Every marketer loves the big wins – the game changers, the four-baggers that propel results to a whole new level.

Man, those are fun.

But those of us who’ve been doing this long enough know that as much fun as those big wins are, it’s really our ability to find those small wins tucked in a corner here, or a tweak there, or an experiment here, that add up to success.

Back when I was heading up marketing at FreshBooks I wrote a post on this topic and called this approach to growing business playing small ball.

Let’s play some video marketing small ball, and start sharing ROI wins

With video marketing still being so new and in such a massive transition, I thought it would be a great idea to take the small ball concept and start sharing some of the little tests and experiments I’m doing here with respect to finding some easy and consecutive wins using video.

When we re-launched the Vidyard website in February, we took the opportunity to really put some techniques to the test and I’m excited to start sharing the results that we’ve seen from these innovations.

My goal is that, over time, together we’ll keep finding those nickels and dimes that really add up to meaningful growth. And by taking small steps, we take something like video marketing from potentially being overwhelming to something that becomes easy to execute and digest.

That’s the plan. So let’s get started.

How we increased the CTR on our home page video by 15% without actually touching the video itself

I have to admit that how we did this just makes me smile, mostly because it’s just so obvious and way too easy.

We did it by testing different thumbnail images for our video.

The thumbnail image is the static image that shows up for your video before anyone hits the play button. We found that changing that image can have a pretty big impact on how many people will actually click to watch the video.

The first part to getting any kind of ROI on your video content is to have people actually click to watch! So spending time optimizing your video’s thumbnail is a solid investment.

Which thumbnails to test?

Once we had our homepage video completed, it was time to choose some thumbnails to test. When it comes to split testing I have found a few things to be true, at least in my experience.

First, test executions that are “testably different.”

I love that game where you’re shown four pictures and you have to guess which one is different…and it’s usually something so subtle like someone’s face having one more freckle than the others. Well, for split testing, you want to go the opposite route. You want to find a few executions that are as far away from each other as possible. This lets you cover more “directional” ground. If the images are too similar, one still may win in the test, but you might not be able to distinguish why it actually won – and that means you haven’t really learned anything. So, choosing very different images will give you some good direction as to what kinds of images resonate best with your target.

Second, be sure to choose something you are sure won’t win.

Let’s not let our own bias get in the way of a solid mathematical test. I’ve been surprised so many times now by which leg has won in a split test that I’ve stopped being surprised. To that end, I always include executions that I don’t think could possibly win – just to make sure I haven’t biased the test from the start. Having said that, all executions should line up with your brand and represent you, or anyone in the video, well.

The thumbnails I chose, and the one I thought would win

Below is a screen shot (click the image to enlarge) of the three thumbnails I chose . One had charts and graphs only, one had charts and graphs and Steph’s face, and one had the image of our “black box”.

AB Split Screen Test

Before starting the test, I was pretty sure the one with the charts and graphs and Steph’s face would win. It had a great visual of our analytics in action, plus it had a person in it – two elements that have proven to engage. I also put the “black box” image in there because it was a theme that we’ve started using in our communications so it would be neat if it won, but I didn’t think it would because, well, it’s just a box.

You can see that I was using our own A/B thumbnail feature and this allows me to also set how frequently I wanted to show each leg of the test. I chose to have each leg show an equal number of times.

The results – not what I expected

After quickly generating close to 5000 impressions for each thumbnail and between 400-500 clicks, the results were in.   The winner, with a 9.69% click-through rate, was the “black box” thumbnail. Glad I decide to include it! My favourite to win came in second, at 8.61%. In third place was the image with just the charts, at 8.39%.

With a 9.69% click-through rate, the “black box” thumbnail outperformed over the third-place thumbnail by 15.5%. That’s almost 1.3 more clicks for every hundred impressions.

AB Split Testing Results

That’s pretty good.

Now, imagine the impact if you did this kind of test for every video you are executing.

Let’s crunch some numbers

Let’s make up some numbers here to see what kind of impact this kind of small change could have.

Assumptions:
# of active videos: 100
# of average impressions per video per week: 2000
# of total weekly impressions = 200,000 (100 x 2000)

If we take our 15.5% CTR lift or incremental 1.3 clicks per 100 impressions and apply the assumptions above, the impact would be 2,600 incremental clicks per week (200,000/100 x 1.3) or 135,200 clicks per year across our entire campaign.

All from something as simple, easy and inexpensive as testing your video thumbnails.

The final word

A 15.5% gain is pretty solid, so I took a look at the aggregate data across our platform to see what other Vidyard customers were experiencing with their split tests results.

What I found was that, on average, A/B thumbnail testing resulted in a CTR improvement of 9.5%. So, our 15.5% improvement was at the high end but an average of 9.5% for the effort involved is still pretty sweet.

So, there you go. Our first swing at video marketing “small ball” is a hit!

Frankly, if you’re running several videos and can see this kind of impact across your entire campaign, it’s a pretty big hit.

I’d love to hear what others are trying and learning as they navigate their way through the video marketing landscape. Feel free to share them in the comments below.

In the meantime, stay tuned for more video marketing small ball learnings in the weeks ahead.

Team Vidyard