Sign up for our daily recaps of the ever-changing search marketing landscape.
Using Natural Language To Improve Conversion Rates
Languages are powerful things. Emotive, passionate, constructive, descriptive—they are all these things and so much more! This post explores the power of natural language for conversions and first I’m going to look at “Mad Libs” forms.
What is a Mad Libs form?
Mad Libs forms are a style of lead generation/contact/sign-up form which are designed so that the form fields complete natural language sentences. The name springs from Mad Libs, a game where one player has to fill in the blanks. They’re best explained visually and the most famous example of this in the wild is the Huffduffer sign-up page:
While this might look flash, you have to ask, “does it work?” Well, evidence suggests that it does work. Luke Wroblewski, Chief Design Architect at Yahoo, ran some tests on his own mad libs style contact form and saw a 25-40% increase in conversion rate:
In the interests of fairness I should also link to this post from software developer Patrick McKenzie arguing these forms don’t work at all! Personally I don’t really like the implementation but that post emphasizes the most important lesson of all: do your own testing.
Thankfully for those of us who are technical and design n00bs there’s a helpful intro post on how to start creating these things. Visit Rustin Jessen’s guide here.
A real-life example
The inspiration for this post didn’t actually come about because of Mad Libs forms. While they are cool and might quite likely increase your conversion rate, it was actually the following post that inspired me: Boarding Pass Fail. The post is from an extremely talented designer who took it upon himself to redesign boarding passes to be “better”:
But “better” is a difficult concept to nail. The original post inspired many other designers to create their own versions. The following design from graphicology was the one that jumped out at me as being the “best.” For me (and I realize this is a personal preference) this boarding pass gets the information across quickest and clearest:
Doesn’t this look quite a lot like a Mad Libs style approach? Behold the power of natural language.
The real lesson here
The point I’m trying to get across here is not that Mad Libs forms are a magic bullet or that you should get a job designing boarding passes. The real point is that natural language is a powerful tool. Let me tell you a story.
We were working with a client who sells kitchens. Obviously they don’t sell kitchens online—no-one buys a kitchen directly online. Instead, people like to come into the showroom first and browse around, play with the taps, open the drawers etc. So the conversion for the client in this case is filling out a contact form. All across the site there are large images of beautiful kitchens and the call to action to book an appointment.
However, the link to “book an appointment” went to their contact page which had a generic call to action to “contact us.” To improve conversions we tried various things on the page, such as modifying the button, the form fields, the color and size of the header but none of these changes improved conversion rate as much as changing the wording of the header.
The winning page had an altered page header that said “book an appointment” and the form title was also changed to “book an appointment” (both of these were previously “contact us.”
While we have seen some improvements from modifying the button to read “book now” instead of “submit” and changing the col our, by far the biggest improvement in conversion rate was simplifying changing the wording of some text.
So, next time you’re looking at a page and trying to improve conversion rate remember that the message is one of the most important elements of the page. And changing the message can be as simple as changing the wording. Figure out what your users are looking for and give it to them in simple language.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.