Often­times it’s pretty easy to look at a land­ing page, whether it’s a paid search land­ing page or an email land­ing page, and say, “Yep, I think we ought to test out a few alter­nate designs – no harm in that, and we’ll increase con­ver­sion.” How­ever, many times there is much more heated debate around test­ing entirely dif­fer­ent con­ver­sion funnels.

Often this reluc­tance to test entire fun­nels stems from the fact that IT will have to be much more closely involved since the tech­nol­ogy may touch an order man­age­ment sys­tem. A cam­paign land­ing page usu­ally does enjoy the lux­ury of not hav­ing to inter­face with those sys­tems or require IT involve­ment – marketing’s front-end design­ers can usu­ally exe­cute a land­ing page test.

Let’s look at an exam­ple, and how to tackle it very eas­ily. A large finan­cial ser­vices com­pany had long wanted to decrease the aban­don­ment of its online credit card appli­ca­tion form. It was sev­eral pages long, with a series of ques­tions and form fields, almost all of which were nec­es­sary in order to process the appli­ca­tion. The response would be that the indi­vid­ual was approved, not approved, or that a deci­sion was pend­ing fur­ther research.

One of the company’s devel­op­ers had explored build­ing an AJAX-based form that would reduce the process to a sin­gle page that built as infor­ma­tion was sub­mit­ted. How­ever, debates con­tin­ued to abound within the com­pany as to whether a multi-page form would con­vert bet­ter or the single-page form. The answer? Test it. But how?

Assum­ing you have a devel­oper to build the new con­ver­sion fun­nel for you, the process to test it is very sim­ple: Build the pro­posed new appli­ca­tion form, and at the begin­ning of the exist­ing fun­nel cre­ate a straight-forward redi­rect test. Send half of your cus­tomers to the first page of the multi-page form, and half to the 1-page AJAX form. Upon sub­mit­ting the form, all cus­tomers get sent to the same ‘thank you’ page where you record the con­ver­sion. Now you can view the com­par­a­tive per­for­mance of the two fun­nels against each other quite eas­ily, just like a sim­ple a/b test of a land­ing page.

But what if your credit card appli­ca­tion fun­nel (or shop­ping cart check­out, or lead cap­ture form) ends up show­ing no sig­nal between such rad­i­cally dif­fer­ent expe­ri­ences? What if the results appear to be the same, even long after sta­tis­ti­cal sig­nif­i­cance has been reached?

That’s when you per­form cus­tomer seg­men­ta­tion analy­sis on the results. You may find that cus­tomers with fast con­nec­tion speeds love the 1-page AJAX form just as much as your cus­tomers with slow con­nec­tion speeds hate it. At that point, rather than choose one expe­ri­ence and push it out to every­one you should max­i­mize your aggre­gate con­ver­sion rate by set­ting up a tar­get­ing rule where you give the 1-page form to any­one on fast con­nec­tion speeds and the multi-page form to any­one on a slow con­nec­tion speed. This way you don’t just get a so-so con­ver­sion rate over­all. You get the best pos­si­ble con­ver­sion rate from both groups.
Good gen­eral rule to fol­low: Instead of deliv­er­ing the best-performing expe­ri­ence to every­one, deliver the best-performing expe­ri­ence for every­one. We’ll talk more about tar­get­ing another day…