As the foundation of best practices for social and search marketing campaigns, A/B testing your ad headlines, copy, and images is standard practice, but A/B testing videos is much less common. We believe that employing an adapted version of A/B testing for video offers one of the most powerful tools in the video marketer’s awareness-generating arsenal.
What is A/B Testing?
Before diving specifically into video intra-optimization A/B testing, let’s define it for those of you who aren’t as familiar with the practice:
A/B testing is the process of running two ads that contain one differing creative or copy element against each other, seeing which one generates the most audience interaction, declaring it the winner, getting rid of the other variation, and then changing another element of the winning ad and testing which of the two performs better.
Rinse and repeat to arrive at the best performing piece of content.
It is a powerful tool when testing how your audience responds to messaging, offers, or subjective qualities. While most often used with ads, you can A/B test e-mail openings, call scripts, or pretty much anything in your sales/marketing funnel.
Challenges of Applying Traditional A/B Testing to Video Ads/Content
- Videos take much more time to make
- They cost more to change after the fact
- From a technical standpoint, editing videos relies on one moment, or clip, leading seamlessly into the next. If you were to change part of the video,the rest of it might not make sense.
- Video also typically requires longer review processes and the opinions of more stakeholders, because the investment is more significant, which further inhibits the likelihood of making enough videos to have options to test them.
Because of these challenges, most video marketers will spend a great deal of time planning and strategizing, and hoping their videos achieve their goals. And when the campaigns fail, the marketers need to explain to their stakeholders that they just blew a ton of money on videos that didn’t drive results. The result? They revert to picture and copy ads and give up on video.
Video Campaigns Need to Listen and Adapt as they Run
How can video campaigns listen and adapt while they run? You guessed it: A/B testing. But how can we A/B test video ads, provided the challenges of cost, time, and capabilities we’ve already discussed?
Instead of making a number of different videos to A/B test, which would take an insane amount of money and time, we believe in micro-optimizing individual videos.
Intra-Optimization is the process of A/B testing high-impact, individual elements WITHIN each video over the life of the campaign to arrive at the best-performing version of each video.
Here is an advertisement my video marketing studio, Vybrary, created to explain the service, which was directly targeted at digital marketing agencies that specialized in Facebook ads. It gives a more fun overview than all this text can provide 😉
A sample micro-optimization process might look as follows:
- Start with the first video in your campaign, which has been designed to speak to a chosen segment of your target audience.
- Determine the most likely individual elements in the video that, if changed, could increase attention-span, shift emotional reactions, or speak to your audience on a deeper level (see below for examples)
- Write these variations into the video’s script.
- Cast and shoot the video, including these variations at the time of the initial shoot.
- The editor will create and gain approval for the first two versions of the video that will test the most-likely variables.
- The campaign launches.
- After actual audience interactions with both videos have been analyzed, according to a predetermined amount of data (10,000 impressions on fb, for example), we’ll declare a winner for each video based on key metrics (CPL, CPC, conversion rates, etc.)
- The editor crafts the next video, keeping that winning element and creating two new versions of the video that test the second-prioritized element for testing.
- This process repeats for the duration of the campaign.
Performing these micro adjustments may seem tedious and insignificant. However, consider how much changing one word of copy can affect a traditional ad. Now think of a video. Wouldn’t changing the first three seconds have a similar effect? We’ve seen simply changing the color grade of the video to increase conversions by 30%!
Compound these benefits over a month-long campaign, and you could be improving the performance of your initial video by 100-1000% or more for proportionally small increase in time/cost/money.
Suggestions for Video Elements to A/B Test
- The opening 3-5 seconds: What order of shots best grabs the audience’s attention? Start with a person’s face? Start with establishing the environment? A title? It’s impossible to predict what will grab the audience, but the ability to test the actual content of the first five seconds is a HUGE advantage.
- The protagonist/actors in the video: Especially if you’re unsure of your exact target audience’s demographics, it can be powerful to try different actors. If written into the original script and planned for during production, this can actually cost very little but have a huge impact on your campaign, as it subconsciously influences how your audience relates to your videos.
- The color finishing & visual tone: Perhaps one of the most under-rated qualities in video advertisements, the color grade can dramatically affect the performance of a video. By altering the contrast, color cast, saturation, or many other color elements, you can improve the performance of your ads by generating different emotional responses in the audience, without needing to re-shoot anything. As a professional colorist, I love taking advantage of this tool, because it’s one of the most powerful, yet under-utilized tools for improving conversions.
A glorious side effect of this process, is that it allows you to gather critical information about your audience, as well as run a successful campaign. Instead of a binary, ‘it worked or it didn’t’ situation, you’re able to ascertain what tone, messaging, appearance, or even specific kinds of content resonate with them. You can then return this data to other teams in your organization to fuel their outreach efforts. This, alone, justifies the cost and added complexity in the process.
