I have reviewed website statistics with clients on numerous occasions. I wish I could say presenting those stats has always been helpful, but after reading The Lean Startup by Eric Ries, I’m not sure I can make that case.
One of the things that draws Eric’s ire is the concept of Vanity Metrics. Loosely defined, they are metrics that make you or your business look good, but in reality don’t do anything to help you get closer to figuring out what your business really needs to be doing to grow.
One example of a vanity metric we have used a lot of in the past is Revenue per Website Visitor. While that metric may be interesting and it may make Strategy6 look intelligent, I’m not sure it brought any of my clients any closer to retirement, so to speak. A slightly better use would be to chart the revenue per customer over time, but even then the author states that unless we are using the metric to validate or refute a specific hypothesis, then it still isn’t doing any good.
One way that Strategy6 might use the Revenue per Website Visitor metric going forward will be as follows:
- Begin by telling the client we hypothesize that if we show potential customers “add-on” options during the checkout process we can increase our revenue per website visitor.
- Then split the web traffic into two equal groups and show half the add-ons and the other half no add-ons.
- Finally, look at the Revenue per Website Visitor for the control group and the test group and see if the test group confirms the hypothesis by producing a higher Revenue per Visitor
Split tests are nothing new to SEO or to Strategy6 but after reading Eric’s book, we plan on doing them as much as possible in 2012. Going forward, we will only present metrics that either confirm or refute a previously stated hypothesis. The best part is that they will all be “actionable,” that is they will reveal to my client the correct action they should take going forward. Anything else, according to Eric Ries, is a waste of time to compile and present.