How do they calculate those numbers though? They don't really seem to indicate their methodology, especially with results as stirring as these.
So each phone has a unique ID. Their analytics software (almost certainly) sends that back to their server to uniquely identify each phone. Said analytics package can also determine whether a copy of an app is cracked or not and relays that information as well. So they store that info in their database. If the app phones home later and isn't cracked, then chances are the user bought the app. Since the phone's ID is the same, they can pair the two and get a percentage of how many people actually paid for the app after having a cracked version. This isn't perfect but it's pretty close. There are loop holes. For example, crackers might change how they crack apps and Pinch's detection might not work anymore. I don't believe this has happened yet.
The Lite-->Full upgrade correlation is a little more fuzzy and they would have had to do some guess work (e.g. substring matching on names like "BillMinder" and "BillMinder Lite" and so one) to get that stat since there's not any way to tell their system (AFAIK) what lite/full versions are a pair. They might have done some other clever tricks but I don't know what those might be off hand. Would have to think about it.
Edit: And I should mention that chances are Pinch Media has some clever hackers (by that I mean good programmers) working for them. Building a decent, scalable analytics app and mining the data isn't easy. Building a crappy one is easy.
Any serious developer is also writing for Blackberry and/or Windows Mobile and/or Android simple to gain exposure and lessen loss.
Yea I gotta disagree w/ you on this one. WinMo isn't exactly a thriving platform right now. And RIM devices are prevalent (3 of 5 top smartphones sold in US ATM, IIRC), but there are different form factors/series to support, which means additional work since one has to write an app a few times (different screen resolutions, OS versions that may not provide as many features/APIs, touch and nontouch screen versions, etc.). Android market is still in its infancy. I think it will grow to a good size over the next 1-2 years, at which point we may jump in. I don't think it's ready to support a business yet. Then again, it's arguable whether the App Store is with its wacky economics. I'm not saying it isn't worth it, just that it's not as easy as just writing another app. Maintenance is not something that should be overlooked. We pour a *lot* (lot lot lot) of time into maintaining/updating BillMinder. And, in my mind, the definition of a "serious" developer is subjective.
Yes but 90% of the lite versions are crippled down or have features missing or its not even enough of a lite to decide yet, I would like to see something like Verizon has with Get it now were they do trial modes that expire in x amount days and you can have the full version to try out.
So it's against the rules to do a time-bombed application because Apple wants "Free" to mean Free on the App Store. Although I wish there were more flexibility there, I get where they're coming from on this one. If the Lite versions weren't crippled somehow, there would be no incentive for users to upgrade. It kind of sucks as a user. We try to leave enough in our Lite app to let users get a good taste of what's available in the full version while still giving them reasons to upgrade. As much as we'd like to help folks with our software, we also want to be paid for our time (as I'm sure everyone does in some fashion -- not necessarily just on the App Store). We're in the process of updating BillMinder Lite -- been busy working on the full version plus other apps. I'm sure you already knew all that but I meant to point out that no one would bother upgrading if the Lite versions weren't slimmed down and that Apple does actually have good reason for this one, as limiting as the reason is.