Join Brad Batesole for an in-depth discussion in this video A/B testing, part of Up and Running with LinkedIn Sponsored Content.
- View Offline
- [Voiceover] One of the best ways to optimize campaign performance is to A/B test variations. And you'll do this by setting up two identical offerings and then adjusting a single variable in one of them. So you might start by trying two different images, then you'll pick your winner, and go on to make two new advertisements, but this time testing two different headlines, then test two calls to action and so on. You can't clearly identify a winner if multiple variables are present per each test, so you want to stick to one change per A/B test and test it until the results are statistically significant and you've been able to determine a winner.
You want to keep your campaign running for a few days to also limit any environmental variables such as news events, site traffic, or the day of the week. Now when you're A/B testing, I recommend you set your ads to even distribution, as opposed to optimized. And you can do this by diving directly into the campaign and selecting the settings icon, which is this gear, next to the campaign at the top of the screen. Choosing rotate ad variations and then selecting rotate variations evenly.
And once you select that setting, your ads are going to be rotated evenly through out the testing phase to ensure balance. Once you've identified your top performers, you can pause the losing ads and switch back to the optimized click through rate option. When A/B testing, build new creatives to get a fresh start. Restarting creatives that have already run come with baggage like past performance, and a relevancy score which could bias the results of the test.
Find more ways to leverage the world's largest professional network with our other LinkedIn training courses.