess, whether or not they’re used shouldn’t matter on its face, though I’ve found them to be pretty bad. Results matter. If results aren’t there, this becomes more important.
Creative Evaluation
We can get into a whole philosophical discussion about creative quality here, and it’s important to remember that results are most important. Polished ads aren’t necessarily better than UGC ads. They may not even be better than the simplest and ugliest ads. What works is not always predictable.
With that in mind…
1. Does the creative provide a clear message?
Ultimately, this is all that matters. Does the creative quickly grab our attention and get a clear message across? Everything else is secondary.
2. Are the size and aspect ratio appropriate?
The first one can be impacted by this one. Your creative will be cropped and resized for different aspect ratios (more on that in the customizations by placement section). That can impact whether you’re able to consistently get the message across.
Important parts of the creative can be cropped out or covered by text.
3. What creative approach is taken (polished, UGC, ugly, human, graphical)?
Ideally, it’s a variation to help find out what works best and to reach different people with different strategies. But there is no right or wrong here. Consider it when evaluating results.
Advantage+ Creative Optimizations
The number of Advantage+ Creative optimizations is constantly growing, and their use is mostly controversial among advertisers — some optimizations more than others. The biggest fear is a client seeing an optimization that wasn’t approved that goes against brand standards. Because of this, advertisers often take a conservative approach.
That said, these enhancements should only be used to your benefit. Meta’s not trying to hurt your ad results by forcing you to use Advantage+ Creative. The entire point is to apply these enhancements at times and in situations that will actually lead to more results.
I understand why this hasn’t been embraced. There isn’t nearly enough transparency behind it. There’s no way to run a breakdown by Advantage+ Creative enhancements to see how ads performed when certain optimizations were applied and when they weren’t. Taking Meta’s word for it isn’t an easy thing to do.
So, let’s consider…
1. Do they automatically accept all enhancements?
I know that I’ll get pushback on this, but if I had a preference, it would be this one. In general, I’ve found that we’re too slow to adjust to automations that help us. While I understand the slow acceptance of some of these optimizations, they may be able to help you — even if it’s by a small amount.
Of course, you can’t just blindly accept them either. There are customizations that you should make to site links, catalog items, overlays, and music. When customized appropriately, all can be useful.
2. Do they automatically turn off all enhancements?
Shutting the door on Advantage+ Creative entirely is likely a bad idea. That said, it might not be worse than blindly accepting them without making sure they look and work appropriately.
3. Do they automatically turn off specific enhancements?
The Catalog Items enhancement isn’t going to apply to everyone, so it makes perfect sense to turn it off in some cases. I also understand turning off the more controversial enhancements, particularly if you aren’t going to otherwise customize them.
Customizations by Placement
Far too often, advertisers are satisfied once they provide the creative variations that Meta requests.
But these aren’t always ideal. Consider the following…
1. Aspect ratio by placement.
Use the Advanced Preview to get a more accurate view of how your ad will look by placement.
You can even view the Advantage+ Creative variations. I’ve found that customizations are often needed. See this post that outlines ideal aspect ratio by placement and format.
2. Format by placement.
As mentioned earlier, static images are ideal for some placements while videos are expected in others. While Meta will convert static images to videos to help, this is where you should consider providing a mixture of images and videos where possible to make the most of each placement.
3. Text by placement.
Different placements have different text limitations and feature text differently. If you’re going to go all out on customizing by placement, you’ll approach each placement differently.
All of these things aren’t necessary, of course. But the higher the budget, the more I’d recommend it. These small things can help performance, even if it’s small.
Website Destination Evaluation
This might be the most overlooked asset in an audit. You can do everything perfectly as an advertiser but still struggle to get results because of the website.
Not all advertising uses a website destination, but it’s ideal if you can eventually send people there. When evaluating the website destinations used in ads, consider the following…
1. Performance.
Does the site load slowly or quickly? This is critical as a slow-loading website will lead to abandonment before the transaction happens.
How seamless is the transaction process? Is it confusing? Is it broken? Or is it very quick and easy to complete?
All of these things impact ad performance, even if they are outside of your control.
2. Ownership.
There are always exceptions, but I’d recommend that the brand own the website or landing page you are using as an advertising destination. What I mean by that is that it should be on your domain, not a third-party domain (like LeadPages, for example).
This is a matter of trust when it comes to selling a product or collecting information. If you don’t even have your own website, how trustworthy can you be?
3. Action optimization.
The landing page should be optimized for the action that you want. It could look nice, but are benefits and calls-to-action clear? Have landing pages been split tested to find a top performer?
4. Branding and polish.
Your company’s domain, logo, brand colors, fonts, and overall polish are important, too. You don’t want this page looking like it’s a template that a scammer set up in five minutes. Trust, trust, trust.
5. Messaging consistency.
Is the messaging on this page consistent with what was in the ad? Or will the potential customer feel like they’re on the wrong page or have been tricked?
Messaging consistency is not only about the words themselves, but the tone, branding, imagery, and expectations.
Summary
When auditing a client’s ad account, there is a long list of items to consider related to the ads themselves without looking at performance. The evaluation of ad copy, creative, format, website, and more will help paint a picture. If performance is a challenge, you have a list of items to refer back to for possible improvement.
Your Turn
Is there anything else that you would add to an audit related to the ads themselves?
Let me know in the comments below!
