Continuing our series on the benefits of closed loop marketing; let’s next delve in to the use of Analytics, in conjunction with multivariate testing platforms, to increase conversion rates, for the traffic you are already capturing.
It’s first important to understand what it means when Google Analytics says you had 39,685 unique visitors to your site in the past 30 days.
That does not mean you had 39,000 potential customers visiting your site. The following table segments traffic, based on the convertibility of traffic.
Segment D: There will always be a percentage of the traffic coming to your site that aren’t actual people. For instance, spiders crawling your site—like the GoogleBot—will count as a visitor. No worries, we like Google and all their cute little bots. You should too.
Segment C: This is the natural, irrelevant traffic, that comes to every site. It’s the programmer looking for a bit of JAVA code, who mistakenly visits a site selling gourmet coffee.
Segment B: This traffic might show some measure of engagement, but are still highly unlikely to convert.
Segment A: This is your target audience—if they have a positive user-experience, they will convert (make a purchase, submit a form, etc.).
So, if your conversion rate is 3%, it doesn’t mean you are missing out on 97% of your potential sales. It would be a more realistic estimation to say you are missing out on converting somewhere around 22% of your traffic.
But even with a more realistic picture in mind, if you can use multivariate testing to convert an additional 2% of your overall traffic, that 67% lift in conversion rate—from 3% to 5%—would mean a 67% increase in your overall revenue. Of course that assumes the AOA is the same, etc.
Conversion optimization is the online version of training your sales team to know how to “close the deal.” The good thing is, it’s not about clever language or pressure tactics, but offering up the user-experience that your visitors like best.
This is done through performing A/B or multivariate tests:
1. Identifying A Potential Test
Let’s say you have a page with a clear goal: getting a visitor to add an item to their cart, submit a form, etc.; but only 1.5% of the visitors to that page are completing the goal. Why? No need to speculate—let’s test it.
2. Identifying Testable Elements
What if we change the main image? What if we edit the copy? Should the Add to Cart button be red, green, mauve? Are people distracted by the content at the bottom of the page? These kinds of natural questions are perfect for a multivariate test.
So let’s say you identify four elements you want to test, and you have three alternative options for each. If you include the current elements on the page, that is over 200 potential variations of the current page!
3. Multivariate Testing Tools
An MVT will be able to automatically segment the traffic coming to the page, serving up different versions to visitors. The great thing about these tools is that it isn’t necessary to equally serve up all 200+ versions. As the system serves up versions, it can quickly rule out low performing options and hone in on the most effective page combinations.
In the following example, the system has determined the top performing alternate versions—reporting the lift in conversion, as well as the statistical confidence in that lift.
In this example, version 3143 resulted in a 289% lift in conversion; but the company decided to go with version 2131, which generated a 163% lift.
Why would they pass on the top performing version? Because there are always brand and voice considerations that have to be taken into consideration. But by first testing the variations, they know exactly what that choice will cost them in short term profits, to preserve the longterm brand considerations.
This data-driven, closed-loop, marketing approach is the present and future of online marketing. It’s only going to be a matter of time before more and more companies begin to understand and allocate budget to these cutting edge methods.
But since a lot of companies haven’t figured this out yet, it creates a situation where forward-thinking companies who invest in testing their way to offering the best user-experience, will make big strides ahead of their competitors.
Got a problem? Let us solve it: Contact Us