November & December theme:
“Choose One Thing”
Posted in :  Brain Waves

First, a quick confession. I was that kid in science class that always had his hand up asking questions three minutes before the period was over.  You can imagine how popular I was on Friday afternoons at 3pm. Today, as an online marketer, I actually get paid to ask questions. In our website work at Fathom, my colleagues and I throw a lot of ideas at the wall and see what sticks. Some of our ideas are spot on while others fall flat.

Future online marketer at work?

The hard part is knowing which ideas are which. “In this game,” says Avinash Kaushik, irreverent website analytics evangelist and author of Web Analytics: An Hour a Day, “the difference between Winners and Losers is that Winners take the time to figure out which of their ideas stink. Losers don’t.”

Are you satisfied with a 2% conversion rate?

Since the dawn of the Internet, conversion rates have hovered around 2%. This means that 98% of visitors to our websites don’t do what we want them to do. We weren’t able to persuade them to complete a purchase, pick up the phone and call, sign up for a trial account, download an e-book or register for a webinar. I don’t know about you but I’m not satisfied with a measly 2% conversion rate.

There is a better way. By knowing what you want to accomplish, asking questions, posing hypotheses, testing alternatives and never being satisfied you can succeed through testing. The good news? Testing is quick, cheap and produces results.

Testing your Way to Success

Here is the process that you will want to follow:

  1. Formulate a hypothesis
    1. Pose a question
    2. Answer your question with a statement
    3. Include the results you expect to find
    4. Define how you will recognize success
    5. Example: The ‘Request a Demo’ form is under performing because potential submitters are worried about how their information will be used. We will test a statement that assures people that we jealously protect their privacy and link to the  site’s privacy policy in a pop-up window.  We expect to see an increase in form submissions as a result of this test.
  2. Benchmark how you’re currently doing
  3. Identify what you’ll be comparing and contrasting
  4. Perform the test
  5. Measure the results
  6. Document your findings
  7. Implement the winning test results

And you thought 8th grade science class would never come in handy!

Quick and Easy Low-Cost Tools

Now that you’re on-board with testing, here are some great tools to get you started. Visual Website Optimizer lets non-technical people quickly and easily test different versions of webpages and starts at $49 a month. You can also tell the tool what percentage of traffic you want to send to each variation.  So, if you’re testing a really radical idea, you might decide to send only 10% of your traffic to that version.

Another tool, that’s a lot like Visual Website Optimizer but is meant for landing pages, is called Unbounce. Unbounce also has a WYSIWYG environment, is very user friendly and also allows you to allocate traffic to the variations you’re testing. Their website also has a lot of great information regarding landing page optimization. Both solutions let you define goals and will tell you which of your variations are winners and which are losers.

Start Testing Today

Now that you have a quick and easy low-cost solution for testing there might be a temptation to test everything in sight. A syndrome that Bryan Eisenberg, author of Always be Testing, calls “slice and dice testing.” Please resist the temptation. You’ll be able to beat your current results by a lot by if you question whether or not an element you’re thinking of testing might actually prevent someone from taking an action like buying, downloading, registering or signing up for a demo. Remember, always test for impact and go for the BIG wins.

What do you think? Does your organization test? If it doesn’t . . . why not? If it does . . . what kind of results have you seen? Do you have any favorite tools?

Comment with Facebook

Posted by: Steve Machesney
Email the author: