Fri, 09 Jun 2006

Yesterday, I posted an article detailing some lessons we've learned in our short time as the developers on the Windows platform. Part of that article included a comparison of the conversion rates (from downloads to sales) between the Mac and Windows versions of Airfoil.

As noted, these numbers were not what we'd expected. Readers had some comments about the specific comparison we made, noting that perhaps we should have measured the first 20 days of the Windows version against the first 20 days of the Mac version. This would also provide interesting data, but unfortunately, we don't have the traffic numbers for Airfoil Mac's release (March of 2005).

We think the most interesting and valid comparison is the conversion rate during a period of relative downtime. As Quentin noted in the yesterday's comments, releases are the exception. We can't yet compare the Windows version in a period of downtime, but I did dig up the relevant numbers on our other applications over the timeframe used yesterday. Yes folks, for the second day in row, it's a table not incorrectly used for layout.

 WinfoilMacfoilAudio Hijack ProAudio HijackNicecast
Downloads:X2.2X3.26X1.98X0.69X
Sales:Y2.0Y2.37Y1.29Y0.36Y
Conversion Rate:~5.4%~4.9%~4.5%~3.6%~%2.9

Conversion Rates Across All Applications, May 12th - May 31st 2006

What Have We Got?
This is all pretty rough, and unfortunately we're not in a lab where we can perform controlled experiments. Without that, it's really just numbers and without similar data from other companies, we can only compare among our own applications. It's clear that in downloads and raw sales, our products rank in order of popularity as: Audio Hijack Pro, Airfoil, Audio Hijack, Nicecast.

We also see that among our Mac products, Macfoil actually has our highest conversion rate, with Audio Hijack Pro just behind it. Nicecast is by far our poorest performer, probably due to its higher price tag.

Finally, given these numbers, I think it's safe to assume that Winfoil's conversion rate is artifically high due to its recent release. Time will tell, however. It looks like Windows is much friendlier to ISVs than we all think, but I doubt it outright beats the Mac platform.

So Where Does That Leave Us?
I was speaking with Cabel of Panic the other day (about t-shirts, of course), and he was bemoaning the fact that no one in the Mac world shares sales data. I'm always curious as well, but Rogue Amoeba keeps mum about our own numbers and I certainly understand when others do the same. However, I'm hoping this article might inspire others to share information on their conversion rates (particularly during non-release times).

So what say you Mac software developers? How about a meme across the Mac blogosphere? (Whee, buzzwords!) Brian Wilson of Unsanity, inspiration for this follow-up, I'm looking in your direction.

If others get on board, we'll see some posts showing the conversion rates for other applications. Hopefully this won't prove foolish on my part. If the average turns out to be 25% and our rate looks pathetic by comparison, I'll certainly look a fool. But never let it be said that Rogue Amoeba was timid, dear readers!

Updates

• Brian has rapidly answered my challenge.

• Numbers from The Little App Factory

• Numbers from Gaucho Software

• Numbers from Zengobi

• Flying Meat's VoodooPad Lite is kicking our butts. I'm not one to call Gus a cheater, but...

• Numbers from Delicious Monster

Posted by Paul | Permalink | View/Post Comments (6)

Comments



Mike Piatek-Jimenez
Fri Jun 9 17:26:34 2006

I've posted some numbers for Gaucho Software here:

http://www.starcoder.com/wordpress/2006/06/09/conversion-rates/


Scott Morrison
Mon Jun 12 17:48:37 2006

To provide a perspective from a software currently offered as donationware, I have seen about a 1% donation rate for MailTags 1.2.1 since its release in early April (Certainly, many of the downloads belong to past donators downloading the new version  -- so to be liberal, lets say that 1.1 to 1.25 % guestimate rate)  And judging by some reports, this is a fantastic return rate for software that is donation based and not crippled.

In anycase, I am going towards a registration model for the next major release and I am curious how that will affect conversion rates.

Ethan Gutmann
Fri Jun 16 13:30:44 2006

Another point to consider is "who are your windows users?".  Given that the numbers for that platform are lower than the numbers for the mac platform (quite substantially)... while the total number of people who use windows is greater than the number of mac users (again, quite substantially), I would suggest the following.  A large number of your windows users are also mac users and, perhaps a large portion of your windows users are people who user your app enough on the mac platform to buy it on the mac platform, thus making the people who looked into it on windows far more likely to buy it than ordinary.  Of course this is all conjecture, but I think the argument is not with out merit.  As you point out, the numbers aren't a perfect comparison.  It would be interesting to see some larger cross platform apps post similar numbers.

Paul
Fri Jun 16 19:14:37 2006

Scott: Indeed, from all I've ever seen, donation rates are just awful. The harder you make it to avoid paying, the better you'll do. We stumbled across a pretty good system with our current apps, where users can test everything but need to purchase to use it (due to noise deterioration). I'd say you'll definitely see an increase, there's little doubt there.

Ethan: While an interesting thought, as far as the hypothesis that most of the Windows users are Mac users, that's not proven to be the case, no.

As for larger cross-platform apps, what's out there? That's a pretty slim market.


This post is archived, and commenting has been closed.