Videos - autoplay or no autoplay? Test results...

[Keith Gilbert]

Hey Warriors,

From time to time, I've seen it debated here on the WF whether or not you should set videos on your sales page to "autoplay" when the visitor hits the page.

Well, I decided to do a little split test in a non-IM/non-webmaster niche.

I recently did an extensive A/B split test and here are my results. Version A and Version B were identical except that in version B, the video autoplayed.

Version A (no autoplay): 93 sales
Version B (autoplay): 100 sales

Pretty interesting, but I'm not saying here that you should always set your videos to autoplay. I'm not on all my sites. Rather, I'm saying that you should always split test! (Even though I'm the worst person about it!)

Hope this is useful to someone.

Keith

[yesacpow]

You are very correct. It can also be very amazing to see the big difference a small word can have on your entire sales page.

Thanks for the tip

[queenbuzzy]

That doesn't surprise me...especially if you get the first few seconds to be really engaging. I've just started calling the internet video presentations: "webfomericals" because that's what they really are.

And man, do we ever get drawn into them. Just like their tv counterparts, they're hard to get away from, and if they're compelling enough, people will get sucked in and buy!

7% more, apparently, from your test!

Congrats!

[DrewG]

Step one of AIDA-

Attention --- what better way to grab the visitor's attention with an attention grabbing video?

Nice case study, thx for the share-

~D.C.

[artwebster]

This is not a significant statistical variation (3.6%) so would tend to show that those of us of who don't like autoply will buy anyway if the product is what we want and those who do like autoplay will, also.

What these results aslo tend to show is that the number of people NOT buying from each type of video site is the same IF the visitor capture rate is the same.

HOWEVER, if version A got 93 sales out of 200 visitors and version B got 100 sales from 500 visitors, you will have a significant variation on which you can base your future strategy.

[SimonHarrison]

On a side note, if your using YouTube videos on autoplay embedded somewhere on your site, every time somebody hits that site your automatically getting a free view on that video bumping it in popularity on Youtube so you kill two birds with one stone. You don't get the same advantage when not on autoplay.

[sajeena]

Thank you Keith for posting your split test result.
Simon, Your side note is very helpful. I plan to upload my salespage videos to youtube.

Best Regards,
Sajeena.

[Jenie0109]

i'll prefer to disable autoplay. Sometimes autoplay affects the speed of our internet connection specially if our internet connection isnt that fast

[Josh Anderson]

i'll prefer to disable auto play. Sometimes auto play affects the speed of our internet connection specially if our internet connection isn't that fast

It does not effect the speed of your internet connection. It's actually the other way around. How fast your internet connection is effects how fast files load.

1. Just because you turn auto play off does not mean the file does not load. Many video players do not have an "auto load" control and start loading the minute you land on the page whether it auto plays or not. Always choose a player that allows you to control auto load.

2. If you optimize your video by encoding it to a bit rate that is less than the average speed of your internet visitor it will not cause page load problems.

This is important for auto play because you want it to play near instantly. I now set most of my videos to play in just one second because they are optimized at 252kbps. Also if there are multiple videos on a page I make sure that only one is set to auto load.

The key is to encode your video at a bit rate to target fast delivery at the bit rate of your target market.

[adam123]

Originally Posted by artwebster

This is not a significant statistical variation (3.6%) so would tend to show that those of us of who don't like autoply will buy anyway if the product is what we want and those who do like autoplay will, also.

What these results aslo tend to show is that the number of people NOT buying from each type of video site is the same IF the visitor capture rate is the same.

HOWEVER, if version A got 93 sales out of 200 visitors and version B got 100 sales from 500 visitors, you will have a significant variation on which you can base your future strategy.

The core of A/B split test is driving the same amount of traffic to offer A and offer B. So the visitors number should be the same

[Mark Collard]

I concur with Josh's advice re making sure you're comparing apples with apples - make sure you're getting the same level of traffic for each test, otherwise, it could be hard to compare. And only have one video per page set to auto load (otw pages load much slower).

Also, I discovered similar results when I added (just) audio on auto play (sorry don't have the clinical figures to quote right here), but it was the only thing that changed on my site, and sales experienced a boost.