Conversion rates

Monday, June 12, 2006

Over at Rogue Amoeba, Paul noted some interesting statistics with regards to Airfoil's conversion rates. That is, given a time span, what percentage of those downloading their product end up purchasing it?

I've been a stats hound since starting Zengobi. A few months ago I even started going crazy with Google Analytics to churn some more numbers. Yeah, I need a life but darnit if this stuff isn't fascinating!

Our conversion rate for Curio for the month of May was 4.93%. Our average for this year is 5.2%, with a best month of 8.77% and a worst month of 2.75%.

Two quick points to make:
  1. We don't know how many downloads are made by people who already own Curio (Group A) vs. people who are trialing a slightly older Curio (Group B) vs. people who are downloading Curio for the very first time (Group C). They all can't be Group B and C trialers either since some percentage of Group A are certainly updating themselves when a new update is available (like from 3.0 to 3.0.1). That means our number of "downloads to potential customers" is certainly a smaller number than our overall downloads number. So those conversion percentages above a probably a bit higher. Fortunately, with our last release of Curio, I've added some anonymous Google Analytics tags to the "An Update is Available" alert so, when we come out with our next update, we can further drill into the stats since we can see how many registered and trial users are coming in via that mechanism.
  2. You can download Curio, anonymously, and we kick in a 15-day trial. We also support a 60-day extended trial if you give us your name, title, and email address. So, technically, giving a month timeframe we don't know how many downloads-to-sales were made within the same month. That's why the average spread across a number of months is perhaps more useful.
I'm happy that our number is around the same as Paul's and many others responding to his posting. I had a worry that everyone else out there had a double-digit conversion rate. I'd love to know what Delicious Monster and OmniGroup are doing just to see what we can expect as traffic grows.

Kudos to Paul and everyone else for discussing this openly. Also, I'm happy that Linda from OmniGroup opened up a discussion on marketing with this blog entry. As a small Mac startup, I'm glad the developer community is generally pretty open about their inner workings.