Facebook Split Testing: Should You Be Using It?
If you’re going to have a successful Facebook marketing campaign, you need to constantly monitor your ads to see what works and what doesn’t. Before, this took a lot of very slow trial and error, and while there is still a lot of that involved in any successful marketing campaign, Facebook has made things a little bit easier with its split testing feature. This can help you get the most out of your marketing campaign by determining exactly which variables are the most effective and which ones need to be dropped. It effectively turns you into a more data-driven marketer, and we’re going to discuss how you should be using it.
How Split Testing Works
Facebook’s split testing feature is intended to help answer questions about different components of your campaign instead of simply making assumptions when you aren’t getting the numbers that you want. It allows you to test five different variables – ads, delivery optimization, audiences, placement, and multiple variables (essentially a miscellaneous category).
To get a better idea of how this works, let’s take a look at the “Audiences” category. If you were to select this category, Facebook would divide your audience into random groups with no overlap and deliver data on each one. This kind of randomization ensures that the test is conducted fairly without any influence from other factors that could skew the results. You can have up to five different variations of your test variable and determine the cost of each one. Facebook will automatically calculate this and let you know which variable will have the lowest cost. Facebook will then use the data from the test along with thousands of simulations based on it to provide you a confidence level in the results. This will be presented as a percentage, which will represent the chances that you will get the same results if you run the test again.
This is much more useful than simply trying to determine the effectiveness of your ads by looking at your sales. For starters, the fact that you’re testing the ads means that you are getting relevant data without spending the time and money to actually run the campaign. Granted, you still have to produce the content itself – something that can get expensive if your most effective ads involve video content – but any tools that give you the chance to tweak a marketing campaign and make sure that it will deliver positive results will always be welcome. After all, you will be effectively testing your campaign against a cold audience when you’re just starting out. You need some kind of reassurance that what you are doing will be worth the effort.
How to Use Split Testing for Your Own Campaign
To create your own Facebook split test campaign, first click on the green “Create” button on the Ads Manager dashboard. You will be able to use either the Guided Creation option or the Quick Creation option for your campaign depending on how you’ve created campaigns in the past. Quick Creation is preferable for the purposes of split testing, so switch it to that if your dashboard defaults to Guided Creation. As with any campaign that you create on Facebook, your first step in a split test is to decide what your campaign objective should be. For example, if you’re creating a video ad campaign, you might want to choose Video Views from the Campaign Objective list. You can also choose to look at traffic, app installs, conversions, lead generation, or any other analytics that might be relevant for your campaign.
From there, you should have the option to turn the Split Test switch on. When you do this, you will be able to choose the variable that you want to test. If you were to choose the Audience variable, you can choose which ad sets to test. If this was the only variable you chose, Facebook would test the same ad for the different audiences that you’ve selected. For example, you might choose to test your ad with a saved audience that represents your main demographic along with a random lookalike audience.
In order to get sufficient results for your test, you will need to set a budget for your campaign. Facebook will suggest a budget that will prevent you from having insufficient data, but it’s ultimately up to you to set a budget. If you haven’t run a split test campaign before, you will most likely be given a minimum suggested budget that will produce relevant results without costing you too much money. You can choose between a daily budget or a lifetime budget. Since you’re just running a test of your ad campaign, you’re better off with a daily budget at first, especially since you can continue to apply this budget to the winning ad when the test is over.
Your next step will be to select a schedule for your test. Although a four-day testing schedule should be enough to give you the information that you need, it’s recommended that you start with a seven-day schedule at first to make sure that you achieve the best possible results without wasting too much money.
Once the test is over, you will receive an email about the results. After that, it is up to you to review these results to find the ad set that performed the best. All that is left after that is to implement the ad groups and run your campaign.
Even though split testing is a very effective tool in determining how a Facebook marketing campaign should operate, you should never rest on your results for long. Things change quickly on social media, so keep on running new tests while paying close attention to the campaigns that you already have running. This is the only way you can stay on top of how your brand is perceived and how your potential clients are reacting to your ads, so don’t be afraid to throw out what isn’t working and tweaking the elements of your campaign when needed.
Latest posts by Heather Hart (see all)
- New Photo Stickers and Photo-Tagging from LinkedIn - May 18, 2019
- Facebook’s New Guide on Creating Effective Stories - May 15, 2019
- Top Tips for Getting More Leads Through Social Media - May 15, 2019