So funnel testing is analyzing all the way through the funnel, from acquisition in the search engine optimization end to conversion. So it’s SEO testing also CRO testing together. I’m going to write a little bit more about some of the inspiration for this. However, in brief, it essentially boils down to the fact it is perfectly possible, in fact we’ve seen from the wild cases of tests that win in SEO terms and lose in CRO terms or vice versa.
Testing for only SEO or only CRO isn’t always perfect. Some changes lead to higher conversions and decreased site traffic, for example, while others might rank more highly but convert well. In the current Whiteboard Fridaywe welcome Will Critchlow as he shows a technique of testing for the top-of-funnel SEO adjustments along with your conversion-focused CRO changes at the same time.
In other words, tests that perhaps you make a change and it moves better, but you shed organic search traffic. If you’re only testing one, which can be common — I mean many organizations are just testing the conversion speed side of matters — it is perfectly possible to have a winning test, roll it out, and also do worse.
Click on the whiteboard image above to open a high profile version in a new tab!
Then you can compare the conversion rate among the group who obtained the control and the group who got the version. That’s very straightforward. Like I sayit sometimes happens on a single page or within an whole site. SEO testing, a little bit broader. For the purposes of the consideration, there’s basically just one Googlebot.
So let us step back a tiny bit. Conversion rate optimization testing works in an A/B split kind of manner. It’s possible to test on a single page, if you’d like to, or even a website segment. The way it works is that you split your viewers. So that your audience is broken. Some of your viewers gets one version of the webpage, and the remainder of the audience receives a different edition.
Over at Distilled, we’ve been investing pretty significantly in building out our capability to do SEO evaluations and specifically built our optimization shipping network, which has let us do a new kind of SEO testing that has not been previously accessible to most of our clientele. Lately we’ve been focusing on a new enhancement to this, which is full funnel testing, and that is what I want to talk about today.
Hi, everyone. Welcome to another Whiteboard Friday. If you’ve been following what I have been writing and speaking about around the internet lately, today’s subject may not surprise you that much. I’m going to be talking about another kind of SEO testing.
Therefore the way that people do an search engine optimization test is we really divide the webpages. To do it, you want a substantial website segment. So This was one of our silly examples we came up with on a few of our demonstration sites. The consumer lands on this page here. What we do is we then set a cookie. This is a cookie cutter. This user afterward, as they browse around the website, wherever they go within this site section, they get exactly the exact same remedy, either the control or the variant. They get the same treatment across the entire site segment. This is much more like the conversion speed test .
Users get whatever their very first page impression resembles. They then get that remedy applied across the whole site section. So what we can do then is we could evaluate independently the performance in search, evaluate that on entries. Thus do we get significantly more entrances to the version pages than we’d
Assess SEO evaluation on entrances
Googlebot = stateless requests
What I didn’t reveal in this diagram would be if you’re conducting this evaluation across a site section, you would cookie this user and make certain that they always watched the exact same treatment no matter where they navigated around the website. So because Googlebot is creating stateless requests, in other words just independent, one-off asks for each of these of these pages with no cookie collection, Google sees the divide.
We have seen this with, by way of instance, removing content from a page. Sometimes having a cleaner, easier page can convert much better. But maybe that was where the keywords were and maybe that has been helping the page rank. So we are trying to avoid these sorts of situations.
Some pages are unchanged. Some pages get the hypothesis applied to them. As far as Google is concerned, that is the conclusion of the story, because on any person petition to these pages that is what we serve back. But the seriously important thing is I’ve got my small character. This is a human browser performs a search,”Exactly what do badgers eat?”
That’s where complete funnel testing comes in. So I want to simply run through the way you conduct a complete funnel test. Everything you do is you first of all put up it in the exact same way as an search engine optimization evaluation, because we are basically starting with SEO in the peak of the funnel. So it is set up exactly the same way.
Then, crucially, search engines and users see exactly the same experience. There’s no cloaking going on. There’s no duplication of articles. You simply alter some pages and not alter others. You then apply kind of advanced mathematical, statistical investigation seeking to work out perform these pages get statistically more organic search traffic than we think they’d have done if we had not made this change. So that is how an search engine optimization evaluation works.
Imagine, for instance, an e-commerce site with thousands of goods. You might have a notion of something which can help those product pages perform better. You take your hypothesis and you simply apply it to some of the webpages, and you also leave a few of the pages unchanged as a control.
Full funnel testing
We believe this will only be the more data-driven we get, the more answerable search engine optimization testing makes usthe more important it is likely to be to combine these dots and make sure that we’re getting accurate uplifts on a net basis when we combine them. So I hope that’s been useful to some of you. Thanks for joining me this week’s Whiteboard Friday. I am Will Critchlow out of Distilled.