I want to share some Facebook ads mistakes that I keep seeing in ad accounts audits that are just scorching money yet are super simple to fix.

No System for Testing

I see far too many businesses simply throwing shit at the wall and hoping it sticks. This method quickly turns Facebook ads into gambling at the casino. The house always wins. Systematizing your testing helps you find winning ads much faster and more reliably.

Only Change 1 Variable At A Time

Let’s say you have 1 campaign that’s doing OK. You launch a new campaign with a different audience, new creative and new copy. The new campaign does amazing. What made the impact? It’s very hard to tell, therefore hard to replicate. 

Let’s say instead you only changed the creative and everything else remained the same and the results were much better than the original campaign. Then you can be pretty confident that the creative was what made the impact. 

This strategy leaves little doubt on what improved or degraded performance.

Apples to Apples Comparisons

Meta does not treat all creative types equally. What I have noticed is that Meta’s algorithm tends to prioritize spend towards dynamic catalogs first, then videos, then carousels, then single static images. If you test various creative ‘types’ in one campaign, it will generally skew spend accordingly and may not give certain creatives enough spend to judge them. If you want an apples to apples comparison of different creative, group images with images and videos with videos. This will give you a more ‘fair’ assessment of each creative. 

Using ‘Constants’

Let’s say advantage shopping campaigns (ASC+) work for you consistently. Then that’s your constant and you can test various creatives against that. See which creatives are giving you the highest ROAS, lowest cost per result, or whatever metric you use to measure success. Side note: ASC+ tends to skew spend towards 1-3 creatives so if you notice there are some other creatives that are getting sales but not a lot of spend you can duplicate the ASC+ and exclude the creatives that took all the spend in the previous and force it to test the others.

Let’s say no matter the audience you select, a certain creative always does well. Well now that’s your constant. You can use that to test various audiences like ASC+, Manual Broad, Original Audience Options, Interests, Lookalikes, Etc to see what gives you the best results. I see many people saying that audience targeting does not matter but using this strategy it is very easy to see what audiences are giving you the best cost per result and lowest CPMs.  In some ad accounts Broad gives the best results, lowest CPMs, etc. In other accounts, Broad is terrible and doesn’t work at all while more narrow audiences like interests or lookalikes give good results and significantly lower CPMs.

Label things clearly

Ad accounts with no system for labeling become a mess, fast. If you label every campaign, ad set and ad clearly – you will be able to easily assess the data. If you can easily assess the data then you can make decisions based on the data much more efficiently. 

I tend to name campaigns and ad sets by the settings I selected. Here is an example of a campaign structure labeled effectively. It’s a 1 CBO with 1 ad set and 2 video ads.

• September 12 | CBO | Adv+ Broad | Top 2 Videos

  • Adv+ | 18-65+ | Men | Adv+ Placements

    • Video | Woman Actor | UGC Testimonial | I Can’t Get Enough of X Hook

    • Video | Man Actor | UGC | I Did X So You Don’t Have to

This strategy definitely saves time. And time is money so…

Traffic, Awareness, Engagement…Oh My!

Ask and you shall receive. Meta is going to give you exactly what you ask of it – traffic. That doesn’t mean it has to be good traffic. Don’t be fooled by impressive vanity metrics like a super low cost per click, low CPM or high CTR. If it’s not getting sales, it’s most likely driving up your expenses with not much to show for it.

Awareness campaigns are pretty similar. They are going to get your ad out in front of as many people as humanly possible but generally don’t lead to many sales or conversions. Some argue that it helps get your brand out there and can help improve performance on other campaigns – however my biggest gripe is how difficult it is to objectively measure the results of an awareness campaign.

Engagements campaigns can get you likes and comments but again the traffic is generally very low quality. Run a low budget Facebook page like campaign and see who is liking your page. The results may surprise you. I have seen engagement ads work to some degree for driving DMs and calls but I see Leads campaigns outperform engagement most of the time. Some argue that engagement campaigns can supplement existing sales ads with the same PostID but again its impact is not easily measurable. Last, I feel like engagement ads tend to incentivize advertisers to focus on metrics that ultimately don’t pay the bills like number of views, comments, likes etc. 

All this being said, I also believe there are no absolutes with Meta ads. Are there traffic campaigns out there that crush? Absolutely. Are they few and far between? I believe so. If I see a traffic, awareness or engagement campaign that is profitable I won’t turn it off just because I don’t ‘like’ them. Be pragmatic. Keep what works, ditch what doesn’t.

At the end of the day, businesses need sales and more importantly profit to thrive. Sales and Leads campaigns are most effective for doing just that. Profit over everything.

Campaigns Set Up for Failure

Let’s say that your average cost per purchase is $50 and you consider that ‘good.’ What I often see are businesses running CBO’s with 10 ad sets, several creative in each and a campaign budget at $50/day. With this setup, it’s likely going to take ~10+ days for each ad set to get a sale. The budget is spread way too thin. Facebook ad campaigns need conversion data to optimize. If you only give an ad set $5/day with an average cost per purchase of $50, it’s generally not going to optimize well. My rule of thumb is to set up my campaigns in such a way that I should expect to get at least 1-2 sales per day per ad set. 

A better way to set this up would be a CBO with 1 ad set and a campaign budget of $50-100. Now I can be confident that I should in theory get at least 1-2 sales per day. Let’s say I go 4 days with no sales. Well I can be confident that this campaign is not going to work and I can close it and try something else. If I see 1-2+ sales per day, then I know I am on the right track. 

Next time you set up a campaign, ask yourself if you are asking ‘too much’ of Facebook. Am I spreading the budget too thin?

Thanks for reading. If you found this helpful, please share it with someone who could benefit from it. Leave questions and comments below, I will respond to all of them. Cheers!