Conversion Immersion – Making the Most of Existing Traffic
Carrie Hill is the SEO team leader and also specializes in Post Click Marketing and Online Brand and Reputation Management for Blizzard Internet Marketing. She is also a member of the Research and Development team at Blizzard and is responsible for keeping their products and clients on the cutting edge of technology and innovation in website promotion. As a regular columnist for SearchEngineWatch she helps small business owners gain insight into the complex world of online marketing. She is a frequent presenter for the Search Engine Strategies conference series where she has presented or moderated panels on Post Click Marketing, PPC Research Strategies, Online Reputation Management, and Small Business Search Engine Marketing.
The times, they are a-changing, people. The internet is becoming an increasingly competitive market and the share you’ve captured for your keywords is a hot (and finite) commodity. You can assume that the search volume for any particular keyword is what it is and is what it will be for the foreseeable future. Barring temporary phenomena such as a viral campaign that might skyrocket queries – and hopefully traffic – the available volume isn’t going to increase anytime soon – at least not sustainably. The economy is not in very good shape right now. Although search doesn’t seem to be suffering much at the moment, buying more traffic may not be an option as it becomes more and more expensive to list in high-quality directories or PPC engines. The best way to increase your revenue is to take better advantage of the traffic you’re already receiving!
You’ve been very diligent with your website – you’ve optimized your pages, written killer content, and are effectively driving traffic to pages designed to sell your products or services - but sales are still not where you’d like them (or need them) to be. So what’s wrong? You’ve followed all of the SEM & SEO rules and the sales should be beating down your door, right? Perhaps you’ve overlooked something (don’t worry, everyone does it at some point in time). You forgot to factor in human nature.
Or more specifically, you forgot to optimize your website for conversions.
Traffic is really just a portion of the battle. Your job isn’t done once the user hits your landing page. It’s your job to help them decide YOUR landing page offers them exactly what they’re shopping for – and that it’ll be easiest to spend their dollars (or sign up for that information they need) with you.
Conversion optimization takes place offline all the time. When you walk into the grocery store you have a list because you already know what you need. The store offers a map of sorts - signs that help you find the right aisle to check that next item off your list. How do they further structure their offerings to steer you where they want you to go? Let’s say a loaf of bread is on your checklist. The bread is all in one aisle – so finding the loaf you want should be easy, right? What do you find when you get there? Look closely at that shelf. The most expensive bread is at eye level and above. You need to bend over to grab the cheap stuff from the bottom shelf. Basically, the grocery store has effectively optimized their conversion rate by making it easier to buy the expensive stuff. Are you doing the same with your website?
At SES San Jose this year, Tim Ash talked about testing and tuning your landing pages. The idea that struck me most during his talk – and has stayed with me since - is the idea that you can use your visitors to design your conversion paths. Just because Aunt Bess thinks the website you designed is perfect doesn’t mean your visitors feel the same way. When was the last time Aunt Bess put some money in your pocket?
An integral part of increasing your conversion rate is testing. You’ve got to try a variety of options before your website visitors will tell you which one they like best. This doesn’t mean tearing a page apart and testing 10 things at a time. I’m not a big fan of multivariate testing. I prefer a slower more methodical test contained in the context of A/B or Split testing. Tom Leung of Google Website Optimizer advocated A/B over Multivariate testing at SES San Jose this year.
“I’d agree about the power of A/B testing. At the end of the day, people get the best results from very small tests. Small tests make you focus. Multivariate tests can make you lose your focus.”
Remember going to the eye doctor and staring though the contraption at the letter panel on the wall? They changed lenses testing each change and asked you a series of questions as they did:
A or B
1 or 2
They offered up a change with of two choices – varying the new choices based upon the previous decision. Eventually, they dial into the perfect combination to provide you with 20/20 vision. You can utilize the same methodology with your website. Start with a control page and a test page configured to reflect ONE change to that control page. Split your two pages between your traffic, showing 50% the control page, and the rest the test page. Wait a bit and see which version works better. Make the best performer the control for your next test and then change something else. Not every test page will perform better than your control. A/B testing will actually keep you from making big mistakes in a redesign or retooling of your website.
The consensus on what constitutes a valid sample size seems to be 100 conversions per test – but not every site gets a large volume of conversions. So how do you test without spending six-months on each version? Sites with high-dollar items (hotel or vacation rental reservations immediately come to mind) can benefit from A/B testing without having to wait the 3-4 months it takes to reach 100 conversions Not every test needs to be a measured against whether a visit resulted in a sale. You can make “time on page” your testing criteria, or a decrease in bounces, or the “correct” click into the next page in your desired conversion path.
This type of A/B testing isn’t reserved for the math genius in your office. With the advent of Google Website Optimizer, anyone with a small amount of know-how and a little time can set up a test. You may need to invest a little in designer time – or in help getting the codes on your website – but it’s pretty simple and straightforward with the Website Optimizer help section.
If you’re looking to A/B test on a larger scale, consider using an enterprise application designed to handle a high volume of tests along with providing help with analysis and projection based on those tests. Look at LiveBall (www.ioninteractive.com/liveball-platform/) from Ion Interactive or TuningEngine (www.sitetuners.com/tuning-engine.html) from SiteTuners.
In our unstable economic climate, taking advantage of traffic and visitors you’re already getting is a great way to increase revenue without increasing the costs involved in buying more traffic. Improve your conversion rate –then buy more traffic, you’ll be taking better advantage of that traffic and realize a higher return on your investment.
