Why everything you know about web-conversion optimization is wrong

Big buttons are better. Creating a sense of urgency is important. Simple and clean and elegant is better than complex and busy. Most of us who build websites have thoughts and ideas about conversion optimization: the art and science of transforming visitors into customers.

Most of our ideas just might, however, be completely wrong.

[aditude-amp id="flyingcarpet" targeting='{"env":"staging","page_type":"article","post_id":889388,"post_type":"story","post_chan":"none","tags":null,"ai":false,"category":"none","all_categories":"big-data,business,dev,entrepreneur,","session":"D"}']

“Best practices don’t work,” Chris Goward of Wider Funnel said today at GROWtalks. “Most of what’s been said out there … people make it up … they have to come up with new content for their blog posts.”

Goward proved his point by showcasing several versions of a popular, 1-million-subscriber wine sales site, WineExpress.com, that he tested. Most of the web-savvy technical audience picked a simple, cleaner version with a larger buy button and larger sales text.

We were wrong.

“As much as you’re a smart marketer, you won’t know what they’ll respond to best,” Goward says. “In this case, visitors were high-net-worth people looking for classy wines. Anything that smacked of marketing turned them off.”

Which means that while there’s some standard dos and don’ts that are almost always valid — rotating homepage messages are distracting, leading with video hurts conversion — each unique site and each unique set of target customers has enough variation that only one good overall solution exists: testing everything.

And that’s more than something you do occasionally or episodically, according to Goward.

“Testing is not just a tactic … it’s a strategy,” he said. “It can inform the entire business … you can use the results to get insightful findings about your target audience.”

Goward has used that strategy to help Silicon Valley startups like Expensify realize a staggering 47.2 percent improvement in conversion of homepage visitors to signed-up users, accelerating the company’s trajectory to a million users. Using his insights document management company Iron Mount achieved four times its previous rate of lead generation. Google is a customer as well.

[aditude-amp id="medium1" targeting='{"env":"staging","page_type":"article","post_id":889388,"post_type":"story","post_chan":"none","tags":null,"ai":false,"category":"none","all_categories":"big-data,business,dev,entrepreneur,","session":"D"}']

The things Goward looks at for conversion optimization include relevance and clarity, which can improve your odds, and anxiety and distraction, which will tend to decrease your odds.

Plus, of course, it has some marketing common sense.

“Most companies talk about their points of parity with competitors, but the most powerful place to be is points of difference,” Goward said. “The word is points of irrelevance: features you’ve worked really hard to create … and no one cares.”

One challenge for startups?

[aditude-amp id="medium2" targeting='{"env":"staging","page_type":"article","post_id":889388,"post_type":"story","post_chan":"none","tags":null,"ai":false,"category":"none","all_categories":"big-data,business,dev,entrepreneur,","session":"D"}']

To get the best data, you need 30 to 50,000 monthly visitors, Goward said, to get reasonable results quickly. However, even small-traffic new companies can test and get valuable data.

“The key when you have low traffic is to test boldly, making different options that are very significant, including your offer,” Goward said. “And then, be comfortable running it for a while.”