1. Nick Hill Silver

    Head of e-Commerce at www.proporta.co.uk

    14 January 2005 16:29pm

    Nick Hill

    I manage a product that has rather a complex sign-up process, with two separate registration processes. The first step establishes the users’ site username, password and mobile phone number details; the second step verifies credit card and address details. Probably mainly as a result of this, the failure rate for users successfully signing up for the service and using it is unacceptably high. I wonder if anybody has any knowledge of what an acceptable success/failure rate would be for this kind of dual registration process, and what success/failure rate would be achieved by a single sign-on process? 

    Thanks

    Nick

  2. Alex Chudnovsky

    Fndr at Majestic12.co.uk

    14 January 2005 20:00pm

    Avatar-blank-50x50

    I did a fair amount of work in this area in my previous life. It is not uncommon to have 70-85% of users who started to checkout to fail for one or many reasons. It is important to be able to model all steps of checkout and assign failure rates to specific pages, this way you will know where problems are, and most importantly regular monitoring will help see changes in dynamics that can often be due to mistakes, or on the contrary better design decisions.

    Generally the less steps in registration you have the better, and the less data you ask straight away the less people will leave. Having a single page signon however introduces a degree of uncertainity that results from not knowing the difference between two distinct groups of people: those who clicked and were not even going to try to fill in the form, and those who tried but failed for whatever reason.

    regards,

    Alex

  3. Bryan Eisenberg

    Co-Founder at Eisenberg Holdings

    14 January 2005 21:21pm

    Bryan Eisenberg

    Nick,

    This is a terrific question. I have data that we documented using FireClick’s Index of average shopping cart abandonment rates in the 70-75% range. However, this number in my opinion is completely unacceptable. A recent case study was written up in MarketingSherpa about how my company took CafePress.com who already had an incredible 35% abandonment rate (what’s most incredible is that they have thousands of store with their own look and feel -check ours at www.cafepress.com/futurenowinc- and then take them to a checkout process that changes the look feel), down to only 15% abandonment. If you are looking for some ways to understand how you can reduce your current abandonment rate, you can read my article at ClickZ.com titled "20 Ways to Reduce Your Shopping Cart Abandonment Rate."  Here is a link to part 2 http://clickz.com/experts/crm/traffic/article.php/2248551 which has a link to part 1 in it.

    Best wishes,
    Bryan Eisenberg


    "The Conversion Rate Specialists: Creating Persuasive Architecture™ Online Home Page: http://www.FutureNowinc.com Weekly ClickZ’s ROI Marketing Column: http://www.clickz.com/experts/crm/traffic/
    Co-Author: Persuasive Online Copywriting & The Marketer’s Common Sense Guide to E-Metrics

    On 16:29:07 14 January 2005 Nick Hill wrote:

    I manage a product that has rather a complex sign-up process, with two separate registration processes. The first step establishes the users’ site username, password and mobile phone number details; the second step verifies credit card and address details. Probably mainly as a result of this, the failure rate for users successfully signing up for the service and using it is unacceptably high. I wonder if anybody has any knowledge of what an acceptable success/failure rate would be for this kind of dual registration process, and what success/failure rate would be achieved by a single sign-on process? 

    Thanks

    Nick

  4. Matthew Tod Enterprise

    CEO at Logan Tod & Co.

    16 January 2005 21:09pm

    Matthew Tod

    My company has worked on about dozen projects on process completion in the last 18 months, and I think there are three things we would say in response to your question - but no definitive answer!

    • Firstly look on the bright side and measure completion rate not failure rate! The reason for this is that failure assumes that everybody who starts really intends to finish - and that is almost never true. Have you established the number of people who actually are really interested in completing the process? A survey works quite well in this area.
    • Secondly look at your process as two elements, typically the major part of the drop out occurs on the fist stage, and then between 60% and 85% complete the second stage. If you can break your analysis to cover both elements it should yield clues as to where the issues might be.
    • Thirdly you need to look at the nature of the visitor starting the process. Conversion rates from existing users will be far greater (300% greater in one case) than visitors generated from banner campaigns. You need to take the mix of visitors before setting a completion target - your average figure will be covering up a wide ranges of conversion rates for each audience.

    So I'm afraid there is no simple answer to your question!  Do drop me a line if I can help anymore.

    Matthew

    Matthew TodLogan Tod & Co  Email:  Website: www.logantod.comTelephone: 020 7717 8447Mobile: 07961 042870Skype: matthew.tod

    On 16:29:07 14 January 2005 Nick Hill wrote:

    I manage a product that has rather a complex sign-up process, with two separate registration processes. The first step establishes the users’ site username, password and mobile phone number details; the second step verifies credit card and address details. Probably mainly as a result of this, the failure rate for users successfully signing up for the service and using it is unacceptably high. I wonder if anybody has any knowledge of what an acceptable success/failure rate would be for this kind of dual registration process, and what success/failure rate would be achieved by a single sign-on process? 

    Thanks

    Nick

  5. Steve Jackson

    Chief Analytics Officer at Kwantic Oy

    17 January 2005 11:10am

    Steve Jackson

    Nick,

    Acceptable depends on your offer. We've worked on shopping carts with an optimized process and low cost to the consumer which (when we'd finished with it) abandoned at only 17% average. However you only had to spend €17 and you didn't need a credit card, you could use a debit card system as well and we know this improved take up by 19% on that particular service.

    On the other hand we've also worked on carts that we couldn't improve past 40% abandonment overall. The catalog included higher priced products at over $2000 and i think that people require a lot more time over that kind of decision.

    Of course price isn't the only influence at play. There are a number of articles on our site about ways to improve abandonement ;
    This one about shopping carts,
    http://www.conversionchronicles.com/page.php?PageID=69&tracking=article10ways
    This one about forms in general,
    http://www.conversionchronicles.com/page.php?PageID=77&tracking=article10elements
    And you might want to consider split testing at later stages;
    http://www.conversionchronicles.com/page.php?PageID=47&tracking=articlesplits

    To answer your question in general terms I'd say that an acceptable rate of abandonment of an optimized cart is around 40%. At least that's what we aim for with our customers as a base for continued testing. As I say it depends on the product or service but sometimes it's possible to overtake that. The 10-20% level is exceptional and I believe represents the "best in class" as far as abandonment rates go.

    On 16:29:07 14 January 2005 Nick Hill wrote:
    >I manage a product that has rather a complex sign-up
    >process, with two separate registration processes.
    >The first step establishes the users’ site
    >username, password and mobile phone number details;
    >the second step verifies credit card and address
    >details. Probably mainly as a result of this, the
    >failure rate for users successfully signing up for the
    >service and using it is unacceptably high. I wonder if
    >anybody has any knowledge of what an acceptable
    >success/failure rate would be for this kind of dual
    >registration process, and what success/failure rate would
    >be achieved by a single sign-on process?
    >
    >Thanks
    >
    >Nick

Reply to this thread

Log in to reply to this thread or join Econsultancy for free so you can post to our forums along with other benefits.