Register

How to Use Split Tests on Facebook to Find Your Target Audience

Marketing

photo

Facebook’s Split Testing tool enables you to test your campaigns to find out which one is the most effective.

We are excited to announce the first round of leaders who will bring our 2020 theme HUMAN.X to life at our global conference in New York on May 5-7.


With approximately 2.23 billion people logging on to Facebook every month, the social networking service has definitely secured its place in the ecommerce industry as the top platform for both B2B and B2C businesses.

But with over 80 million small- to medium- business pages vying for the coveted Facebook user’s attention, how do you optimize your ads for your marketing objective (whether it be for brand awareness, views, traffic, lead generation, engagement or conversion) and at the same time, make sure you get your money’s worth?

Thankfully, you won’t have to aimlessly conduct trial and error experiments. Facebook’s Split Testing tool enables you to test your campaign by setting a defined budget and duration and then choosing any of five vital ad elements (more on this later) to find out which is most effective.

So how does this Facebook feature actually work? Continue reading on, but before you get too excited, below are five no-nos when conducting a Facebook Split Test:

  1. Don’t go blind: You can’t conduct split tests based on gut-feel. Make use of the insights and analytics tools Facebook provides to help you tailor fit your test design.
  2. Don’t go too big: You can’t be overly-ambitious. While it may be tempting to assume that you can potentially reach Facebook’s billion-people population… Let’s be realistic, you can’t. It’s not like you have an unlimited budget to spend, and nor is your product or service for everyone too.
  3. Don’t be stingy and impatient: Reliable results can only be generated if you spend enough (if you have no clue, the tool provides an ideal figure) on the campaign and let an ad run for at least four days. As well, don’t be tempted to cancel an ad if it’s doing poorly.
  4. Don’t major in minors: While testing which design elements and ad copy your audience engages with the most is a good strategy, you shouldn’t waste precious ad spending on a test that will simply determine which font type or background color works best. Choose a vital element such as the ad headline copy, image, or video to differentiate. Then try to strike a balance between variation and similarity, as shown in the example below that tries to work out which is better—a text only ad or an add with an image.
  5. Don’t disregard optimizing for multiple devices: 95.1% of active Facebook users access the social media platform via their smartphone. In the same way that mobile app testing is conducted to make sure a developed application works seamlessly across different devices and operating systems, the same needs to be done with your adverts. You may create your ads on a desktop computer, but make sure the designs are responsive when viewed in multiple devices with varied display sizes.

Source: Jeff Bullas

The Ultimate Guide to Creating a Facebook Split Test to Determine Your Best Target Audience

Now that you know what not to do when creating a Facebook split test, below is a rundown on how you can make the most out of this innovative tool that will help you strategize for meaningful and engaging Facebook content.

1. Setting up your campaign and choosing your ad objective

Once you decide to conduct a test, the first step is to Create a campaign in your Facebook Ad Manager. This is button 2 in the photo below.

Source: AdEspresso

Then, you will be prompted to Choose your objective. Your Facebook ad can either be for awareness, consideration or conversion as shown in the possible options below.

Source: AdEspresso

2. Turning on the Split Test feature

After choosing your desired objective, you can now tick the Create Split Test box. Note that the feature has to be enabled in advance and cannot be simply activated for already ongoing campaigns, unless you edit your ad set (which will interrupt the current campaign).

Source: Social Media Examiner

3. Selecting the variable

There are five variables to choose from when testing out your campaign, each addressing a crucial question in Facebook content creation:

  • Creative: Which design elements work best?
  • Audience: Which specific audience engages with your ad the most?
  • Optimization event: Which ad delivery method will help you achieve your campaign goal?
  • Placement: Where should your ads be shown?
  • Product set: Which inventory items does your audience engage with the most?

Keep in mind that these variables can only be tested one at a time. For this guide, we will be focusing on choosing the appropriate target audience.

Source: Social Media Examiner

4. Making Your Ad Sets

With your campaign basics set-up, you can now create your ad set that will prompt you to choose each set’s corresponding audience (more on this later) and budget, schedule as elaborated in the next step. A maximum of five ad sets can be created and tested against each other.

Source: Business2Community

5. Determining the budget and schedule

If you are new to Facebook advertising, the social media platform makes it easy by setting a default minimum and ideal budget for each ad set you are testing for. The budget can either be split evenly or weighted.

As for scheduling, Facebook recommends a minimum of four days and a maximum of 14 (if you have budget to spare) to get valid results. Remember, scheduling isn’t just about the campaign duration, but also choosing the best time for the post to go live.

Source: Social Media Examiner

6. Zeroing in on your Target Audience

Split testing to determine the most suitable target audience can be done by specifying which group of users will see each ad set, and then customizing other demographic information such as location, age, gender, languages, interests and behavior.

Source: Social Media Examiner

If you have not created a custom audience yet, you can do so based on the different types of users who have already engaged with your page.

Source: Social Media Examiner

You can also choose to generate a lookalike audience based on the demographic of those that have interacted with your enterprise. Take time to understand how customizing and retargeting your audience works.

After you have differentiated the audiences for your ad sets, you should be able to view your Split test summary.

Source: Social Media Examiner

7. Designing Your Ad

After setting up the parameters for your audience split test, now comes the fun part of creating your ad. Facebook has five different layouts (carousel, single image, single video, or canvas) to choose from that features different ways of combining text, image and videos.

Source: Buffer

After making a selection, you can now upload your image and fill in the copy such as in the example below. Then, repeat the process depending on how many ad sets you will be creating.

Source: Buffer

Keep in mind that since you are split testing for the audience variable, other ad elements must be identical or have minimal differences only as exhibited in the example below that had identical copy but modified images.

Source: Business2Community

8. Narrowing Down the Audience Split Test Feature to Target by Key Traffic Type

Your Facebook page traffic can be narrowed down into three types—cold, warm and hot:

    • Cold traffic – users that have zero knowledge or interaction with your brand
    • Warm traffic – users that have heard of you brand and have interacted with your page but have not been converted in buying your product or service
    • Hot traffic – users that have recently purchased from your business and can be encouraged to buy again (upsell) or those that have been converted before and now need reminding (re-engagement)

Attracting cold traffic primarily involves creating ads for awareness by providing content in the form of blog posts, surveys, guides or reports that is relevant to your enterprise’s industry as displayed in the example below. Split testing at this first stage in the conversion funnel usually entails setting up lookalike audiences and experimenting with creative, optimization or placement elements.

Source: Business2Community

On the other hand, testing engagement of warm and hot traffic can be further narrowed down to see which segment will respond best to your goal. For example, if you want to know which ad will drive the most webpage visits, choose traffic as your campaign objective.

Source: Social Media Examiner

Then, set the Audience variable of your ad sets by choosing from any of the readily available Custom Audience category types Facebook provides (as shown in Step 5) or even compare with a specific warm audience that has interacted with a video you have posted.

Source: Social Media Examiner

Custom audiences can reflect users at whatever stage they are at in the sales funnel. By using this split testing feature, you can optimize your ads for each audience segment to ultimately lead the audience from awareness to conversion.

9. Assessing your results and what to do next

Once your campaign duration is over, you will receive a notification and an email of the results. The best performing ad set is determined by the advert that yielded the highest result at the lowest cost possible. Other aspects included in the outcome summary are:

  • The number of times your ad accomplished the intended goal of the campaign
  • The average cost per achieved result
  • The total amount of money spent for the split testing exercise

The results can either equip you with better insights for your next campaign, or you can choose to adopt or renew the winning advert. Facebook provides a confidence level percentage so you can be assured that you will most likely get the similar desired outcome. The general rule of thumb of a clear an actionable result is a confidence level of at least 75%.

Conclusion

Facebook’s Split Testing feature allows you systematically test a variety of integral ad components through an isolated approach to see what blend works best for your target audience. Keep in mind that you shouldn’t be disheartened if an ad you run does poorly. Making these kinds of “mistakes” are not a waste of money. Learning what doesn’t work is a worthwhile investment as well.

Now, it’s time to put new learning into practice. Head over to your Facebook Ads Manager, up your game by using other ad tools and management platforms, and create those split tests!

Join 100,000+ fellow marketers who advance their skills and knowledge by subscribing to our weekly newsletter.

WATCH OUR 2019 PROMO




Newsletter Subscription

Get the latest insights, trends and best practices from today's leading industry voices.


Learn More

Write for Us

Interested in sharing your ideas and insights with the world? Become a SMW News contributor and reach 300k readers each month.

Apply Here