Amazon cross-selling

This is clever. Well I think so anyway. Amazon are cross-selling their new Mastercard credit card via the medium of your actual shopping basket. They give you a £10 credit if you apply for the card but take this one step further by applying this credit in a hypothetic, yet obvious way to your current check-out total. Automatically your eyes are drawn to the discounted price of your purchase. Nothing seems unusual as we're now finely accustomed to the concept of discount codes et al. Personally I think this is fairly slick marketing example.

Online polls

I'm always looking for tools and services that compliment the core optimisation activity we do using MVT and AB testing. In the past I've tried getting the general public to rate a selection of test variants using an on-line survey (see earlier post here for details). To me testing is always going to be the ultimate customer feedback on what works for on-line sales conversion but there's no harm in garnering opinion of your winning web design and page content.  I'm now piloting on-line polling on a few landing pages using a Polldaddy  poll widget. This simply embeds a rating widget in the top corner of selected pages enabling visitors to either give the page a thumbs up or thumbs down . Obviously the shortcoming here is that you cant tell if people are giving the page design or page content a good or bad rating, but at this stage I am trying to get a very high level user response to the page as a whole. Ideally I would have this widget on every page on the site as a means of flagging up any under performing pages or highlighting pages that people have an affinity with.

Use of Awards endorsement in web pages

In the past we've included any awards the company or product have attained based upon the assumption that they can only be a positive thing to have on the web page. Well I finally got around to measuring exactly what effect the use of an award logo had when added to a page. Basically I had concluded an earlier optimisation exercise on a landing page using Maxymiser and tweaked the winning page combination to show an award logo instead of the image of two people. I let the adapted test run for 3 weeks and the winning combination fell from a 9.96% uplift to minus 4.05% thus showing that the use of award imagery in this context was ineffective. Click on the image below to enlarge for a summary of what happened.

Is Fatwire any good?

Looking for information on Fatwire and MVT testing? 


 If you didn't know Fatwire is a content management system...




We have been doing MVT and AB testing on a Fatwire (FW) website for the last 10 months. FW dictates that your website is constructed through a series of templates which dont lend themselves to conventional mvt tagging; well not easily anyway.

We decided upon using Maxymiser as a testing solution primarily because their one touch solution could be worked into the header of the websites master template, thereby enabling us to conduct tests across site with zero (any I really mean zero) IT department input from then on. Putting Maxymiser (or any third party testing tool)  tagging in the header means it needs to be in a site-wide 'asset' of FW, thus enabling you to conduct MVT or AB testing anywhere within your website. Doing this allowed Maxymiser to then render out alternative/test content on top of your 'restrictive' Fatwire content.

If you want to be consistent with your Fatwire templates, you really need to come up with test variants based on the content restrictions of your current, available templates. This is a high level choice to make. Personally because Maxymiser could render any designs I liked over the top of the current suite of Fatwire templates we had, I decided to go with page designs and content that didn't bear much,if any,  resemblance to the Fatwire content beneath. This was because I wanted to see what gains could be made by testing alternative layouts and use any uplifts as leverage with our IT department in getting new templates built, becauseI could prove the benefits of doing so.  Additionally while your running an MVT test over the top of the Fatwire you are effectively 'making hay while the sun shines'. By that I mean that even while your running a test, nine times out of ten, because your layouts are not restricted by the Fatwire template business rules, you're already going to benefit in customer conversion during the test period alone which you wouldn't have got otherwise. I will talk about using your testing tool as a CMS in a later post. It's a bit controversial : )

In all honesty if I had the choice I would not be working with Fatwire at all, especially not the very old version we have. It's effectively a step backward in terms of web design, it's highly restrictive in terms of design flexibility. The illustration I've taken to recently using if that a one man band building an e-commerce website in his bedroom using something like Microsoft Front-page, actually has the potential means to add 90% more functionality to his website than someone using a traditional content management system such as FW. Think about that. If you have some JavaScript based web service or dynamic content you want to have a go with using, just by way of a trial, with FW, how much heartache do you have to go through to get that tagging, or JavaScript included into your website body header or body? If it's not a major hassle to you or your IT department then fair play but you're probably not the norm.

Anyway here's my true feelings about good old  FATWIRE laid bear through the medium of satirical animated video.....

Google Optimizer - Landing Page A/B Testing

I launched a split test on one of our highly trafficed Current Account landing pages last week.

This basically saw the rather stale current champion page design challenged by a much more creative led design for the same page. The graphic above shows the readout for the visitor conversion rate which shows the creative led design as the outright winner (16.9% uplift in click to apply rate). However, I have tagged both pages so that we can identify which pages actually result in submitted current account applications, and the results show the original design leading over the creative led design. This is yet another example of how pretty design may compell people to click an apply button but if they havn't read the fine detail they're less likely to go through the entire end to end transactional process.

The original page design (champion) is shown below on the left and the creative led design (challenger) is shown below on the right.