I’m frequently asked about the best way to start testing on a corporate Web site. The options are endless, but until you get started and get all the powers that be within your company aligned with the ideas of constant testing, it can be hard to get started.
There’s no right or wrong way to get started, as long as you run tests correctly and that will improve site performance. Let’s start with a few key tips, then get into a few great test types to start with.
- Find a sponsor. Having executive support is key. And it shouldn’t be that hard to find someone who has a bit of an entrepreneur spirit, who is willing to shake things up to improve the business. Your sponsor can help get the horsepower you need to break down some barriers you’ll encounter and help shift the culture.
- Focus on what matters to your business. Spend time and efforts on the things that matter to your business. That could be more qualified leads, more successful checkouts, or something else. Don’t waste your time testing things that don’t drive your business.
- Monetize the impact. I’ve written a lot about monetizing site behaviors; the power of monetization is in forecasting and measuring the impact of potential changes to your site. Estimate what impact your tests could have if they’re successful and use that as your guide of what to test. More important, use monetization to celebrate the tests’ success within your company to get people more excited about optimization.
- Place multiple bets. When testing you don’t want to put all your eggs in one basket. Even the most obvious tests that you’re most confident about may not return the huge impact you thought they would (funnily enough, the ones you’re most sure about are the ones that often don’t pan out). Again, this is why we test. Don’t stake the entire idea of live testing on your site on one test; set up a series of tests with a number of different variables you want to check out and understand. And remember, even when you don’t outperform the control, the lessons from that test are often well worth the effort. In many cases, the lessons of a failed test can be a plus.
- Spend the time. Too often people skim over the importance of setting up a test correctly to deliver statistical significance to get the most accurate data possible out of it. Have a hypothesis and make sure the measurement methodology is correctly defined. A test set up incorrectly can look like a failure when actually it was a success. Spend the time!
Where to Start Testing
While we want to focus on what drives your business, there are a lot of ways of doing so. Testing of landing pages from campaigns, partners, and paid search can be a great place to start testing and can return a high ROI (define). Often this traffic is just dumped on a generic landing page, product page, or — worst of all — the home page. Take a look at that traffic’s bounce rate. You spend all sorts of money on that traffic, then watch it land, turn around, and immediately leave. Test different versions of landing pages that tie specifically to what they clicked on to get them to the site. Try testing different messages based on the creative, type of site they came from, time of day, and day of week. Even doing a few small things to make the page more relevant to visitors can have a big impact.
In my next column, we will dig into another great place to start testing: tuning the message based on whether the user has been to your site before.
Now is the time to test: we’re into fall and the holiday shopping season is just around the corner. If your business focuses on consumers, the fourth quarter most likely makes the year for you. Even if you are business-to-business, you know things slow down greatly after mid-November; you want to finish the year strong between now and mid-November. The next few months are key. Get in and make some changes and improve your site. Just a few small changes can have a significant impact.
Be sure to check back to read part two of this series that focuses on tuning your message to first time or returning visitors.
Join us for ClickZ Presents: Online Marketing Summit, September 25 at the Sheraton San Diego.
While ad fraud has become part of every marketer’s vocabulary, attribution fraud—the practice of gaming outdated attribution models to justify self-serving means—has ... read more
When you’re just starting out as a business owner it’s easy to become wrapped up in the seemingly endless number of metrics ... read more
Something I’m asked frequently at conferences and from marketers is what metrics they should be striving for from their social media marketing. ... read more