Recommended reports
Checkout Optimization Guide: 70 ways to increase conversion rates
Problems with the checkout process are still the biggest single source of revenue loss for most e-commerce businesses. This report identifies 70 awesome recommendations and includes lots of practical insight to help you reduce your checkout abandonment rates.
Alternatively, see our E-commerce Platforms Buyer's Guide or Selling Online: a How-to Guide for Small Businesses.
Head of Ecommerce at Lovehoney
26 October 2009 14:47pm
Hi there. I'm a stickler about conversion rates, and one thing that consistently bugs me is that our IE6 visitors have a lower conversion rate then their IE7/IE8/FF brethren. We use all sorts of prgressive enhancement to make sure the "base experience" for IE6 users is usable and intuitve, and yes, like many of my peers we do have a lot of bugchasing needed to make sure this stays, but still, conversion rate for IE6 users is generally half that of the average. I break this down into 3 factors:
1) The computer they use is likely to be slower
2) They're more likely to have a slower internet speed
3) They're less likely to "trust" the internet, and therefore, less likely to shop online.
However, I could do with some benchmark figures to compare against. Does anyone else have a similar problem? One thing that is special about www.wiltshirefarmfoods.com is that our average site visitor is 73, and as such, there are a lot of trust issues and apprehension our visitors have shopping online.
Matt
Managing Director at indiumonline
28 October 2009 13:08pm
Hi Matt,
Really interesting post!
Looking at a few out clients, we're seeing a similar pattern. Always a lower conversion rate for IE6, often substantially so. I'm afraid I can't name names (client confidentiality), but I looked at a range of eCommerce sites, with varying demographics.
I think your factors form good hypotheses, have you looked at the following:
- browser & operating system breakdown - would help explore some of those hypotheses
- browser and internet speed breakdown?
- Does the AOV also differ? Is it better or worse for those customers?
- what are the other site stats like for them? Do they exit at a different page? Do they view more or less pages? What's there bounce rate like?
These latter questions may highlight some areas for testing to see if you can improve it - eg if the bounce rate is much higher for them than others on the homepage then maybe you need to reduce the weight at bit, or you might need to drop the flash image and replace it with a gif.
Of course, if these are the older customers - they may be researching online, but buying over the phone. Harder to extrapolate but there's plenty you can do to encourage those customers to still place their order:
- put the phone number more prominently on the site - in the header maybe?
- also make the quick order more obvious too
I'd be really interested to hear where it takes you. Or if you want to pick my brains a bit I'd be happy to help, contact details below.
Chloe
Indium Web Management
www.indiumwebmanagement.co.uk
01865 339 470
chloe at indiumwm.com
Head of Ecommerce at Lovehoney
01 November 2009 20:06pm
Hey Chloe, I think I'm going to have to do a fair bit of datamining on this. I think your suggestion about internet speed is a winner. We have some progressive enhancements, whilst they "did" work in IE6, in my mind were far too technodisco for this audience.
Becuase they're jquery based, it's not the easiest to do multivariate testing on them ( well, there "is" a way, but it involves fudging User Defined Variables in Google Analytics - oh for an Omniture icense! ) - so this morning I commented them out for the day.
And promptly, WinXP IE6 conversion rate comes only a couple of % point below site conversion rate.
Obviously, this needs to be tested. Keynote (who we've used in the past) can't test jquery elements, so I think guerilla testing over a 3G signal might be in order!
Do you like Quick Order? It's quite successful, so obviously, I'm thinking of getting rid of it....
Managing Director at indiumonline
02 November 2009 17:26pm
Hi Matt,
if you're mailing a lot of catalogues then quick order's great, it let's the customer order fast without getting side tracked. And often they're coming to your site with a shopping list - so you want them through as fast as possible.
BUT of course you want dwell time, and them to see all the other great things on the site.
There's no easy solution I'm afraid. You've just got to be aware that you have 2 types of customer - those who use the quick order, and those who don't know it even exists.
For those who love quick order, you need to keep them using it (I'm guessing you're conversion rate from that page is very strong - if not you need to have a look into it). And use other marketing methods to let them know about other things that may be of interest. Can you do a "add to your order" after it's been placed? then you could add some extra marketing to the order confirmation page/emails. OR you could market to them about other services via the email / rss / twitter (whatever their preferred mode is) or even via the catalogue...
It's a balance every company with a catalogue needs to find, website as marketing tool vs website as order taking channel. Afterall, it's probably cheaper for you to have customers place their order on the website, than via the phone or by post - so there's a cost saving to having a quick order, as well as a good conversion rate.
hope that helps
Chloe
Head of Ecommerce at Lovehoney
04 November 2009 12:30pm
That does help a lot, thanks Chloe. QO has conversion rates of around 97%, but yes, there is only 1 reason why you'd visit it :-) - we send out around 1M catalogues a year
Very good idea on the follow-up "add to your order" emails. We don't offer a post-order change facility yet, but it's something I want to do within the next 6 months, thanks for the inspiration!
Matt
franchise holder at Les Bons Voisins
15 November 2009 13:55pm
Matthew I agree with your points about older users having slower pcs, connection speeds etc. but I'm less sure about the underlying failure of conversion.
Isn't it possible that they simply browse for reference and then prefer to complete a catalogure form for the order. That might , for example, be down to any online payment preferences or fears they have, which in older customers may be of high importance.
Could you try something on your catlaogue forms as a tick box that asked whether people had used your serbsite when preparing the order? Even if it didn't give you data on what browser they were using it would help to close the gap on hits with no apparent purchase.