Facebook Rolls Out Experiments to Simplify the Campaign Testing Process for Marketers

It combines elements of what was previously known as Test and Learn, Split Testing

Facebook rolled out Experiments, a way for marketers to combine the results of multiple types of tests of their ad campaigns in order to optimize their performance and develop a deeper understanding of their impact.

The social network said in a blog post that Experiments combines testing elements previously referred to as Test and Learn and Split Testing to give marketers on its platform a single place where they can A/B test campaigns, as well as measure conversions and brand lift.

Facebook wrote, “Testing is a powerful tool for marketers that allows them to make more impactful decisions by learning what works, what doesn’t and where new opportunities for performance growth lie. We’re launching a new approach to testing on Facebook. Experiments help marketers optimize ad performance and give them a deeper understanding of the true impact of their ads, allowing them to make better business decisions.”

The social network said that prior to testing, marketers should evaluate the effects of different social and environmental factors on consumer behavior and adjust their plans accordingly, adding, “When measuring effectiveness of a campaign, carefully consider the implications of using those results to inform future decisions, while A/B testing can provide relevant short-term insights.”

Facebook pointed out that A/B tests enable marketers to compare strategies for creative, audiences and other variables, adding that while automatic campaigns eases the process for advertisers, A/B tests can be used to compared manual and automated optimizations and gain additional insights.

Experiments also enables brands to conduct holdout tests, comparing the conversion rates of people who saw ads with those who were intentionally withheld from seeing them so that they can calculate the lift the ads provide in conversion events such as purchases.

Brand surveys, another element of Experiments, give marketers the chance to poll people who have seen ads and the holdout group that hasn’t, and then aggregate and compare results.

The social network wrote, “Experiments are designed to help marketers make better strategic decisions with more powerful testing tools. It’s easy to get the right test up and running, fast. Just select the test, add information and Experiments will set up the test.”