People
- Marius Andra (Team lead)
- Michael Matloka (Full Stack Engineer)
- Paul D'Ambra (Full Stack Engineer)
Mission
Makers everywhere get better at building products because of PostHog
Q4 2022 Goals
Objective 1: Nail Data Exploration
- Build v0.1 of the data exploration system
- Run a design sprint with Cory
- Ship v2
Why?
Product reason:
- We have a pretty good interface for data visualization (trends, funnels, retention, etc), but not so much for data exploration (events, persons). Users have been asking for more.
- The biggest pain comes from not being able to move between different types of related data. Imagine this scenario (© @rcmarron)
0) Found an anomaly about feature usage in trends 1) "I wonder which paying organizations have used feature Y" 2) Search for that event/action and change the tab to organizations 3) "Interesting CompanyXYZ is using it, I wonder who" 4) Flip to persons + add an organization filter for CompanyXYZ 5) See the persons 6) "I wonder how/why they're using it" 7) Click recordings + see the recordings It would be pretty clunky/impossible to explore your data in that way with today's interface.
- Currently you need to re-create this same filtered query on 3-4 different pages and manually copy over the properties. It's doable, but not a great experience.
Code reason:
- We have a lot of bespoke bridges to show related data in the interface (people on insights, recordings on people).
- Every new one adds more complexity.
- This compounded by ever so slightly different use of filters on different data types.
- The further we continue down this path, the harder it'll be to have a fully integrated product across all data.
Meta reason:
- We now have separate product teams.
- If we're not careful, we'll end up with inoperable products.
- This builds the first shared bridge between all of them.
Potential ideas for how?
- Universal Search: https://github.com/PostHog/posthog/issues/7963 (Note: this doesn't mean natural language text search)
- An "explore" interface/table, which supports input from various swappable data sources, with a unified set of filters applied to data.
- Support for custom columns and formulas on those tables.
- Ability to save as a cohorts, save as an action, or export as a CSV when the data permits.
- The ability to link to these tables from anywhere. "Explore further" links wherever we show any data in the interface.
Future steps.
- Replace all other data tables in the app (events list, persons list, etc) with this interface.
- Interconnect all the ideas here: https://github.com/PostHog/posthog/issues/3765
- Add this power to the toolbar. E.g. add an action, see user recordings and explore further.
Objective 2: Support performant querying of every insight type at a scale of billions scale events
- Phantom is happy with query performance
- Instrument and significantly improve query performance for this scale of events
Roadmap
3 year goals
- You can explore data across all insights and dimensions
- You can trivially share any insight anywhere
- Onboarding is as easy as a video game
- Tight integration with developer workflows
- No more complex than it is today
- Using PostHog sparks joy
- We support trillion event querying
Feature ownership
You can find out more about the features we own here