I feel quite fortunate to have an app that has been popular over the long term. I try to provide a decent app at a very reasonable price.

However, I've also found that even at $1.99 or even $0.99, many consumers in the iOS world are just not interested. I think my product is great, but it's difficult to convince potential buyers of this, especially in the (very) limited space of the App Store. Blog reviews help a lot, word of mouth helps even more. But there is a segment of the population that won't give you app a chance if it's going to set them back a buck or two.

In September I decided to put out a free version of that app, with vastly reduced functionality and the addition of banner ads through Apple's iAd network. The assumption at the time was that I'll get quite a bit extra exposure and still be able to generate some income along the way.

Of course, this wasn't an easy decision. It's a bit scary to decide to take a product that generates income and decide to give a form of it away for free. I mean, what would happen if everyone stopped buying the paid version? And if the iAd income wasn't all it was cracked up to be. I could potentially be destroying the very product that I am trying to expand. Not a great prospect.

I decided what I needed was to really take out a LOT of the functionality of the original, so the free version was "just a taste." If people are content with the taste, well, then they probably wouldn't have bought the full version anyway, so even if iAds generate a penny per user, that's better than nothing.

Converting the paid version to the free version was pretty simple, and integrating the iAd was a snap thanks to their quite excellent documentation.

I was fully prepared to yank my free version from the store the moment that I detected a plummeting of the paid version. For me, that would be disaster. So, on September 10, the free version was approved and hit the App Store. The first days returns were not wonderful. $3.56. Ouch. Of course, this was the first day and really it takes around a day for new releases to really propagate throughout the App Store.

Luckily, things increased by fair amount. Now, the free version is no blockbuster. In about 2 months, there have been around 20K installs. Not terrible, but nothing to write home about. Really, it only cracks the top 300 free list in the Entertainment category on the weekends.

However, I'm pleased with iAd. The eCPM is quite high (consistently around $10). The fill rate isn't great (typically around 20-30%, but occasionally as high as 80% or as low as 5%). I've seen complaints from other developers about the fill rate and how you can get closer to 99.5% with a network with AdMob.

Wow, sounds great to me.

So, like many others I decided to add AdMob as well into the app. When iAd would fail to load (since it has a much lower fill rate), I'd fall back to AdMob. I was ready to start rolling in the AdMob money since some days I'd see 20K or 30K requests go unfilled with iAd.

I didn't quite work out that way. Yes, the AdMob fill rate ends up being really really close to 100%, but the eCPM is so much smaller than iAd that the revenue doesn't come close. In addition, the click-through rates on AdMob are a bit lower than iAd. I don't see why that would be...are consumers actually aware of what iAd is and that the ads really are more interactive? I doubt it.

Here's the last seven days of data for iAd vs AdMob:

For a developer, iAd is the clear winner. But AdMob is a great fallback. Not only can you fill in those times that iAd won't be able to provide a fill, but you can also use their beta House Ads to serve ads for your own products when both iAd and AdMob doesn't have anything. My AdMob house ads actually have a click through rate of over 1%.

Of course, iAd is the clear winner for now. Who knows what the future holds...perhaps they will be able to increase their fill rate, but will it be at the cost of a lower eCPM? Even if the eCPM on iAd dropped by a factor of 10, it would still be over 3 times higher than that of AdMob (at least in my tests).

A couple of things I want to note. My paid app sales haven't really seen any effect from the free version (good or bad). I still make the assumption that those downloading the free version probably wouldn't purchase the other version anyway.

A really really interesting metric is the effective price paid per download of my free app. That is, the total revenue via ads divided by the total downloads. It's currently at nearly $0.15 per "free" copy and increasing every day (by a fraction of a cent). How long will it take to reach $0.70, the developer royalty from a $0.99 app sale? I'm guessing another 6 months, maybe less.

At that point, the "free" app truly becomes a paid app, or even better.