In my previous post, I've talked about A/B testing and in a future post, I'll cover what multi variate testing is. This post is an interim between the two; last week my wife had our second child, so blogging time is a little hard to come by at the moment!
In this brief post I'd like to list a few things that MVT is not. There seem to be various ideas about it, most of them described as a panacea for all online woes.
It isn't having more than two versions of a test image on a page; that's just A/B/C/n testing.
It isn't really about simultaneously optimising different parts of a page, either. In its purest form, MVT is about measuring and studying how changes to multiple areas of a page affect conversion, including the interactions between the parts that are changing. It's the collective sum of all the parts of the page that contribute; optimising each individual component may lead to reduced performance for the page as a whole. Taking these interactions into account, for me, is the difference between MVT and just running multiple A/B tests on one page. I'll cover this in more detail in my 'proper' post next time.
MVT isn't, by itself, the cure-all for a poor customer experience either. Setting up test versions of pages on a website won't provide long term help to a website, in the same way as a quick blast of keyword optimisation won't fix a poor Google ranking. MVT is a long-term process, and it's prone, as all computer-related activities are, to the Garbage In Garbage Out problem. If you don't think about the testing, and develop a proper testing program, then you won't learn anything or improve anything for yourself or for your site visitors.
Apologies that this post is so short, and brief; think of it as a trailer or a primer for my next post!
In this brief post I'd like to list a few things that MVT is not. There seem to be various ideas about it, most of them described as a panacea for all online woes.
It isn't having more than two versions of a test image on a page; that's just A/B/C/n testing.
It isn't really about simultaneously optimising different parts of a page, either. In its purest form, MVT is about measuring and studying how changes to multiple areas of a page affect conversion, including the interactions between the parts that are changing. It's the collective sum of all the parts of the page that contribute; optimising each individual component may lead to reduced performance for the page as a whole. Taking these interactions into account, for me, is the difference between MVT and just running multiple A/B tests on one page. I'll cover this in more detail in my 'proper' post next time.
MVT isn't, by itself, the cure-all for a poor customer experience either. Setting up test versions of pages on a website won't provide long term help to a website, in the same way as a quick blast of keyword optimisation won't fix a poor Google ranking. MVT is a long-term process, and it's prone, as all computer-related activities are, to the Garbage In Garbage Out problem. If you don't think about the testing, and develop a proper testing program, then you won't learn anything or improve anything for yourself or for your site visitors.
Apologies that this post is so short, and brief; think of it as a trailer or a primer for my next post!
Get your Free SEO report in 3 steps. We provide free and instant online reports of
ReplyDeletewebpage rankings in top 5 Search engines, Google, Bing, Yahoo.
Search Engine Optimization is the practice of improving the visibility (traffic) of
a web site, or page, in search engines organically.
website conversion optimisation