Skip to main content

How great would it be if you could set up your ad campaign and simply walk away while the sales come in? While it is possible over a short period of time, unfortunately, creeping frequency and ad fatigue will inevitably kick in at some point, and results will dip. This is why testing is a crucial part of keeping up and improving performance over time; you always need to be on the lookout for a new audience or a creative refresh. Having a testing roadmap can facilitate the process and make sure you are always challenging the account (and yourself) to strive for better.

So how do we go about this?

  1. Familiarise yourself with the account

Get to know the ins and outs of the account and understand the business’ needs, and, more importantly, identify the areas that offer room for improvement and testing.

  1. Find the right opportunities

Start at the “root of the problem” and begin by looking at campaign level, finding elements which can be tested and improved: is the campaign objective appropriate for the business’ target? Could you challenge the budget optimisation? Then, move on to the adset level and look at things such as audience and placements.

When it comes to ads, there are many variables to test, such as ad type (video, static, carousel, portrait vs square), copy, headlines, CTAs, colour, image etc.

  1. Plan out the tests

Now that you have a comprehensive list of tests, set up a timeline, starting with the most probable wins and those which will make the biggest difference: e.g. test images first, rather than CTAs or other subtle elements.

To get conclusive results, try not to run more than one variable at a time, and make sure the testing elements are dramatically different. Use your top audience to test new ads, and, if you want to bring in a new targeting segment, run with your best performing creative.

  1. Test and find the winner

There’re more methods of setting up tests in Facebook; the quickest way is to just use the A/B test tool from the Ads Manager main table, which lets you select the variables you want to compare. The A/B testing tool is quite prevalent in the Business Manager interface, and it’s likely you will stumble upon it on various occasions. Alternatively, go to the Experiments tool, if you are looking for a more comprehensive testing strategy.

Set up your test and let it run for a few days, ideally until it acquires a minimum of 4-5k impressions. Facebook will determine the winning variable for you, based on your chosen objectives (ideally these would be stronger metrics, such as purchases).

  1. Make the winner work harder

Once the winning version has been identified, roll it out across your campaigns. It is ideal to keep a log of the test results, to reflect and see what has been tried and successful. Use these insights to make informed decisions with future strategies, and feedback to the relevant departments in your business.


Discover more about how our paid search expertise can help you to improve performance, or get in touch directly.