Back when Obama was running for president, here’s what happened on the campaign’s website in December of 2007:
Visitors were coming to the campaign website — but they weren’t signing up.
The campaign realized this problem, and the director of analytics Dan Siroker decided to run an experiment. He tried 24 variations on two variables: featured image/video (six options: three images, three videos) and call- to-action button text (four options: “Sign Up,” “Sign Up Now,” “Join Us Now,” and “Learn More”). 310,382 people visited the page during the testing period, which means about 13,000 visitors saw each combination.
The sign-up rate on the stock page was 8.26%. By simply trying different combinations, they found a variation that worked significantly better — an black- and-white photo of a family coupled with a button that said “Learn More” had a signup rate of 11.6%. That is a 40.6% improvement. Two tweaks. Same visitors. Same cause. Documented on Optimizely, the testing company Siroker would later co-found:
“We conservatively estimate that those extra signups resulted in 2,880,000 new email addresses and $60 million dollars in additional donations.”
I want you to focus on one detail of that result: ALL SIX VIDEO OPTIONS PERFORMED WORSE THAN ALL THREE IMAGE OPTIONS. Also consider this: “Learn More” — the least strong, least committal button text — outsold every other option, including “Sign Up Now,” by 18.6%. They assumed certain choices would perform better than others. Instead of hoping, they ran a test and found out which actually worked.
What Your Website Is (Actually) Doing
Your website is not a brochure.
People do not land on your website to receive information. They land expecting to learn about something and leave without anything if your page isn’t clear. A brochure is a document that provides information and asks nothing in return. Your website is different, because every page is a conversion machine, designed to transform a visitor’s mental state from curious/skeptical/unfamiliar to interested/trusting/ready-to-do-something.
Everything on that page is either moving them forward or holding them back.
Consider: when a visitor first arrives at your page, they will quickly (and subconciously) ask themselves four questions:
1. What is this?
2. Is it for me?
3. Can I trust it?
4. What should I do next?
If your page doesn’t answer those questions quickly — above the fold, in plain language, before they have to scroll or search — they leave.
Did you know that 53% of mobile visitors will abandon your page if it takes longer than three seconds to load? Google Consumer Insights. Speed is the first test your page must pass. Here’s the second: as soon as they arrive, visitors are scanning your page for answers to those four questions. If your language is confusing, jargon-y, or otherwise opaque, they leave.
Here are the Four Things that Actually Drive Conversion:
Number one is a headline that answers “what is this” and “is it for me” clearly in one sentence. Not a catchy slogan; an actual statement.
“Tools for freelance designers to send invoices in under a minute”
Converts better than:
Work smarter, not harder.
Be specific. Sell your one great thing. Put that in your headline.
Number two is page speed. We covered this already, but Google Consumer Insights also found that over half of mobile visitors will leave a page within three seconds of it loading poorly. Another study from Portent showed pages that load in one second have a conversion rate of roughly 39% - at two and a half seconds, that conversion rate falls to less than 2%. Every second matters.
Number three is one clear call to action above the fold. Remember the Obama example above? “Learn More” outperformed “Sign Up” because it seemed less daunting as a first step. When you decide what action you want visitors to take on your website, make that the smallest reasonable step. Not the biggest, scary one you hope they’ll eventually take.
Number four? Trust.
Your visitor is deciding whether they trust your page before they do anything. Believe it or not, there are people who will browse your entire website and leave without taking any action simply because they don’t trust it. Elements that increase trust — usage of real names, real images, a physical address if applicable, privacy policy, and prominent social proof (think reviews, testimonials, logos of recognizable partners) — are not optional parts of your website. They are answers to your visitor’s implicit question, “should I trust this?”
You’re Already Running The Test
Remember how the Obama campaign tested 24 combinations?
Most business owners test zero variations. They pick a design, publish it on their website, and wonder why they’re not getting conversions. But your website is being tested. Right now. By every visitor who lands on it.
Replace your homepage.
Because your visitors are the test. Everything they do — where they move their cursor, how far they scroll, where they ultimately exit your page — is data. When you look at a webpage and see something that “feels like it would convert,” you’re seeing the result of hundreds of visitors downloading that information in a matter of milliseconds. The websites that convert and the ones that don’t are not defined by their design. THEY ARE IDENTICAL. Great pages and great conversions are built with clarity.
Two tweaks to one landing page earned the Obama campaign $60 million dollars.
There are that specific. The results are that measurable.
Treat your website like the experiment it already is.
