Today's guest post comes from James Ellis, Digital Strategist at closerlook, inc. He blogs at digital-pharma.tumblr.com and pretty much lives on Twitter (@digital_pharma) if you'd like to reach out.In the last blog post, I showed you that your website only had three parts and that unless you optimized how each part worked, you were throwing money away. Now, you’re going to learn how to improve each step of your website. These are data-driven improvements you can make at any stage increase the overall bottom line, but you need to embrace the data and commit to making changes based on the data.
What kind of changes? That depends on the stage. Let's talk about your landing page. The task is to turn the spark of interest that clicked a link into a fire of attention. So are you? Are you making your case clear, or are you distracting the user with a dozen buttons and links? Is it clear what the user needs to do next? And by "clear" I mean drop-dead obvious. A button, a big one, with a contrasting color, that jumps off the screen? That's obvious. A button directly tied to a reason to continue? That’s essential. Don't just say "Next" or “Submit” when you can say "Learn more about a 15% increase in efficacy."
At stage two, do you know what questions your targets have as they come in? Not what you want them to be, but have you surveyed your targets to learn the misconceptions and uncover what they really want to know? If so, are you providing them the answers to those questions? Is it clear that these are the answers they want?
In the last stage, when they are ready to take the next step (however you have designated it), are you making it easy for them? Or are you the register operator who feels the need to ask the ten questions before you take the customer's money? You've done your job and gotten your target enthused and excited, now close the deal!
Finally, you'll need to apply a funnel metric to the entire process. Here are the numbers, all easily obtainable from Google Analytics that will tell you exactly how well each stage of your web site is doing:
- Percentage of landing page visitors click the call to action
- Percentage of them read more than a single page before leaving
- Number of pages they visit before they convert
- The last page they visit before they register (Hint: make that the next page they see after more of your landing pages)
- Percentage of people who click to convert complete the process
Now you’ve laid the groundwork. The real work is looking at that data honestly every day and asking, "How do I move that number up?" Don't try and increase all the numbers at once. Focus on one at a time. Remember, increasing any one number increases the overall flow of valuable traffic, so pick your battles.
Measurement without action is worthlessHow do you increase each number? Since you’re smart, you planned a few alternatives in the original design process. At some point, your agency had a long conversation about whether the "click to learn more" button on the landing page was "right." You pay them to make the best choice possible, but their choice is still a guess without the data to back it up. Take their top two or three options and throw in a couple of crazy alternate versions and have them ready to test. You think the button is too small a thing to test? Ha! Amazon tested the way their registration buttons worked and increased sales $300 million the first year. One test showed that the color of your buttons can increase click-throughs as much as 21% by itself.
Test something new each week. The first week, test different button colors and compare the results (the results on your dashboard will be obvious). Declare a winner, stick to it, and pick the next test. Maybe you wonder what the button text should be, or if changing the order of the fields in your registration form changes things. Repeat. This optimization process is vital to increasing your overall conversion rate. Each increase trickles down to the next stage, widening the funnel at its most important point: the bottom.
At the end of six months, you will see an increase of your conversion rate. Let's be conservative and assume it's a 10% increase. A 10% increase in conversions means that every $50 click is worth 10% more. It means that you could trim 10% off the email/ad budget and get the same results.
But if you are diligent about testing something new every week, and committing to the winner every time (even when the winner isn't something you expected it to be), you should expect something closer to a 25-50% conversion increase, at which point you will write this up in a memo, send it to your boss and your boss's boss and start picking out your new corner office.
None of this is difficult. In fact, it's pretty easy. Plan for these tests. Create the assets you want to test ahead of time (to save money). Partner with your agency to select even more elements worth testing. Put the numbers on a dashboard (Google Analytics will email you the numbers every day if you want) and begin.
You can make those $50 clicks look more and more like $30 and $20 clicks before your next performance review.