What User Analytics Tools do You Use for Side Projects?

Hey @viperfx, great questions. I'll dive right in:

Wouldn't mixpanel be better for feature usage? It's easy to tie users to tracked events.

Your point is good, and let me make my point a little more specific. It's easy to tie individual known users to tracked events, which has a lot of great uses ranging from a pseudo session recording / UX interview, to detecting account sharing, to detecting fraudulent patterns. And unlocking the personalization (more on that in a sec).

But if I want to know if my feature has wide adoption among, say, 1000 customers, then I don't want to look at each user individually. I want the aggregate. GA does this particularly well, thanks to making available the aggregate navigational data of where the user was previously and where they went next. Additionally, using a GA Segment to compare filtered conversion rates (say, trial-to-paid) while broken down by another property (say, a UTM parameter) is just <chef kiss>.

Of course, you could get some such reporting in Mixpanel, just querying for a cohort of users who have had event X with property Y (say, have had a Page View for URLs containing /my-new-feature/). But you lose GA's navigational richness, the comparison of first-touch/last-touch marketing channels, and cross-device tracking (if that matters to you).

The debate between two major, powerful tools like GA and Mixpanel largely comes down to personal preference. Both are good and everyone will have a favorite. I choose to draw the lines there.

What personalization can mixpanel do? Do you mean messaging such as email or in-app?

Yep. You could ostensibly do a really sophisticated and customized onboarding sequence using Mixpanel, which benefits from having all of the behavioral data and some marketing data available.

There's also a webhook option, which opens up possibilities ranging from better data integrations to crazy Zapier workflows.

This area is where Mixpanel wins over GA.

I see huge benefit in having both available.