The most common Facebook Ads mistakes to avoid

Working at an agency and as a freelancer has allowed me to review lots and lots of Facebook Ads accounts. Even though I have reviewed accounts from multiple industries, there are some mistakes that keep repeating. Here is a brief blog post about the most common Facebook Ads mistakes and some tips on how to avoid them

Pixel Events

The most common mistake is not going after what you really want. One example would be ‘purchases’. Another would be ‘leads’. It used to be true that Facebook needed data in order to optimize an ad account. This is not the case anymore. Now you can start a campaign by optimizing for the objective you want instead of creating a funnel that might prove costly to create and/or maintain.
I have audited many accounts that opted for every step of the funnel only to realize that 75% of the budget could go to ‘purchases’.
Another thing to keep in mind is that, pixel events should be functional and CAPI [Conversions API] should be present.
Especially if you have a store, CAPI should be fully functional.

Age groups

Age groups can be a tricky thing to optimize for various reasons. The first reason is that data reporting is gone. When you visit your Facebook analytics you will see that the demographics tab is deprived of data. One could argue that you can bypass that hurdle by visiting your Google analytics and filter the converting traffic by the appropriate demographic metric. The problem with that thought is that Google Analytics doesn’t always have that data, the people who buy are not always the ones that see the ad. This is pretty common between family members where one member influences another in order to make a ‘purchase’. Also, data-sensitive extensions and browsers will never reveal such data.

What you should consider here is that unless you have trustworthy data that one age group buys more compared to another, it would be wise to run some tests in order to make sure if you can trust the data that flow through your website.

–Placements

I have seen a lot of accounts that use ‘Automatic placements’ as the only placement strategy. I won’t argue that such tactics will reach more people and increase your audience. It can even make your ads more memorable according to Facebook.
One notable problem is control. There are groups of people that only have one device for accessing the internet and that is their mobile phone. So, there is no point in adding desktop placements to your placement strategy.
Another problem. that might be catastrophic for your ad spend, is that automatic placements optimization won’t optimize for media. When you create ads for mobile, you need to make sure that the media are appropriate for mobile devices.
This means little to no text.
Conversely, when you create ads for desktop devices, your media have more room in order to present your products or services.
Media for desktop ads will get cropped or resized in mobile device placements and media for mobile might not look good for desktop placements.

That said, you should be more vigilant with your placements and consider Manual Placements for:
Brand Safety
Creative Considerations
&
Overall strategy

Pro tip
If brand safety is a major concern for your client or your Facebook Ads account then you should always remove ‘Audience Network’ from your placements. If that isn’t a concern of yours and you are only interested in purchases, then you are free to use ‘Audience network’.
Do remember though that you will have little to no control over where your ad shows up.

More info about Automatic Placements

More info about Manual Placements

More info about Image Guidelines

Bottom line: Placements are very important for the success of your Facebook Ads account[s]

Exclusions

The problem with exclusions is that accounts will either have too many or none whatsoever. Accounts that have too many exclusions are essentially limiting their potential pool of buyers. This (almost) always translates as reduced sales and/or leads. Conversely, accounts that have no exclusions might run into the problem of sending ads and creatives that might put some people off.
Example
Sending a remarketing ad to people who just bought from your BOTF campaigns. It gets even worse if they haven’t received the item bought and you hit them with a better offer than the one already paid off.
This is “adding insult to injury” type of moment and you might want to avoid that.

In essence, exclusion lists should work together with the ad, media, offer, and stage of the funnel.

Bottom line: Exclusions can be a balancing act

Too many ad sets

Facebook will inform potential clients that CBO campaigns are best suited for campaigns with at least 2 ad sets but no more than 200 ad sets. Even though this sounds good in theory, in reality however CBO campaigns will optimize per audience size.

“Remember that audience size may affect budget distribution. If your audiences are significantly different in size, ad sets with the largest audiences will likely receive the most budget with CBO.”

This will wreak havoc on your sales and ROAS.

What you need to remember about CBO is

You shouldn’t exceed 70 ad sets per campaign
Add additional ad sets in bulk
It takes up to 2 hours for CBO campaigns to re-adjust
Don’t pause and unpause your ad sets
Input changes in bulk to avoid long learning phases
Ad sets with low delivery are ‘bad’

Follow the link, If you want to learn more about CBO best practices

If you are short on time then you could add multiple ad sets as long as the audience sizes are ‘fairly’ similar in size.

But if you want control of a CBO campaign then you should put only one asset per campaign run your tests on the campaign level instead of the asset level.
The problem with that is that you need more time to manage your campaigns and quite possibly a bigger screen because you will need to scroll through pages of campaigns.

Bottom line: Consider the number of ad sets

Lookalike audiences

Lookalike audiences are usually a hit or miss in most of the accounts I review. The most common mistake is Lookalike audiences with limiters such as age, language, etc. When you create a Lookalike audience, the system leverages information such as interests, behaviors, demographics from an existing source audience. When you put extra limiters, you essentially reduce even more the pool of potential buyers.
Another mistake I often see is people creating Lookalike audiences from insufficient source audiences. This can be everything from audiences that barely have enough people to audiences filled with content viewers.

Follow the link, If you want to learn more about Lookalike Audiences Best Practices

Bottom line: Lookalike audiences shouldn’t be limited unless they are enormous [over 5 million users]

Duplication of ad sets

Many times have I see duplication of multiple campaigns with no changes whatsoever because the client read somewhere on the Internet that by duplicating campaigns and ad sets every two hours you are forcing Facebook Ads to serve you a ‘fresh’ audience. That is simply not true.
What is however true is that sometimes, campaigns that were performing optionally [eg 10 ROAS] start to generate 0 sales.
Personally, I have seen that if you duplicate once, you will get them “unstuck” and, eventually, start to generate sales again.
So, don’t duplicate campaigns that didn’t run enough and/or are generating zero sales.
Let the campaigns run before you consider duplicating.

Bottom line: Duplication of Ad sets will not skyrocket your ROAS [Return on Ad Spend]

Small Interest Groups

Even though enticing, small interest groups rarely work. In the post iOS era, they work even less. When I hear Facebook Ads account holders saying that “the audience group is spot on but I can’t it to spend the budget” then my immediate thought is too small an audience.
My advice is to begin introducing more relevant audiences to your campaigns.

Bottom line: be more liberal with your targeting

learning phase

‘Learning phase’ is an often misunderstood concept in Facebook Ads. People believe that if they get the campaigns out of the learning phase then ROAS will increase. That is rarely true. It is however true that exiting the learning phase will improve the performance of your ads but the key phrase is, improve and not skyrocket.
If your ads are not performing well then exiting the learning phase will not perform miracles for your ROAS.

You can fix the ‘Learning phase’ by increasing or reducing the daily budget, changing creatives, switching to CBO, utilize dynamic ads, experiment with new audiences, create a new offer, etc.

TIP

One way to find the appropriate budget is to find the CPA of your industry and multiply by 50 and then divide by 7. This will give you the average daily budget that should [hopefully] get you out of the ‘learning phase’. This doesn’t mean that if you don’t have the budget needed your Facebook ads will tank, but it is a metric to help you understand the environment and the competition.
Because, when it comes to advertising dollars, someone is willing to outspend for each and every placement and channel.

Bottom line: ‘Learning phase’ is more of an indicator than a warning

Creatives

When it comes to creatives, things are less forgiving. Facebook Ads accounts with creatives that are not liked by the users will be marked as
‘Low-quality’. This greatly reduces the reach of your ads and undoubtedly reduces your ROAS and/or MQLs.
Additionally, bad creatives will cause problems with dynamic ads since all the variations will look unattractive.

Creatives need to be liked by the user, the system [A.I.], and fit the placement. Users need to click them, A.I. needs to approve them, and the creatives need to look good on the [specific] screen. Creatives designed for Desktop will contain information that will get cropped on mobile devices. This will confuse the user and will not click the ad. This means that you will lose sales.
Also, the cost per link click will skyrocket and that is not good.

Bottom line: Creatives [media] are extremely important for the success of your Facebook Ads account.

Retargeting

The most common mistakes I see in Facebook Ads remarketing campaigns are big budgets, wrong audiences, high frequency, wrong ad copy, and wrong media. When it comes to budgets, you should always be mindful of the size of the audience and the size of the budget. The majority of the budget should go to cold audiences because you need to find new customers. Keep that in mind.
When it comes to frequency, I have seen remarketing campaigns with a frequency of 10. That is just bad because you are, essentially, driving people away.

Bye for now
Panos