My instinct is to explain @Jack Brickey 's answer with a bit more tech detail, as well as echo @Walter Buggenhout 's suggestion that you will probably want to look at each app separately.
The problem here is actually that apps have a lot of flexibility and power. This is a good thing on the most part, because it means apps can be built to do some really useful things, but a consequence is that there are a lot of very different entry points to them.
To track usage, you need to identify and log the usage of the entry points. But every one of those is unique to the app in use, although some can be grouped (for example, many apps provide workflow functions - same entry point for each type of function).
But a lot of apps provide many functions and in different ways, and it's easy to build one that is very hard to track usage of.
One example would be an app I wrote for a client which I know I could track - it had a bundle of project reports, field types and workflow functions. To track
- workflow functions - I'd search the workflow xml for my app's class name
- reports - I'd have to parse the server logs for people hitting the urls
- field types - I'd need to read the list of custom fields and check if any of them are of a type I'd written for
And then you'd need to cross-reference that with at least project config, but maybe also activity.
So, from three relatively simple additions to Jira, you can see there are already three totally different and potentially complex ways to identify usage. There are actually hundreds, each of which would need its own dedicated code to id...
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.