70% Increase In Conversion And A Lesson In Human Nature

Although every website defines conversion a little differently, we all need visitors to take some desired action. We need them to advance one step at a time until hopefully they become a customer. Because ours is a service business we need a phone call. And because the particular site I’ll refer to in this post is a national franchise that repairs and installs garage doors, we need our visitors to find the local franchise in order to make a phone call. So, the desired action is simply to enter a zip code into the zip code box. Which was happening at a rate of about 8%.

So we did a redesign, here’s what the 2 pages looked like:

Original Page

Original Page

New Layout

New Layout

The goal was to create a cleaner look that emphasized the desired action by isolating it above the fold. The result using website optimizer for a month showed only the chance of a slight improvement.

Original Vs New Layout

Original Vs New Layout

Then we added a video spokesperson to the new layout.

New Layout With Video Spokesperson Added

New Layout With Video Spokesperson Added

The video contained a spokesperson that highlighted one or two important reasons to choose us and asked visitors to enter their zip code in the box. We pitted this new page with video against the original in 2 separate experiments. In the first, the video was present but did not start automatically, the player needed to be clicked to launch the video.

In the second, the video was launched automatically.

You want to take a guess which was the winning combination?

The combination where the video needed to be activated by the searcher was the only combination that showed significant gain and it was dramatic. The desired action was taken 70.5% more often and Google was very confident that should continue – 99.9%!

Website Optimizer Shows 70.5% Increase

Website Optimizer Shows 70.5% Increase

I learned a long time ago when selling yellow pages that when people get information on their own they trust it more. I always believed that was true of search as well. Now, I’ll add video to that list.

My takeaway from this experiment is that if you want someone to do something you have to tell them to do it— just wait until your asked to tell them. We’ve added this same spokesmodel to some local sites. She’s telling the site’s visitors to go ahead and a make a phone call, but only after that information has been politely requested.

(Full disclosure: At this time I have no affiliation with Live Face On Web. However, based on this experience, we are currently in negotiations to become a reseller of this product. Whether we do or not I do not yet know.)

This entry was posted in Analytics. Bookmark the permalink.

6 Responses to 70% Increase In Conversion And A Lesson In Human Nature

  1. SPK says:

    Data, data, data. I love data.

    Amazing how little differences that seem trivial make such a significant impact.

    You remind me of google doing all the side by side testing. Clearly, we are still learning the psychology of buyers on the web. The good thing is the web is static and we will have it down very soon.

  2. SPK you would love the stuff we’re doing. If that year-end bonus ever gets quite not-so-generous, let’s talk.

  3. kittu says:

    a great idea, it works beautifully for one of my clients!!

  4. Josh says:

    Different niches different percentages.
    It brought me sometimes 40%, sometimes even 77% depend on the niche.
    Bottom line – its working!

  5. I am an old hand at putting up walk-on videos, but pretty new to trying to get a handle on analytics, so this may seem a newbie question. How do you test 2 versions of a site? I mean, what do you actually do to send half of your traffic to one iteration of a site and half to the other? I have a WordPress site I would like to set up that way to test your findings. Terry

  6. So, what your data shows is that the addition of the autoplay video brought the conversion rate up from 8.8% to 9.13%, but if they had to click on the video to see it, it went up to 15.8% of all visitors: put in their zip codes. What do the data show in terms of what percentage actually watched the video in the 2nd variation?
    Is this data holding up in 2013? Are people more accustomed to Autoplay videos(or more burnt out)?