March 29, 2022

12 Tips for Facebook Ad Testing in 2022

Do you want to know how you can test your Facebook ads to lower your costs and produce more results?

Here at McLernon & Co., we often hear stories from our clients that Facebook ads they’ve tried to run in the past were ultimately very expensive per web traffic click or per lead form, or that those leads were low quality and did not result in a sale or another intended result. If this sounds familiar, please read on.

Often in these scenarios, there was no Facebook ad testing performed. Before diving into a large campaign spend, it is important to test different creative, against different messaging, with different audiences, in different post formats, so you can delineate what is effective from what is not before scaling up your Facebook ad campaign.

Here are 12 tips for Facebook ad testing.

1. Set Realistic Goals and a Timeline that Align with the Sales Funnel

We get it. You want that sale NOW! However, how realistic is it that somebody on Facebook who has no awareness of your company is going to make a purchase on the first piece of content they’ve ever seen from you? Not likely.

Before you start delivering conversion ads to users, they need to be taken on a journey and learn about your company, and your product or service, then understand why they are a fit with your product or service, and see evidence of quality before they are ready to make a purchase. 

Most social media activities, in fact, take place at the top and middle levels of the funnel. 

This is critical in your ad strategy. If you are targeting new audiences, you will want to begin testing ads that align with the top of funnel (TOF).

2. Confirm You Have Tracking in Place that Follows Your Ad Sequence

The better you are able to measure what’s going on throughout your funnel, the better off you will be able to analyze what is going right and what is going wrong.

Ensure that your Facebook Pixel is in place on your landing page, and every event downstream from there should be tracked. Here is an example:

People often forget to include the thank you page in their Pixel. This is helpful so you know (and Facebook knows) that customer is definitely a lead or made a purchase because they reached that confirmation page. This will be key information in your ad strategy, as you likely will either want to pause ad delivery to that customer or you may wish to deliver a different type of ad to them in the future.

3. Determine and Set Up Your Audiences for Testing

Keep in mind that different audiences may respond differently to the same ad creative. For example, the way a 30 year-old woman might respond differently than a 60-year old woman. Or people who are interested in travel may respond differently to people interested in sports. Think about who your potential audiences would be, and then start dissecting them into sub-target groups.

Your ad testing to different audiences might look something like this:

4. Change Only One Variable When A/B Testing

A/B testing is the process by which you test two variations of an ad to determine which version produces the stronger results. When A/B testing ads, it is important to change only one variable at a time so you can pinpoint what is working in the ad and what needs to be refined.

Visuals are what stop people from scrolling, so it is best to start by testing different variations of your creative before you test variations of your caption copy, headline, and call to action. You may also want to test single images vs. video or vs. carousels, and/or different imagery entirely. (Check out the next tip about getting inspiration for your creative!) 

One thing we have noticed here at McLernon & Co. is that video content where a person is talking directly to the camera filmed with their phone (think TikTok) is delivering a high level of engagement both organically and in paid ads. Facebook and Instagram also prefer and prioritize video content.

Usually, we like to allow ads to run in testing for at least 7-14 days before choosing the better performing ad and making optimizations, especially for website traffic ads where the Pixel is gathering data. Facebook currently tracks customer sequences across a 7-day period. At minimum, ads need to test for 4 days to give Facebook a chance to learn delivery optimizations.

If you have a low-performing ad, this may be due to the audience it is being delivered to or it may be other factors. You may wish to try testing it with a new audience and see how it does before eliminating it from your ad sets.

5. Use Facebook’s Automated A/B Testing Wisely

Facebook offers advertisers the option to test up to 5 ads against each other in an automated way with alternate copy, headlines, or visuals to determine which gains the better click through rate. This automated process is best only when you feel confident in how a particular ad will perform (your control ad), but want to test some of its variables for social proof. When you have multiple ad sets and are testing all equally against each other (rather than testing against one control ad), we find it’s better to manually set up your testing rather than automating, so you can allocate equal effort and equal ad dollars to each ad set.

6. Research Other Pages and Competitors

You can gain a lot of intel on the kind of ads that would work for you before you even begin testing by researching the ads of others in your industry. Facebook makes this information publicly available through the Page Transparency menu.

Head over to a competitor’s Facebook page, scroll down to Page Transparency, and click the See All button.  If they are running ads actively, you’ll be able to see them by exploring their ad library. Ads that have been running for an extended period of time are likely working effectively for them, so you’ll want to take note of these!

7. Look to Your Own Organic Content

Do you have an organic post that is performing well? Don’t forget that all of your organic posts already “out in the wild” are part of your testing! Your organic performance is a real-world demonstration of what your followers will respond to, so use this as inspiration for your ads.

Sometimes, you may want to boost a post. Posts that already demonstrate a strong level of engagement are the best posts to boost with ad dollars. People are more likely to click on ads when they see a significant number of likes, comments, and shares. It’s important to note, however, that boosting posts offers limited options compared to building an ad through Ads Manager. For example, boosted posts allow you to choose objectives like website traffic, post or page engagement, and local business promotions, while ads built through Ads Manager also allow you to choose objectives like store traffic, conversions, and lead generation. 

8. Manually Select Your Ad Placements for Testing

Facebook gives you the option to let them choose where to place your ads automatically across the Metaverse, but we prefer to manually select ad placements as this ensures we are leveraging the channels we want to leverage and ad dollars are not going towards other placements we instinctively know will not be effective. Additionally, by manually selecting your ad placements, this gives you another opportunity to test your ad. You may want to try testing Facebook against Instagram, or see how Instagram Reels performs for you as you continue making optimizations.

Depending on the nature of your business and your target audience, you may wish to test mobile versus desktop as well!

9. Watch Your Ad Every Step of the Way

Monitor your results closely and every step of the way from testing through optimization, as each event will affect what happens downstream. Some items to looks for include:

  • CPM: How high is the cost per 1,000 people?
  • Link Clicks: Your CTR should be more than 1%. If it is not performing well, you should try new creative, new caption copy, or target it to a new audience.
  • Website Actions (monitored via having the Pixel set up): 
  • Are people taking the action requested (completing a registration, completing lead form, etc.)? You should see at least a 10% conversion rate. If you are not, you may need to optimize your landing page.
  • Add to Cart: Should be at least 10% of landing page views.
  • Sales: At least 10-20% of Add-to-Carts should be purchasing.

10. Increase Your Budget Gradually for Winning Ads

Now that you have a winning ad and you are ready to scale it up, be sure to increase your budget gradually. Facebook needs to readjust its learning and algorithm every time you make a change to an ad, so you can see your results crash with too large of an increase in budget at once. A good rule of thumb is to increase your budget by 20% every 24-48 hours.

11. Don’t Be Fooled by “Ad Fatigue”...Unless the Data Shows Your Ad Isn’t Working!

Many marketers can occasionally question if the creative for an ad has been running for too long and people will get tired of it or annoyed by it– aka “ad fatigue.” However, if the data shows it is still working– then it is still working! This doesn’t mean you can’t rotate in new ad creative. In fact, you should! When you introduce new creative/messaging, see how it performs against your existing ad and only then make a determination about the most effective way to proceed.

Similarly, don’t become emotionally attached to an ad because it once performed really well. Look at the data and let the data tell the story. Long-term it is a good practice to always be testing to get the best ad performance for your dollars.

12. Use Your Ad Learning from Facebook to Influence Ads on Other Platforms

Each social media platform has a distinct purpose and use, so an ad that performs well on Facebook or Instagram may not see the same results on other platforms like LinkedIn or TikTok. However, your successful ads may be helpful in determining which content has potential for success with some tailoring of the creative, messaging, and landing page. Other platforms may also be able to target certain audiences that you are unable to target on Facebook.

For example, Facebook has a lot of data about users on a consumer level, but not as much on a professional level. Facebook may know a user’s industry and job title, but that is not tied to their company’s growth rate. In this example, you may want to repurpose your Facebook ad for LinkedIn, targeting users who work for companies with a significant year-over-year growth rate. Knowing that people use LinkedIn as a platform for networking, job searching, and for professional learning, you will want to be sure to tailor your messaging to fit this mindset.

Interested in learning more about how to run an effective ad campaign for your organization? McLernon & Co. offers intensive strategy sessions including hands-on guidance to help you build your ads. Additionally, McLernon & Co. can directly manage your paid ads monthly or even for your one-off campaigns. Connect with us for more information or a quote! Please email amanda@mclernonandco.com with the subject line "Paid Advertising 2022.”