This Just Tested: An aesthetic design that produced 189% more leads
So often, beautiful design gets trumped by marketing objectives – and rightly so from a marketing perspective. The graphical elegance of a Web page might be worthy of an art exhibit, but if it doesn’t sell anything but “oohs” and “ahs,” what service does it really provide to anyone?
True, but it is at this point that we tend to divide. It’s them or us. Are you a marketer or a designer? Whose side are you on? As one popular design blog analyzes its relationship with marketing, “You can spell ‘team’ from the word ‘marketing,’ but I’ve yet to see a sense of it in marketing.”
But I think we (marketers) can and should live in both worlds. I believe design can be done in such a way as to actually contribute to the perceived value of an offer without being a distraction. I think marketing, whether they can measure it or not, is leaving money on the table when design is viewed as optional icing on the cake. Yes, I have a dream…
But, feelings aside, we must always default to testing – not our gut instincts. And so I was glad to see a recent experiment bring a little shimmer of hope to those of us who long for the day when these two often opposing worlds come together.
The Original Page (click to enlarge)
The original page was for an Australian company that offers dedicated hosting solutions. The primary goal of this page was to get the visitor to request a quote. These quotes were essentially leads that were then followed-up with and nurtured.
Nathan Thompson (who happens to be making his debut appearance on today’s Web clinic) was the key researcher for this page. Upon analysis, he identified the following three conversion threats:
1) There is significant friction due to the volume of questions (many unnecessary) posed in the form.
2) There is not enough value being communicated with headlines, images, copy and CTA.
3) The overall low aesthetic quality of the design may be creating anxiety as to the actual capabilities of this company.
Of these issues, number three stood out as the most interesting hypothesis and I was eager to see the treatments he would put together to address this issue of design-induced anxiety.
The Optimized Page (click to enlarge)
The page that Nathan and the team created:
1) Significantly reduced the form fields required on the first step from 20 to four.
2) Removed distracting banner images and strengthened the visibility of the headlines to better communicate the value.
3) Included a six-point expandable/interactive display of value proposition copy.
4) Added testimonials and cleaner, more modern images.
5) And then, finally, Nathan and the team focused on creating a more aesthetic design in order to improve the perceived credibility of this company. The thought was, if you are a business looking for a company with professional online capabilities, then a more professional design would generate better response. Let’s now go to the numbers.
Well, as the title reveals, the results of this newly designed page were significant. The new page outperformed the control by 188.46% with a statistical confidence level of 95%.
There are so many things that can be said about this test, and we will be drilling down much deeper on this experiment and others on today’s web clinic, but what is most notable to me right now is that the more graphical appealing page was the clear winner. This is different than what we have seen in past experimentation.
Now, we cannot forget that this is a radical redesign, multi-factorial test. There are many factors here being tested at once, and no doubt the results reflect that. But, with that said, all of these changes did happen within the context of a purposefully and strategically more aesthetic Web design. That’s noteworthy in my book. And for those who often feel the tension between marketing and design, it’s my peace offering for today.
What do you think?
Is this page more aesthetic or am I being too kind? How would you improve it? Did the more aesthetic design actually contribute to conversion or was it the other elements being tested?