Checkout Process Split-Testing Tip from Bryan Eisenberg

Path TestingHow should you approach split-testing your checkout process?

This question was asked of Bryan Eisenberg in yesterday’s Google Website Optimizer webinar. Bryan recommended split-path testing, reducing the number of steps in your process and using your analytics data to determine what part of your checkout path needs attention.

What is Split-Path Testing?

The definition of a split-path test, according to GrokDotCom:

Split-Path Test — This test will split your traffic among different linear paths containing multiple pages for each path. This is different in that you’re testing the performance of grouped pages against other grouped pages. For example, you could test a checkout process by splitting it into two variations; one with four steps (or pages), and another with only three steps. Each variation of grouped pages will have the same Goal Page (e.g., order confirmation page). Once the data is collected, the winning checkout process will be the one that converted a higher percentage of visitors.

Reducing Checkout Steps

Different ecommerce stores have different checkout paths, ranging from one-page AJAX checkouts to 6 steps or more. Bryan believes less is more – in fact, he recommends going under 4 steps. But you can find out for yourself if this is so for your website by doing your own testing.

I’ve gathered some examples of checkout steps (many are generally the same aside from labeling) that can give you some ideas for how to simplify your process. For example, you may want to test a new path with a combined billing and shipping page vs. your existing separate steps. Or you may want to ditch a step that may be clogging your funnel, such as “Rewards Program.”

Checkout 9

Checkout 11

Checkout 3

Checkout 4

Checkout 5

Checkout 1

Checkout 2

Checkout 6

Checkout 15

Checkout 7

Checkout 8

Checkout 10

Checkout 13

Checkout 12

Checkout 14

Alexis Brion discusses checkout usability improvement with shares his own screenshots here if you need more inspiration.

Using Analytics Data

You’ll want to look at analytics to identify a “hole” in your funnel where your customers abandon most often. (If you’re using Google Analytics, click on “Goals” then “Funnel Visualization”).

Funnel Visualization

You’ll want to come up with a test that potentially plugs that hole, rather than a new process for the sake of designing a new process.

It’s a good idea to give your new checkout process pages their own URLs so you can pinpoint abandonment in your analytics more effectively, depending on which analytics solution you are using.

Further Reading

The good folks at GrokDotCom have written 2 follow-up posts to the webinar and this post, definitely worth checking out:

Using Funnel Reports to Improve Conversion Rates by Ron Patiro


Website Optimization and the Price of Perfection by John Quarto vonTivadar


Related Articles

17 Responses to “Checkout Process Split-Testing Tip from Bryan Eisenberg”

  1. Great post! I like the example of using the funnel report to find the points where people are exiting the site. I have seen a lot of clients focus solely on the page where people are exiting to run their tests. However, this is not always the case. Sometimes the problem is a lack of information on a previous page causing people exit at a later page in the funnel.

    When creating tests to plug a page with a high exit rate take the whole funnel into account. Ask yourself if there is any information you could have provided people that would make them more confident and possibly not exit the site at this page.

    Good luck and keep testing!

  2. I agree that less checkout steps is important because this is the formality part of the procedure and users don’t want to spend longer on it than they have to. Also, making the steps clear is vital for usability and it’s amazing how many sites still don’t implement this simple feature.

  3. This is something that came up after we did our most recent redesign to our checkout process. Coming from the developer side it would be cool to have a way to test different pages in a checkout process but then you run into the fact you have two totally different pieces of code that both need css/javascript that do the different things and of course different backend logic for different commands and forms or what not. Most likely you will have to change these commands to get the new way working properly. So now you have to invest in a lot of man hours and a lot of confusion with all these different pieces of code doing the same thing.

    Sure it would be great to have this but I’m not to sure how much other people outside of the development team know that this is a pretty complex job.

  4. What, there isn’t an automatic button in your backend admin to do this sort of thing? :)

    This is a lot more technically difficult to do than most marketing or business people realize – agreed.

  5. That idea of split-path testing is great! It can probably be used in many other areas of testing too. Say you want to test what background color is more appealing to shoppers. Send some customers to one, and other to another, and after some time if you see somewhat of a pattern in pages that converted to sales, you’ve got your answer. It’s really worth a try.

  6. A bit off topic but have you guys used the Google Website Optimizer for any of your sites? I know my boss had me look into it when it was first around and then there really wasn’t much use for it for a complex dynamic shopping site. Except to do some a/b testing on content areas in which we already have that functionality built in our CMS.

    I went back there a month back and saw that it was possible to work on dynamic driven websites. But I just haven’t looked much more into it.

    Also being a perfectionist front end guy, I can’t stand to take content out of my sites and host it on google then slap a Javascript tag in place.

    I have talked to our all things Google consultant and he told me he just doesn’t think it’s useful for a database driven ecommerce sites. but what site isn’t? If you guys have any more thoughts on it I would love to hear them. Of if they went more into that in the webinar.

  7. Hi Dan, yours is an excellent question.

    I’m assuming your question is posed to all our readers and yes, we’d love to have you jump in with your experiences as on other posts.

    Playing with GWO is on my ever-growing to-do list, which is why I want to learn all I can about it through Webinars and such. My eyes glaze over analyzing data and statistics so I am never going to be the A/B testing guru :)

    Perhaps someone from Future Now wants to take the challenge, this would make an excellent blog topic to explain how to perform split tests on dynamic driven sites, with what tools etc.

  8. Dan,
    It’s completely doable to apply GWO on database driven sites and on dynamic sites in general. All GWO cares about is finding the test script on the test page and the goal script on the Conversion page. That’s it, there’s nothing more complex than that.

    As for the objection that you want to hold the content close (and not have it be on the Google server), the answer is that it IS on your site — Google simply makes a copy of it on their server so that when a visitor is earmarked to receive variation X or Y that they get it, while at the same time ensuring that any problems at all in any form that occur during the process will mean that your visitor will see what the original variation was — which is, after all, what you’ve got on your site right now.

    I’ll follow up some more on the issue of perfection on our own site

  9. John,
    Thanks for the info, and I would love to read how it went for you guys. I added you to my reader, you guys look like you have a lot of good information to dig through.

  10. Here’s the post, thanks John for your insights!

    The Price of Perfect Optimization

  11. glad I could give some inspiration :)

  12. I am planning on making a new status bar graphic for my shopping cart process. I presently have 3 steps,

    shopping cart, billing information and order complete.

    Right now a customer can click on previous steps in the status bar and go from the 2 step to the 1st step or from the 1st step to the step, by clicking on the wording in the status bar. The status bar graphic is an image map.

    I have noticed many sites to NOT offer this functionally. When I build this new status bar graphic, should I just make it a regular graphic or build in the image map functionally to click on the different part of the status bar?

    Any opinion?

  13. Whenever something is not offered on most sites, you risk that people won’t get how to use it – especially image navigation. Why not build a prototype and test it on a few people – the least internet savvy people you know ;)

  14. TraiaN says:

    I’am always kind of late with your posts, but if anyone is interested on website optimization and wants to test the shopping cart process, I wrote a guideline for designing a more usable progress bar, here: Testing the Checkout Process – The progress bar design. Enjoy!

  15. Amazing post! Thanks for the visuals of other check out processes.

  16. I recommended it on stumbleupon. The only thing that it’s missing is a bit of color. However thank you for this information.

  17. Michael says:

    So how do setup a split path test in google website optimiser? Any help much appreciated.

Leave a Reply

© 2014 Get Elastic Ecommerce Blog. All rights reserved. Site Admin · Entries RSS · Comments RSS