I’d Test That: Determining the Conversion Power of Video with a Split TestPOSTED ON 14 March 2012
This week for I’d Test That, we’re taking a look at a product page on Cisco.com, the global networking company.
Cisco is a perfect example of a company that uses Adobe Test&Target and has tons of opportunity for testing and optimization. We’ve chosen a product page for our testing assessment. Now, this isn’t a typical page that takes you into a shopping cart experience. Cisco works with companies that buy large quantities of product at a time (we’d be curious to know the average order value, as it could be tied to some great metric goals).
Looking at this particular page, you are presented with a model series—the 3900—and four models within. There’s a lot of information on the page, including a downloadable whitepaper, 3D model, video and bulleted copy about the series. Additionally, the side nav has links for further information and specs on each model. And the conversion point is the “Let Us Help” module in the top right. Ideally, this page will drive a potential customer to convert using one of the options in the module.
After taking a closer look at the structure of this page, we watched the video and saw that it’s an overview of all four models, shot in one take. And considering how expensive it is to produce video content, we figured that was a good place to start optimizing. So how can Cisco effectively test video and come away with some valuable learnings for optimization?
We came up with a smart split test that could end up having a lot of value for Cisco. By testing these two cells against the control (test cell A), it will become very apparent what’s working (ie, driving conversions) and what’s not.
To Video, or Not to Video?
Test cell B would be pretty straightforward. We’d run the control, with the existing video, against a page that had all the same elements, minus the video. That would tell us whether the video was influencing people to convert either way.
Make the Information Easy to Access
A six-minute video for multiple products can cause some friction, especially if the customer knows the exact model they want. To really test the value of video for Cisco, we’d advise slicing this one into shorter clips so that each model has its own short video. Each video could be titled by model and then followed by the model’s specs in an expandable text format. This would give Cisco valuable learnings on which models were most popular, allowing for reshuffling of order. It could also give additional proof that customers aren’t responding to video, allowing an opportunity for cutting out the high cost of video production.
What’s Next for Cisco?
This brainstorm proved to be pretty fruitful. From the video test, we came up with two additional tests that would be logical follow-ups to the learnings gained. Of course, we’re not going to give all our genius away, but it’s great news for Cisco that there’s a wealth of testing and optimization that can be done on their site.
The Moral of the Test Is…
This is a great test for any company serving up products that have a lot of features and/or technical specs. So many sites bombard us with information, and given the choice of reading dense paragraphs of copy or watching a lengthy video, we might end up bouncing off the page. Testing what drives people to convert or to ask for more information is a solid place to start. Take a look at your product pages. Is there room to pull back a bit and test how you’re selling the product?
Brooks Bell is the industry's leading online conversion firm. We help large online retailers and subscriber-supported companies maximize their online sales funnel by testing and optimizing all points in the conversion path.