[Guest Post] How to Run High-Impact Split Tests on Your Video Landing Pages

Editor’s Note: We’re excited to share this guest post from Vidalytics founder Patrick Stiles. After 7 years spent using video to sell millions of dollars of his own products and services, Patrick created Vidalytics as a comprehensive platform to help video marketers improve their conversions. Today he’ll share what he’s learned about video split testing, including how to use Vidalytics with Leadpages for best results.

If you have a landing page that’s getting traffic, you need to be split testing.

Especially if you have a tool like Leadpages, which makes it really easy.

Think about it: you can double your traffic, which is HARD WORK… or you can double your conversion rate.

My personal record is a 278% lift by changing one thing on my landing page. I’ll tell you what it was in just a moment.

But when I started selling online and split testing, I wanted to test EVERYTHING.

I’d try tweaking buttons, menus, colors, you name it. I’d even test things like adding a 1-800 number, live chat, exit-intent popups, or timed buy buttons.

Some of these were winners. But none of them had much of an effect. At best, they’d give me a 5% bump in conversions.

These were all low-impact split tests.

Over the years, I wised up and realized I should focus on high-impact split tests. Split tests that had the potential to improve my campaign results and level up my marketing across the board.

So, what can you test to get a high impact winner?


If you’re looking for a high-impact split test… I vote for video every time. A simple video change was responsible for that 278% lift I mentioned above.

Video is a dynamic, engaging sales medium that has the potential to convert better than practically anything else you can put on your landing pages.

So, how do you know what to test in your videos?

For most of the last 7 years, even I didn’t have a good answer to this question.

I realized I had all of this data for my entire website and business from tools like Google Analytics.

But when it came to my videos—the content responsible for driving most of my conversions—I had very little insight. Let alone the data I needed to guide my split tests.

So, my team and I created Vidalytics to give businesses the most advanced video marketing analytics available.

It’s a perfect complement to the split tests you run in Leadpages, too. Here are some of the ways it enables you to run high impact split tests.

How to pinpoint the messages that make sales

The most important thing you can know about your marketing is what specific message makes your prospect pull out their credit card and buy from you.

Video is uniquely qualified to answer this for you, by using a systematic approach I call message pinpointing.

This approach is part art and science. It starts with getting the right metrics to successfully identify the point in your video that makes your prospects buy.

This is where Vidalytics can you find that winning message. Think of it as a very comprehensive heatmap for your videos.

In Vidalytics you can see what point your audience is watching to, and the exact spot where they buy.

You can also see if certain spots in your video are boring or extra engaging by tracking skips and rewinds. Check out the data from a recent video we launched:

In this case, since 70% of conversions happen after 10:30, we know that those first 10 and a half minutes were critical in getting these people to buy.

Now, this data contains EVERYBODY who watched your video. To drill down further, you can remove the noise from the data by looking at JUST the people who converted.

Segment your buyers to see how the people who ended up converting engaged with the video.

Again, we see that 10:30 mark as a major milestone in this video getting conversions.
In this case, 80% of people who bought made it to this spot.

And of the buyers who watched past this point, most stopped the video at random points before buying.

The only other point in the video that has such a dramatic jump in conversions is the very end.

You can also apply segments, such as new or repeat viewers, to see what different groups skipped or rewound.

That said, we didn’t see clear patterns for those actions with this example… Probably because we removed the controls within Vidalytics that allow people to skip around. 😉

But say a significant number of your buyers had skipped a certain part—for example, the second half of your testimonials section. You can safely conclude this section was boring, chop it out, and get people to your call to action sooner.

Now… the real goal of all this analysis is to find the message that makes your prospects convert.

What speaks most directly to your prospects’ intimate fears, hopes, dreams, or desires; or the most painful symptom of the problem you are trying to solve for them?

Let’s look at what we found in the video above.

The first critical 10-plus minutes happened to contain the personal story of how the product creator struggled with anxiety and then discovered a natural anti-anxiety supplement. (That’s the product the video was selling.)

When you find this type of message, you pepper it throughout all of your marketing materials. It becomes the foundation of your Unique Selling Proposition (USP). You put it in headlines, email subject lines, abandoned-cart emails, and anywhere else that your prospect encounters your brand on their way to convert.

Which is exactly what this company did, like in this PPC ad:

Which, yes, worked like magic! And helped drive $40,000 in sales with this video its first month.

All this sounds very good… but what do you do if your video isn’t converting at all? In that case, you can’t pinpoint the right message because it isn’t there yet.

Over the years of losing lots of money on videos that didn’t convert, I discovered something very interesting about all the eventual losers.

It’s the 25×2 Rule, and you can use it to cut your losers and find a winning video much faster.

Here is how it works:

If a video fails to get 25% engagement at the 2-minute mark… then it will very likely fail to convert.

You can spend your hard-earned cash to get enough data to be sure, and maybe you’ll find the outlier where this rule doesn’t apply, but in my experience of selling a few million dollars online, all with video, I have never seen this happen. And I’ve seen dozens, if not hundreds, of video tests.

So, when my money is on the line, I use the 25×2 Rule.

If your video fails the 25×2 test, the biggest problem is the lead, which is typically the first one to five minutes.

So, you don’t have to worry about the rest of the video. It could be totally fine, but you’re losing your audience early.

Just like in this video:

This is actually an earlier version of the video from our message-pinpointing example.

It was far from profitable the first time out. But after running through this exercise several times, we got the results above ($40,000 dollars in sales the very first month).

What you do when your video fails the 25×2 test is simply rework the first few minutes as a new lead. Splice it in and leave everything else the same.

And BAM! You have a solid, potentially high-impact split test to carry out.

The simplest way to run high-impact video split tests

Here’s where Leadpages comes in.

Most of your videos will be more effective if they’re on a good landing page, so hopefully you’re running traffic to a great page that sets your video up for success.

The nice thing about a video split test is that it’s incredibly easy to set up. Once you’ve created an updated version of your video, just find your original video landing page in your Leadpage listing and select Create Split Test.

Then, choose Copy Control (since all you’re changing is the video). Edit your new variation by dropping in the new video embed code, and you’re off to the races.

After your successful split test, you may know that one video worked better than another… but you may not know why. Or how it affected the rest of the video performance.

So before you celebrate, take a moment to analyze. The perfect tool for this job is the Vidalytics Compare View.

This was designed for plugging into a split testing system just like Leadpages. After a test you can see the two videos’ graphs side by side.

As you mouse over one video, the corresponding spot on the other one is highlighted. This way, you know how your changes in each test affected the rest of your video. And you can find not just which video is better, but which parts of which video are better.

Because while you may be happy if your split test simply produces a higher conversion rate, the real benefit of testing comes from learning what works and what doesn’t in your market.

These are the lessons that will separate you from your competitors and give you a business that survives and thrives in the long run.

From what I’ve seen, video split tests give you the greatest potential for huge gains in sales. And once you know a few shortcuts like the ones I’ve outlined above, you can systematically optimize your videos and achieve those big results.

Any questions about running a high-impact split test? Ask them in the comments!