How a Cloud-Based Video Creation Service Uses Testing to Better Understand What Customers Want


What assumptions do you make about your customers? How do you validate if those assumptions are true or, instead, actually damaging conversion?

A/B testing can help you discover what really works with your customers.

Animoto, a cloud-based video creation service, usually has at least one test running every week and runs a total of about 250 tests with millions of customers on its website every year. I sat down with Brad Jefferson, CEO, Animoto, to get an inside look at his company’s testing practices and see what he’s learned about his customers along the way.


One of the tests that Brad’s team ran was to determine what type of sample videos would be most effective.


Vertical versus horizontal segment test

In the Control, Animoto took a vertical approach and showed different industry-specific videos on the landing page, like:

  • Real estate sample
  • Professional photography sample
  • Internet retailer sample

It wasn’t performing as well as the team thought it should be, so they tested a treatment with a horizontal approach. The landing page had video samples that would appeal to any industry, like:

  • Company overview video
  • Product description video
  • Customer testimonial video

By changing the type of videos that were shown above the fold, this test resulted in a double-digit lift in conversions.


How do your customers see themselves?

Animoto’s vertical approach made sense. Many companies break customers down into vertical segments in an attempt to deliver more relevant marketing and content to increase sales.

At least in this instance, customers did not see themselves as only living in (and caring about) a vertical-oriented solution. The customer was more focused on the types of solutions Animoto could offer, not how these solutions could be specifically applied to its own vertical.

Remember, a segment is only a hypothesis that you’re creating about your customers. Your customers may not self-identify in the same way you’ve chosen to break them up. Or, they may identify with certain segments for one kind of product or service, and an entirely different segment for a different type of product or service.

Make sure to test those segment hypotheses to see if your assumptions are correct and learn something new about your customers.


How can you get a better view of customers?

Keep in mind, A/B testing is a means to an end — that end being discovering what really works with your unique customers. Actually talking to customers, systemizing and documenting customer service feedback, recording customers using the product — all of these techniques can help you better understand your customers.

They can also provide qualitative data to help you interpret the results of your A/B testing, as well as come up with hypotheses for future tests.

“A main way that we’ve changed over the last three or four years is, not having that internal bias, but actually putting the product in customers’ hands, and watching them,” Brad said. “Videotaping them using our product. Videotaping when they get excited about it. Even though that’s qualitative, because you can’t do that for all customers, you start to see trends really, really quickly.”


You can follow Daniel Burstein, Director of Editorial Content, MarketingSherpa, @DanielBurstein.


You might also like

43 interviews from the MarketingSherpa Media Center at IRCE

Here’s Why Most A/B Testing is Boring and Not Really Worth the Trouble

The 2015 Video Marketing Cheat Sheet [Infographic] (by Megan O’Neill on the Animoto blog)

Online Testing Certification course (training from MECLABS Institute, MarketingSherpa’s parent organization)

You might also like

Leave A Reply

Your email address will not be published.