UIEtips: The $300 Million Button

Jared Spool

January 14th, 2009

In today’s UIEtips, I tell a story about a client who found a way to dramatically increase their e-commerce site’s revenues with a couple of simple changes. While the story is interesting, the story-behind-the-story is just as interesting.

The client had hired us because they were concerned about checkout-process abandonment. Their analytics were showing a 13% drop off in sales, which, based on the average value of the abandoned shopping carts, was worth about $1.2 million a year in additional revenue.

Checkout-process abandonment is common in e-commerce sites and something that you can easily detect with your site’s usage logs. You just look at the number of people who get to the first screen and then the number of people who actually complete the transaction. Everyone who doesn’t make it is an abandonment.

When the team contacted us, they’d already pretty much decided what the problem was and how they were going to fix it, even though they had never watched any shoppers make purchases. And they were dead wrong. Not only was their fix not going to help, our research showed that it was going to increase abandonment.

Two weeks of usability testing on the live site (and on competitors’ sites), followed by two weeks of iterative paper prototype testing produced a streamlined checkout process, which, once implemented, showed a dramatic increase in revenues. It’s amazing what you’ll learn when you actually watch your users.

Today’s article, The $300 Million Button, talks about the bulk of that increase — how a simple change to a common screen produced $300,000,000 of additional revenue over the next year. I’m sure you’ll find it interesting.

Improving forms, like a checkout process, can show immediate results in your design’s user experience. We’re fortunate that at this year’s UIE Web App Summit, we have Luke Wroblewski repeating last year’s top-rated Web Application Form Design full-day seminar. If your site has forms (and what site doesn’t these days), this is a must-take course!

Have you seen results from changes to your forms? We’d love to hear your experiences. Share them with us below.

32 Responses to “UIEtips: The $300 Million Button”

  1. Terry Fitzgerald Says:

    I tried the “Don’t need an account”process within one of our local book sellers sites. What was interesting was that in order to complete the purchase all of the same information required to register was asked for e.g. mailing address and payment information. I don’t see how this simplifies the purchase effort. I do understand about the email/id + password issue. I have a file I keep of all my various account IDs etc. to avoid this problem. I have also started to trim down the number of different account IDs I keep. Perhaps the example site was different but I’m not sure how you can order stuff on-line without providing this information.

  2. Robert Joyner Says:

    One question did you get paid a fixed rate or a percentage???

  3. Andy Bright Says:

    When I read this article while reading Luke’s book I just had to say ‘wow’.

    This is a great success story to use while trying to persuade those clients that don’t believe user testing can really deliver ROI. It illustrates so clearly that user research doesn’t just offer gains but stem what are essentially losses.

  4. Catherine Williams Says:

    I am not at all surprised at these findings. I myself have often abandoned carts because of being forced to register, for all the reasons you outline.

  5. Josh Holmes Says:

    Jared – that’s another great story. I can’t thank you enough for continuing to share these great stories. This is the type of story that, as I say on my write-up of your article on my blog (http://www.joshholmes.com/2009/01/14/JaredSpoolReconvincesMeThatUXMatters.aspx), seperates the graphic designers from the usability guys.

  6. Daniel Szuc Says:

    Just experienced a similar story in booking flights online to the USA.

    Its nice when a site/business helps by registering/remembering key information for you without having to re-type it in but gets out of the way when in the “buying flow” Its annoying though when the site will not accept key information like international credit cards at the end of the whole “choosing and entering personal information” stage 🙁

    Question — In your experience, do you ever notice that people feel more rushed or they perceive the need to get to the end faster when buying stuff online? (so any additional step in that process to make them think more increases their anger-o-meter?)

  7. Malcolm Bastien Says:

    The $300 Million Button was a great article. If anything it really helps to prove UIE to be a true value add to the bottom line of organizations.

    On part about UIE that I usually find more often than not than a lot of the time, most big problems are easily fixed by using Best Practices, and that those Best Practices make a big difference whenever implemented because most implementations as-is are usually very bad!

  8. Carlos Celi Says:

    Yet again another good practical and very usable example from the folks at UIE! Thanks Jared!

    Making the “Registration” step not required I’m sure helped unregistered users wanting to move ahead with their purchase. But what about registered users? was Login also made optional at this point? (I’m thinking Amazon, where it is required early in the checkout process) If not, was there anything done to alleviate the 160,000 passwords requests(and subsequent 75% non purchases) that this step generated? Just curious!

    Thanks again, great article!

  9. Lee Morey Says:

    They say one picture is worth a thousand words, but apparently the right word can be worth $300 million? 🙂 And people wonder why we UX/UI/IA/ID folks are so persnickety about finding the right labels, eliminating unnecessary steps, and trying out realistic tasks with potential users…

    Your story resonates strongly with me because I have so often worked on improving government web sites and forms which require a lot of information. However, (false) assumptions that a so-called “captive audience” has to do what you want — so “usability isn’t all that important” — certainly isn’t limited to one arena!

    Back in the early part of the century, I started to shop at a web site which required me to create an account simply in order to put an item into their shopping cart. I badly wanted one particular product: a trademarked T-shirt design which I knew was only available from that brand. So I dutifully — if reluctantly — registered.

    However, after shopping around for a little while longer because I had gone to the trouble to register, which was very probably what they wanted to achieve, I suddenly realized that my anger about their ill-considered process had trumped my desire to own the item I originally wanted. What I would associate with wearing it would be the irritation I was feeling right then, the online equivalent of encountering a salesperson who turns you off so much that you leave a store and never return. So, I did exactly that… no doubt the opposite of what they had wanted to achieve by forcing folks to register!

  10. Paul | OptimalInfluence.com Says:

    I’ve actually spent $10 more with an online bookstore that DIDN’T require registration, because all the others I found required me to sign up.

    My own personal reason is that I don’t want my credit card details stored with ANY online store, no matter how trustworthy.

    I’ve often presumed that online stores like Amazon have tested, and found that requiring registration increases sales (presumably repeat business?) – but maybe they didn’t test after all, and just assumed…

    Thanks for the great article!

    Paul Hancox

  11. Jonathan Smiley Says:

    Great article, and an awesome example of designers identifying a problem that at first glance doesn’t even seem like one. There are so many retail sites out there for which registration is just QED – how else could a retail site work? Well, turns out it could work like this, and work better.

    Hopefully this article doesn’t just lead to rethinking retail registration problems, but removing roadblocks for users across the board.

  12. Pete Williams Says:

    Thanks for a very enlightening article Jared.

    Interestingly though there seemed to be no solution offered to the problem of multiple registrations (45% is unbelievable!). So after some thought, I have proposed one, of sorts, on my blog : http://petewilliams.info/blog/?p=7

    I’d love to hear people’s thoughts on my suggestion.


  13. AJK Says:

    I posted this article in one of our UX groups and the following critique was given on the article regarding the numbers and I was wondering if Jared could respond.

    1) At 25$ billion annually there isn’t many retailers according to the fortune 500 list. Even if you go from 20 to 30 billion.
    2) The company isn’t a pure e-commerce retailer, Amazon is leading the pack there and they are not even doing $25 billion annually.
    3) If it is not a e-commerce retailer it is likely that their e-commerce portion is a small portion of the $25 billion due to historical reasons. (point 4 validates this)
    4) The article says that purchases increased by 45%, which suggests that they had a turnover of about $30 million a month online prior to the change.
    5) A 160k pw requests a day? wow that is impressive. That means they are having probably over 250k to 400k visitors a day and make just about 1milion a day. (doesn’t prove anything but that this retailer
    has a headache, lots of feet and a very crap conversion ratio, they make around 4US per visitor)

  14. Elise van Looij Says:

    Interesting article, and interesting comments, especially from AJK. One passage that bothered me a bit was this one: They took away the Register button. In its place, they put a Continue button with a simple message: “You do not need to create an account to make purchases on our site. Simply click Continue to proceed to checkout. To make your future purchases even faster, you can create an account during checkout.”

    I read over it several times and I still wonder what the purpose of this screen is: the customer clicks on the button ‘checkout’ and gets a screen with a ‘continue’ button that they need to click in order to actually checkout, as they originally requested. Uhm, since those designers (and developers, I presume, since the workflow needed to be reworked) are busy anyway …

  15. A.J. Pape Says:

    Jared –

    Very interesting parallels with the training projects I do in large corporations.

    Ideally we learn as much about the people who will be going through the training as possible before we design it, but some clients see that as a low-return investment of time and money.

    Happily I’ve gotten better at being able to show the value of it, but at times it’s still a tricky balance – giving them enough of what they believe they need now, to help them refine their sense of those needs and what will work as the project unfolds.

    This is my first trip to your writing. Thanks, I enjoyed it.

  16. Lynne Says:

    Nice article! I found this to be very true when I logged on to Home Depot to purchase a dishwasher. not only was I not required to register, but there was a chat box that popped up with help offered if I needed it. In that case, the assistant was WONDERFUL, and helped me to make the right purchase. At the end of checkout, I was able to decide if I wanted to register or not, and they give you a choice of email notifications for sales, deals, etc. I found it better to think about registering after I had completed the purchase.
    Now, if someone could figure out why ebay is so crammed with more and more roadblocks, that will be an achievement. That site is getting worse with each passing day.

  17. Joanna Says:

    This is another great example of copy as UI. The *messaging* is improved here — nothing has changed about the actual design or interface. Gooooo copywriters! Be sure to hug your writers… and read more by Erika Hall here: http://www.slideshare.net/mulegirl/copy-as-interface

  18. Jim Griesemer Says:

    Thank you, Jared, and Luke! It’s another win for the 100 year mission!

  19. Patrizia Bordignon Says:

    I encountered a similar problem when evaluating an online selling website.

    For users to determine the cost of advertising their wares to sell, they had to register first. Many users abandoned the process at this stage. We moved the Register button to the end of the costing process, at which point users had made a significant effort in creating their ad and they understood the value proposition. Then they were happy to establish a relationship.

    The result was a 25% increase in transactions in the first month.

  20. Bob Says:

    Great article!

    This was my experience exactly, trying to purchase concert tickets on line with Ticketmaster.
    The registration process (required) fouled up when it realized I already had an account. (of course I didn’t remember my password) Same process. Add to it the fact that you are under a time limit, 2 or 3 minutes before it cancels and you have to start all over again. PLUS my credit card number had recently changed and so was different than the one they had on file, which meant I had to delete all information in my account and re-enter… and that’s when the time ran out. I gave up and called their automated phone line to purchase tickets.
    This was an even worse process which I won’t go into now. I went back on line, took a deep breath and started again. In all it took over 25 minutes to purchase two tickets. This was today. I will avoid Ticketmaster whenever I can.

  21. web development Says:

    very interesting. i would never have guessed that could cause such a big difference.

    I guess google checkout helps as users dont have to register anyway.

  22. Sean Bowman Says:

    Just out of curiosity, you didn’t mention the effect on user registrations. Did they go up or down? Or did the “Continue” button just register everybody automatically?

  23. Jay Godse Says:

    Although this change has been billed as a simple UI change, it is a radical change in the business process. The old process says that a “relationship” (by registering) is a pre-requisite for purchasing products, while the new process says that a “relationship” is a way to optionally speed up the purchasing process. This difference is huge. (The old process is also an example of premature optimization).

    Technologically speaking, authentication & authorization should precede an online purchase, but a credit card is sufficient authentication and authorization, and registration is not needed.

  24. The 4 Essential Steps to Successful Online Marketing —FishMarket Online Marketing FishMarket Says:

    […] Here is a rather sensational example of what happened when Amazon.com fixed one part of their websit… […]

  25. william leech Says:

    When I get to the register bit they lose my sale..
    I am a customer and they are selling something.
    I want to choose pay and leave …simple isn’t it?
    If I click to see an item,then I want details of that item only….don’t waste my time, if you do you lose a sale

  26. Jared Spool re-convinces me that UX Matters | Josh Holmes Says:

    […] with usability. This point is very clearly demonstrated in his latest article on his blog – UIEtips: The $300 Million Button » UIE Brain Sparks. In this article, he talks about a fairly straight forward ecommerce site that was loosing millions […]

  27. Avinash Says:

    Just wondering in the check out phase users would require to fill their Credit Card or Bank Details. How is the security measure handled here to check frauds don’t happen!

    It is surely a great article but I have been having this discussion after this and people are concerned about the payment part where they ask what if a person uses a stolen card.

    Awaiting for your solution 🙂

  28. Jared Spool Says:

    Hi Avinash,

    Having an account doesn’t prevent check or credit card fraud. There are other security measures in place to ensure that.

    Keep in mind that to receive the product, the customer still had to enter their name, address, shipping, and billing information. All of that the business uses to prevent fraud.

    The only real difference here is the customer didn’t have to set up an account (username & password) for returning visits.


  29. Quora Says:

    How is Design ROI measured for the purpose of justifying the value of a design?…

    It would be difficult to define generic metrics for the ROI of design, in most cases it will need to be done on a project by project basis. First, on a related topic, I find the Cennydd Bowles “User Experience Validation stack” very helpful when it c…

  30. Proving the value of user experience design to organizations and clients Says:

    […] The $300 Million Button, where a simple design change resulted in an enormous revenue lift. Spool writes: When the team contacted us, they’d already pretty much decided what the problem was and how they […]

  31. d3vkit Says:

    I was directed here from UI Stackexchange – the website I manage actually uses a non-register, non-login method which I felt could use the extra steps – apparently I really am more programmer than designer 😛 Whenever given the option of continuing without registering or registering, I do. I don’t usually find it a barrier, but after reading this, it is clear how much of a barrier it can be. Can’t thank you enough. I will be changing the order process a bit but this has convinced me to look at less “traditional” methods (likely very similar to the one presented here).

  32. d3vkit Says:

    I do means “I register”. A good example of why registering helps – I can edit what I post 😛
    Oh and that error page – that throw to the simple, white screen, “please enter the spam check” – that is NOT good design. Even I know that!

Add a Comment