What is everyone using for their tech stack, specifically for product analytics?
Has anyone used Amplitude?
3 Likes
Hi @Natasha, I’ve used Amplitude. Also familiar with Segment, Mixpanel, Braze, Appcues, Braze. Feel free to reach out if you’d like to chat directly.
Re: Amplitude, great as a starting point for behavioral analytics of your users. No matter what, I still find that I go back and forth between direct querying against our data warehouse, Amplitude and Cognos Analytics (BI tools).
3 Likes
We use amplitude; and we spent a month+ with our event taxonomy. And also, because Firefox in Private mode after v70 or so blocks your Amplitude events, it’s worth it to double check vs your backend data events in your warehouse. In other words, totally backing up what @DonovanOkang said.
3 Likes
One thing that’s I find often overlooked with people first starting out in instrumentation and behavioral analytics, is the impact of your event data schema and the ease of use of whatever analytics tool you decide on using.
3 Likes
Pendo, Heap, Mixpanel and my favorite is Gainsight.
@DonovanOkang and @FelipeRibeiro, can you please elaborate?
3 Likes
You mean on Taxonomy? So… You need to consider what exactly you are trying to event. And then we had to understand that the structure we would use will be a) used by business users, not just PMs and engineers and b) by people long after we are gone. So instead of calling an event something like: CFlow - CNCL Comp; we went with more readable CancelFlow - Cancel Complete. I know it sounds really self-explanatory; but it’s something we are glad we took extra time to go through. And we worked with the Amplitude success team on this… They were great to work with.
2 Likes
Yes, agree with @Felipe ^. Intuitive naming of events help your org better understand the data, facilitates self-serve analysis–>very important if part of your goal is to drive a data-driven culture in your org/company.
1 Like
Your data schema also impacts post-instrumentation data analysis depending on whether your data is spread out across multiple tables in your data schema, or your data is all housed in one massive table. Out of the box tools like Amplitude typically have expectations/limitations of how they expect your data schema to be. For us, we instrumented so our events were grouped by components with rich sets of properties. This makes direct querying easier and faster, but we hit snags in Amplitude with some analyses like Pathfinder and Personas not being able to drill down to the level of detail we needed to leverage that amplitude analysis…
1 Like
Wow! That was very helpful. Thank you so much @DonovanOkang for sharing your insights and experience. @FelipeRibeiro, Thanks to you too… 
This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.