Comments on: True conversion – the on-base percentage of web analytics?

By: Betty

Betty — Wed, 19 Oct 2011 06:48:42 +0000

wow, I hadn’t gone that deep but it makes a lot of sense to try to have your own metrics that work, well, for our site too. Just not convinced there’s the right keyed data that will help us identify th intention to buy. THinking cap time though. THanks for the inspiration

By: Wanna be better with metrics? Watch more poker and less baseball. | Retail: Shaken Not Stirred

Thu, 07 Jan 2010 01:02:36 +0000

[...] True conversion – the on-base percentage of web analytics [...]

By: Kevin Ertell

Kevin Ertell — Mon, 14 Sep 2009 23:41:19 +0000

Thanks for your comment, Derek. I realized after reading your notes that I didn’t explain the calculation as clearly as I could have. I’ve made a slight adjustment to hopefully make it more clear. You did interpret it correctly, though, and I completely agree that it’s important to optimize at every stage.

By: Derek Monteverdi

Derek Monteverdi — Fri, 11 Sep 2009 22:41:43 +0000

I’m not sure how to read the calculation but if you measured carts or cart adds per product view then measured checkout starts vs. completions then you’re right on target. Overall conversion is an important metric but New Cart Conversion, Add to Cart Conversion, Checkout Conversion are the additional tools that you need to understand progress through the funnel. The goal should be to optimize at every stage.

By: Ted Vasquez

Ted Vasquez — Tue, 04 Aug 2009 20:55:46 +0000

Kevin, thanks for connecting one of my favorite books to our work. If I may continue the analogy, all batters MUST touch first base before they can achieve the ultimate goal of scoring a run. So, on-base percentage is really a way of measuring how often a batter completes his most basic job, reaching first base. After that, he must rely on his teammates to ultimately score a run.
In this sense, Billy Bean broke down a complex and interdependent goal (scoring a run) into manageable components for measuring the performance of individual batters (reaching first base).
I submit that you achieved something similar with your measure of “true conversion”. You broke down a complex and interdependent goal (completing a sale) into manageable components for measuring the performance of specific page types (e.g., product pages are supposed to get people to add products to the cart).
I agree sites are more than a buy button. They have more than one goal. The challenge is defining those other goals and then breaking them down into measurable components.
***NOTE TO ALL THE BASEBALL FANS OUT THERE: A more accurate definition of a batters job is to reach first base and/or drive in runs. This includes driving in runs by sacrificing or even making an out. By that definition, on-base percentage is not a complete performance measure but still better than batting average.