Hacker News new | ask | show | jobs
by Kjeldahl 2199 days ago
I was recently looking for a good tool that supports both web site analytics and app analytics (custom events, typically pushed by SPAs). I looked at GA, Amplitude and finally Matomo (which I ended up with). GA and Amplitude either did not offer or made it hard to work down to the micro level, essentially tracking known individual users down to the singular event level. Matomo makes this easy, although it certainly looks a bit dated compared to the competition. And the free parts are somewhat limited (you need to buy stuff or hosting).

I would have though that there would be several decent packages offering www + app analytics by now, but as I wrote, options were quite limited. Some of the options mentioned in the subject here looks like good options for just website analytics, but I'm not seeing much as far as "app analytics" (custom events) goes.
2 comments
srrr 2199 days ago
There are many packages listed at https://github.com/onurakpolat/awesome-analytics . Heap is an example of macro+micro+web+app.
link
Kjeldahl 2199 days ago
Thanks for the tip. One of Heap's selling points seems to be that tracking events "manually" is over, everything is automatic. That might work if all "work" is defined as "stuff users do". For other types of "work" (calculation pipelines, job delegation etc) I'm sure being able to "micro manage" events can be useful. But sure, my use case might be different.
link
JackWritesCode 2199 days ago
Less companies are focusing on user level tracking, as it’s an invasion of privacy and compliance doesn’t allow it
link
srrr 2199 days ago
Less companies are focusing on user level tracking because one single user is not a meaningful statistical group.

Companies focusing on user level tracking today provide a different set of tools one might be used to and that can of course be compliant, see https://www.hotjar.com/.

link
Kjeldahl 2199 days ago
And I'm sure that makes sense if you have lots of users and low revenue per customer. If your use case is the opposite, tracking individual usage becomes more important. At least until you have lots of those users. After that, who cares! :P
link