Sprint June 16 to June 27
Sprint Planning
List of incidents + any action items that haven't been scheduled yet
Team sprint planning
For your team sprint planning copy this template into a comment below for each team.
# Team ___
**Support hero:** ___
## OKR
1. OKR, status (red/yellow/green) and action points if yellow/red
## Retro
<!-- Grab the high and low priority items from last time and add whether that item was completed or not -->
-
### High priority
-
### Low priority / side quests
-
Team Naples not work
Support hero: @veryayskiy
@pauldambra out for a week ๐ ๐ฎโ๐น
Retro
- can't believe Alex's move was in This sprint, feels so long ago @pauldambra
- lots of tricky support @pauldambra
- "i had a nice week" really great to join lisbon onboarding, good cross pollination :) @veryayskiy
- happy with new UI, less frustrated now it's shipped @veryayskiy
##ย what we shipped
- comments on recording via annotations
- new filter UI to test impact with users
- created shared metrics for replay experiments
High priority
-
grab bag of plate spinning @pauldambra
- support
- โ work towards deprecate blobby v1? @pauldambra
- โ
Make it super easy to see what filters have results to watch - "show me, donโt make me figure it out"
- ๐งช show recordings in activity by time in the "figure out what to watch" tab @pauldambra
- ๐งช "get rid of the tabs" -> new filters UI without a modal? @veryayskiy
- ๐ throttling mutations
- https://github.com/PostHog/posthog/pull/32609
-
watch recordings from where they are
- polish polish polish @veryayskiy
- โ default saved filters?? -> just suggestions not created
- ๐ "migrate" to add types to playlists
- ๐ recent filters as suggestions too?
- etc
Side quests and lower priority
- RN - session timeout - stop of recordings (we have it in web) @veryayskiy
- better "not found" state for recordings @veryayskiy
- e.g. TTL passed, or recording might be sampled, etc etc
- lazy load session recording in the SDK @pauldambra
- CSP reporting finished - live and revenue (robbie and lucas doing most of this - hurrah) @pauldambra
OKRs ๐ก
- OKR, status (red/yellow/green) and action points if yellow/red
-
โ ๐ Goal 1: Get Alex 2 rocketing 2
- it's great to have Alex 2 on the team... let's make it a success
-
๐ฉโ๐ฌ Goal 2: Find the needle in the haystack
- Many of our customers have lots of recordings, but struggle to find the "useful" ones. We want to test and iterate on a bunch of different ideas to get customers to value faster.
- Understand the needles that people are looking for using user interviews, data analysis and more
- Better options for users to watch recordings from where they are (e.g. in funnels or experiments)
- Some people are scared of filtering, some people have no patience for it - give them ways to avoid it
- Make it super easy to see what filters have results to watch - "show me, donโt make me figure it out"
https://github.com/PostHog/posthog/issues/30460
High priority
-
keep an eye on running experiments @pauldambra @veryayskiy @annikaschmid
- https://us.posthog.com/project/2/experiments/120442
- https://us.posthog.com/project/2/experiments/125553
-
support @veryayskiy
-
zen mode design / investigation @veryayskiy
-
throttling mutations @pauldambra
- https://github.com/PostHog/posthog/pull/32609
-
"migrate" to add types to playlists (we have playlists and collections pre-split where we have to infer type) @veryayskiy
Side quests and lower priority
- CSP reporting finished - live and revenue (robbie and lucas doing most of this - hurrah) @pauldambra
- rrvideo investigation @veryayskiy * (how) does it work * temporal or existing celery export? * upsell or stickiness? * whole video / clip snippet?
Team Experience, Developer
Last sprint of the quarter starts now!
OKR
- ๐ท Build the "Product OS"
- ๐ข New layout and "file-based" project view. Almost done
- ๐ข Shortcuts
- ๐ก Every product is able to export operations (backend) and widgets (components) => AI, notebooks etc.
- ๐ก Suggestions
- too complex, will do something simple/manual instead
- ๐ด Keep iterating on shared UI components. Split out Lemon UI and Lemon Components.
- will do as needed / when someone asks for it
- ๐ด Nested nodes within nodes Notion style
- no time this quarter
- ๐ด Permissions
- environments are a blocker
- โฑ Improve dev tooling in PostHog
- ๐ก Roll out "product folders" for all teams
- ๐ก Convert E2E tests to playwright
- ๐ก Tests take 5-10min max
- โช Standardize dev tooling (~vscode~, flox, docker, self hosted)
Retro
The tree is live! People seem to be digging the latest iteration with category labels. Kept polishing and iterating, plus some devex work.
High priority
- [ ] Global UX efforts @adamleithp
- [x] Save to
- [ ] Top bar
- [ ] Design iteration @adamleithp + review from @corywatilo
- [ ] Design for top bar
- [ ] Tests tests tests @mariusandra
- [ ] Backend tests
- [x] All run in parallel in 5min
- [ ] Tests to product folders?
- [ ] Playwright tests
- [x] Chomp clickhouse migrations (these take minutes to run!)
- [ ] Some way to get them more stable?
- [x] Visual regression flakes
- [ ] Make PR to disable test once flake detected?
- [x] Frontend tests
- [x] Code quality takes forever
- [ ] Backend tests
- [ ] Upgrade to latest typescript
Low priority
- [ ] Figure out index pages VS table views
Team Customer Dance Party (CDP)
Meikel - whole sprint out Marcus - support
OKR
- OKR, status (red/yellow/green) and action points if yellow/red
๐ OMG so many destinations ๐ก ๐ Data pipelines UI sparks joy ๐ก ๐ค Pricing thats works for everyone ๐ด
Retro
- [ ] Do the zapier migration to CDP functions @benjackwhite
- Might get done still end of sprint
- [x] Get Webhook Sources MVP released behind a flag @benjackwhite
- [x] Modify templates api to only load from DB (hog and inputs schema will still be used on the function) @meikelmosby
- [x] Automatic template creation using AI tools @meikelmosby
- Out for hog code on transformations :o
- Goal still to get the full thing in place for Transforms alltogether
- [x] Load test the segment destinations with real data @MarconLP
- [ ] Implement retries for segment destinations @MarconLP
- [x] Add new hog function status to allow showing "alpha" destinations and explicitly hiding "hidden" destinations @MarconLP
- [ ] Deeper dive into Google ads and how CDP competitors are handling it @meikelmosby
- [x] @abheek9 sync with Brian just to be aware of this
- [x] Remove penalty for errors on hog functions (only penalising slow ones) @meikelmosby
- [ ] Move away from plugin-server triggering celery to some sort of daily job to notify about errors @meikelmosby
- [x] Build a cdp dashboard @abheek9
Low prio
- [ ] Roll out site destinations (the new hog js stuff) @benjackwhite
- Make sure RemoteConfig stuff is actually working
- [ ] Migrate plugin site apps (can we remove the opt in flag) @benjackwhite
- migrate plugins to site apps and check that fully works with posthog-js
- [x] Fix memory issue in the PR that gets things moved to mappings @MarconLP
- [ ] Link Tracking RFC - what should we do with this @meikelmosby
- [ ] Get up to speed on cdp competitors / market @abheek9
- [x] start gathering thoughts on cdp pricing @abheek9
- [ ] Fully rollout the new data pipelines @benjackwhite
High priority
- [ ] Get the segment actions fully released (a bunch manually verified as stable, the rest as alpha / beta depending on confidence) @MarconLP
- [ ] Get a PoC first "native" hog function running @MarconLP (with help from @benjackwhite)
- [ ] Pricing RFC @abheek9
Low priority
- [x] Add support for including a whole object in dictionary/json inputs
Team Surveynauts - Explorers of Feedback Galaxies ๐ฉโ๐
- Surveys support hero: @ioannisj
- PTO - 2 days @marandaneto
- PTO - 2 days @lucasheriques
- PTO - 1 day @ioannisj
๐ข=finished ๐ก=in progress ๐ด=won't finish โช=not started
Retro
Retro - Team Notes
What could've been better
- extra PTO that was not anticipated @ioannisj
- spent quite some time on keyboard avoidance issues for single/multiple choice questions with many options @ioannisj
- forgot to triage github issues as suppor t hero @lucasheriques
What went well
- Cory helping out driving research @lucasheriques @ioannisj @marandaneto
Retro - Tasks
High priority
- ๐ด Android surveys - release beta version @marandaneto
- ๐ก Flutter surveys - Finish
onRenderSurveyhandler/delegate in survey configuration for iOS and Flutter. @ioannisj - Will probably be done by Friday - โช Flutter surveys - Android - Start looking into how Android will be integrated similar to iOS @ioannisj - Will probably start but not sure this will be done by Friday
- Help out with Android @marandaneto
- โช Save custom templates @lucasheriques
- ๐ก Integrate MaxAI with Surveys - allow him to create surveys and help discoverability of our product @lucasheriques
- ๐ก Mobile surveys research @PostHog/team-surveys
Mobile replay and Mobile SDKs
- โช Manually start and stop session recordings - React Native @marandaneto
- ๐ข Release RN v4, Node v5, Web v4 @marandaneto
- ๐ข Android session replay do not capture screenshot during screen changes @marandaneto
- ๐ก Review and test Android, Flutter and RN for the migration from /decide to /flags endpoint @marandaneto
Low priority / Side quest
// Nothing
Goals
Click to expand quarterly goals
-
๐ก Mobile surveys, Everywhere
- ๐ก Android
- ๐ก Flutter
-
๐ก Level up survey analysis
- โช Global analytics view
- ๐ก A cleaner, more useful survey responses page
- ๐ก Be able to see surveys result breakdown by week
- โช Better results analysis with metric based templates
- โช Don't calculate NPS score for non-NPS questions
- โช Extra polish and UX improvements across the board, pair up with the Design team to get some ideas
-
๐ก Survey customization
- โช Better wait period controls
- ๐ก Alignment customization here and here
- โช Inline surveys
- โช Add icons
- โช Templated strings
- ๐ข Single-response submit
-
Templates, Templates, Templates
- โช Save custom templates
- โช Duplicate surveys across projects/environments
- ๐ข Add mobile-specific templates (no common patterns identified from existing usage)
- ๐ก Explore using AI to create survey templates
-
๐ก Make Key Customers Super Happy
- ๐ก Be lightning-fast at resolving their issues
- โช Build a case study with a power user
-
Hire a Cracked Product Engineer
- โช Hire and onboard our next teammate
- โช Make onboarding smooth and fast so they can hit the ground running
High priority
- Make onboarding smooth and fast so they can hit the ground running @lucasheriques
- Full-screen surveys @lucasheriques
- Focus will be to get Flutter Surveys Beta for iOS and Android finally out! @ioannisj
- Help with Flutter Surveys / Android side @marandaneto
- Surveys research @PostHog/team-surveys
Mobile replay and Mobile SDKs
Low priority / Side quest
Team Clickhouse
Support:
- CH @fuziontech
Retro:
High priority
Done:
- CHKeeper migration from ZK @fuziontech
- wontfix because of feedback from Altinity with concerns around CHKeeper - will require testing
- EU Performance remedies @fuziontech
- Upgraded node sizes to memory optimized
- combined all resources into one giant pool (leveled the priority between offline and online resources)
- Progress towards TF of coordinators
- AI Prop Map refining and backfilling in EU @fuziontech
- This is backfilled in EU and re-backfilled in US
- Have a look at which other materialized columns we can get rid of @tkaemming
Still in progress / bumped:
- Prop defs to CH read side performance improvements @tkaemming @bciaraldi
- Get started with upgrade clickhouse version @tkaemming @fuziontech
Unscheduled:
- Ingestion limitations on US (outage) @fuziontech
- Offline cluster in the US at capacity due to cohort calc reworking @fuziontech
- Query Observability and Performance Improvements
- Make ClickHouse ops easy
- Use Altinity Antalya in production for events
High priority
- Prop defs to CH read side performance improvements @tkaemming @bciaraldi
- Get started with upgrade clickhouse version @tkaemming @fuziontech
- TF for EU coordinators @fuziontech
- Dagster+ planning / pricing @fuziontech
i7ie.metal-48xlin our AZ @fuziontech- New users for grafana, cohorts and limit resources for those @orian
- clickhouse-connect work to allow us to use that (and http) for querying clickhouse @orian
Low priority / side quests
- Iceberg writer @fuziontech
โ =finished ๐ก=in progress ๐ด=won't finish โช=not started
Team Infra
Support hero: ___
OKR
- OKR, status (red/yellow/green) and action points if yellow/red
Retro
- โช Find fractional CISO / push security engineer job ad @MichaelKutsch-ph - we decided to put down the ad until we know what we actually need
- ๐ก Find replacement SIEM solution @MichaelKutsch-ph - some first meetings happened with SIEM providers, more to follow (Splunk, Panther)
- ๐ก Rollout tracing to prod @bciaraldi - partial eu, finishing django atm
- ๐ก RDS proxy, push to prod @bciaraldi - partial dev, took back seat to tracing
- โ Various Postgres upgrades/migrate to different instance types @bciaraldi
- ๐ก Continue to go through costs -> scale back ingress deploy @bciaraldi
- ๐ก Move persons to separate DB on EU @danielxnj - caused an incident. Root was feature flag table in EU doesn't have the primary key. Want to have everything prepared and run in EU.
- โ Whatever we get back from SOC2 auditors @danielxnj - Two open tickets, they'll be resolved on call next week
- @frankh Move logs to our s3 backed clickhouse cluster
High priority
Low priority / side quests
โ =finished ๐ก=in progress ๐ด=won't finish โช=not started
๐Improve build times for posthog apps ๐ฝ Splitting RDS ๐ AWS Account and networking re-work
High priority
- Move logs to our s3 backed clickhouse cluster OR optimize spend on clickhouse cloud @frankh
- Maybe help out clickhosue team with TF in EU or whatever @frankh
- Persons move done @danielxnj
- Tracing (django first) in US
- check out areas / tooling of security to focus on (appsec, endpoint security, cloud infra sec) @MichaelKutsch-ph
- meetings with Panther SIEM, Upwind, Wiz, Splunk lined up
- PoC / trial with one of the tools would be great
Low priority / side quests
Team Messaging
Retro
๐ข=finished ๐ก=in progress ๐ด=won't finish โช=not started
Haven: ๐ข Land https://github.com/PostHog/posthog/issues/31969 ๐ข DnD workflow editor @havenbarnes https://github.com/PostHog/posthog/issues/32417 โช (punt to later sprint) UX showing basic campaign appmetrics/logs analytics https://github.com/PostHog/posthog/issues/32429
Abheek: โช Pricing RFC https://github.com/PostHog/posthog/issues/32303 ๐ข Starting to dig into CDP Pricing as well ๐ข Land Email Templater UX + full liquid support https://github.com/PostHog/posthog/issues/32423
Mahamad: ๐ข Landing recipient preferences backend https://github.com/PostHog/posthog/issues/32438 ๐ข Preferences landing page https://github.com/PostHog/posthog/issues/32428
Moustafa / Ben (in parallel): โช Get a basic hogflow running in plugin-server https://github.com/PostHog/posthog/issues/32443 ๐ก Finalize database schema (may be done previous sprint) https://github.com/PostHog/posthog/issues/31998 ๐ก Hogflow CRUD API https://github.com/PostHog/posthog/issues/32432 ๐ก [SPIKE] Linking hogflows -> hogfunctions
- Making sure hogfunctions can run as hogflow steps, hogfunctions can invoke hogflows
PTO
- @havenbarnes (maybe 6/12 if movers come)
OKRs
- ๐ก Launch Messaging v1
This sprint
Haven:
- Pricing RFC https://github.com/PostHog/posthog/issues/32303
- https://github.com/PostHog/posthog/issues/33501
- https://github.com/PostHog/posthog/issues/32425
Abheek:
- Pricing RFC https://github.com/PostHog/posthog/issues/32303
Mahamad:
- https://github.com/PostHog/posthog/issues/32421
- https://github.com/PostHog/posthog/issues/32420
- https://github.com/PostHog/posthog/issues/32432
Moustafa / Ben (in parallel):
- Get a basic hogflow running in plugin-server https://github.com/PostHog/posthog/issues/32443
- Finalize database schema (may be done previous sprint) https://github.com/PostHog/posthog/issues/31998
- [SPIKE] Linking hogflows -> hogfunctions
- Making sure hogfunctions can run as hogflow steps, hogfunctions can invoke hogflows
- [SPIKE] Behavioral event listeners to support "event happened N times in last M days"
Team Ingestion
- Support: @pl
Retro
- ๐ข Validate traffic for legacy endpoints in capture-rs in prod EU and US @eli-r-ph
- ๐ข [Stretch] Shift legacy traffic to capture-rs @eli-r-ph
- ๐ก Ship managed migrations @nickbest-ph
- ๐ข Ship incremental person property updates @nickbest-ph
- ๐ข Fix end-to-end integration tests for ingestion @jose-sequeira
- ๐ก Ship batch group property updates @jose-sequeira
- ๐ข Investigate batched person property updates @jose-sequeira
- ๐ข Finish implementing Blobby V2 switchover @pl
- ๐ข Start doing sprint reliability reviews for ingestion @pl
Discussion
OKRs
- ๐ก Fix property definitions service
- ๐ข Finish moving to capture-rs
- ๐ข Develop ownership of ingestion incidents
- ๐ข Improve infrastructure for deployments
High priority
- Refactor and cleanup capture-rs @eli-r-ph
- Remove the old capture-py code and deployment @eli-r-ph
- Experiment with property-defs optimizations @eli-r-ph @jose-sequeira
- Implement batched person property updates @jose-sequeira
- Group batching @jose-sequeira
- Ship managed migrations @nickbest-ph
- Switch Blobby V1 off @pl
- Split out the DB endpoint for group tables @pl
- Review the state of ingestion in the SDKs @pl
Team Error Tracking
@hpouillot @oliverb123 one day each
Retro
๐ @fivestarspicy setup survey and ran first customer interview ๐ต๏ธโโ๏ธ @oliverb123 got to the bottom of batch size limits causing throughput issues (serde deserialization error) ๐ข @daibhin new features: client-side burst prevention, multipart symbol set uploads via the cli, 3x rules ๐ @hpouillot workspace migration branch ready and hoping to get it merged by the end of this week
โ
@daibhin migrated us to use the /products folder
โ
@oliverb123 symbol set / stack frame retention policies and deletion
โ
@oliverb123 shipped Python SDK context. @daibhin replaced old tags for better searching of Django exceptions
๐ @hpouillot NextJS wrapper ready but dependent on workspace migration. Decided on plan with Rafa
OKR
๐ข Make it easier to integrate ๐ข Improve product quality ๐ข Build advanced features for larger customers
High priority
Make it easier to integrate
- [ ] @daibhin GitHub Action to upload source maps
- [ ] @hpouillot Webpack and Vite plugins for source map uploads
- [ ] @hpouillot add React & NextJS packages to the JS workspace (depends on workspace migration)
- [x] @daibhin write docs for rules (https://github.com/PostHog/posthog.com/pull/11924)
Improve product quality (integrate with other products)
- [x] @daibhin finish GitHub integration
- [ ] @hpouillot session timeline (e.g. breadcrumbs)
- [ ] @oliverb123 frontend perf work (step 1: add loading time tracking)
- [ ] @oliverb123 extend CLI to include more git info
- [x] @oliverb123 Django integration improvements
planning ahead for next quarter
- [x] everyone: add features to public roadmap to get signal ahead of Q3 planning
- [x] @daibhin: prep for Q3 planning
- [x] @oliverb123 RFC on shared server SDK core
- [x] @fivestarspicy continue customer interviews
- [x] @fivestarspicy prep & share growth review (focus on new revenue growth)
- [ ] @fivestarspicy watch session replays (focus on issue management & performance / load times)
Side quests
- [x] @daibhin deprecate user groups in favour of roles (link)
- [ ] @fivestarspicy add ability to request SDKs to website
- [ ] @hpouillot integrate MaxAI (explain issues, write filters)
Team Experiments ๐งช
Support
Juraj
Retro
Anders:
- Implemented the backend for frequentist stats method. Established a new structure that we will port our bayesian method to as well, for a cleaner for maintainable implementation. Also fixes some statistical issues we have with the current implementation.
- Created a new experiment result format for more meaningful variable names and easier use in the frontend Implemented support for max, min and avg. aggregations on the new engine
- Rewrite of our funnel metrics to use the same udf as product insights funnel queries. Not completely done (missing support for properties). This fixes a bug (multiple steps of same type, HeyGen ran into this issue) and makes it easy for us to support any ordering (requested by HeyGen)
- Added organization setting for default stats method to use for new experiments
- Support
Rodrigo: Feature Work:
- View Recordings
- Results breakdown viz for the new engine Fixes:
- Search filters on experiment view
- Support
- Helped Dylan with Healf before Juraj took over
- Customer Call Ava Labs
Juraj
- Implemented new metrics UI for frequentist
- Refactored storybooks
- Support for Healf
Annika
- Since both query runner rollout and pricing were not super urgent this week, I worked on non-experiments work
- My goal for the rest of the week is to set up some user interviews, also to get some pricing feedback
- The query runner message admin I also just need to wrap up, should be quick, I will also get it done by the end of the week
Sprint goals
Annika
- Next steps experiments pricing
- Loop Pawel in
- Make list of open questions and start answering those
- Some user interviews to validate unit, and โpaying for both flags and experiments, flags for delivery, experiments for evaluationโ
Anders
- Help out on frequentist UI, plus everything else
- Finish funnel query rewrite (if not done yet)
- If time, port bayesian to structure as freqentist module
Rodrigo
- Refactors: get the query transformation unification done.
- Write principles of state management
- New engine/frequentist polish
- Side quest: turn the AI Summary into a user-facing feature.
Juraj
- Make frequentist production ready: UI components, documentation
- Progress the new query runner
- Merge funnel breakdowns
- Email to all (prep the cohort of target users for Joe)
- Slack message to top paying customers
- Roll out to 100%
- Side quest: Investigate current state of data warehouse integration
Team Max AI
Support hero: @kappa90
Georgiy out 1.5 week.
OKR
Retro
๐ก Aggregate session summaries MVP @sortafreel - Temporal architecture is ready, needs user-facing summarization ๐ข Action summarization live for users @skoob13 ๐ด Cohorts summarization live for users @skoob13 - on hold ๐ Easy way to add real-world cases to evals @skoob13 - rollover ๐ข Continue running chats with product teams @kappa90 - Set up chat with #team-data-warehouse ๐ก Deep research v0 @kappa90 ๐ข UI context @kappa90 - not in plan initially explicitly ๐ข New insight planning architecture @Twixes ๐ด Hunt down 99% of errors @Twixes - rollover โฌ๏ธ Future PostHog UX - done @jamesefhawkins
High priority
- Summaries of groups of sessions, built on new Temporal infra MVP @sortafreel
- Deep research UI @kappa90
- Hunt down 95% of errors @Twixes
- Merge SQL editor SQL skill and insight creation @Twixes
- Evals on actual real world cases @skoob13 (limited availability)
Extra points
- Assign Max AI members per product team: Michael with Data Warehouse, Em with Product Analytics, Alex with Session Replay
Team LLM O11y Support hero: ME
Out next week
OKR Retro Completed all items:
- Resposnses API in Node and Python
- Parse supported in both
High priority
- Improve docs to show all the great things we now do
- Roll out playground to all customers
Team Growth
Support hero: week 1 - @zlwaterfield , week 2 - @joshsny
Q2 Goals
-
Installation & SDK experience: Automate the setup of 50% of PostHog installations
- Setup wizard
- ๐ข Support more integrations
- ๐ข Uploading environment variables to common hosting providers
- ๐ข PR workflow
- Github action
- ๐ Automatic event instrumentation
- ๐ Be the default analytics provider for vibe coders
- ๐ Be the default option in boilerplates for NextJS, React & React Native
- ๐ Improve the onboarding experience for self hosting users
- Setup wizard
-
Make it easier to use multiple products
- โ Project homepage: make this a good place to see an overview of all your products
- โ Gather common pitfalls from user onboarding calls and replays in each product and remove them
-
โ Launch a creative product experiment once a month!
Retro
Sprint
@daniloc
- ๐ด Rough prototype of a starter
- ๐ด Attempt an LLM driven migration - scaffold a POC
- ๐ข Form opinions about MCP
- ๐ด Ship 2 blog posts
- ๐ข DeskHog tasks
- Rewrite insights parser
- Package up some deskhogs ๐ฆ
- Scripting DeskHog video
- Design API for card development/user management
- DeskHog shipping odyssey
- ๐ข Begin building events integration to wizard
"deskhog took over my life"
@joshsny
- B2B2C Integrations POC
- ๐ Disabling billing features for platform orgs
- ๐ข Testing querying data off of S3 directly using chdb
- ๐ข Changing the web analytics jobs to export data to S3
- ๐ข Make changes to the query endpoint to support querying data from S3
- ๐ Figure out some of the rest
- ๐ข MCP: Add tracking so we know how many users are using it
- ๐ข Event previews debugging in celery
Sprint
Time off: NA
@daniloc
- Ship wizard event implementation
- DeskHog: get helper going!
- DeskHog: ship to VIPs
- Begin playing with TanStack start
@joshsny
- ๐ Deploy OAuth
/authorizeand/tokenendpoints - ๐ข Background job for cleaning up old tokens
- ๐ Add authentication & permission classes for OAuth
- ๐ข Move the S3 backed web analytics implementation to use a seperate bucket
- Test S3 backed web analytics on an example team
- ๐ Support (both weeks)
- ๐ข Make SQL tool not use streaming in MCP server
- ๐ Make CI faster
Team Product Analytics
Sprint
Support: Sandy & Anirudh
- Sandy
- Week 1 Support
- Nima Onboarding
- Anirudh
- More CDP destinations for Alerts
- Statistical Analysis
- Week 2 Support
- Thomas
- Wrap things up (query migration, paths v2, colors)
Q2 2025 objectives
- ๐ Query node migration system (Thomas)
- We can evolve the schema of query nodes in non-trivial ways e.g. renaming a property or changing the type of a property.
- Frontend side query nodes are migrated or gracefully discarded.
- ๐ Paths V2 follow-ups (Thomas)
- The V2 version of the insight has feature parity with V1 and is rolled out to users.
- ๐ Insight colors follow-ups (Thomas)
- The dashboard breakdown colors feature supports all insights.
- The insight colors feature supports all insights, by adding a modal to allow customization of colors and handling the customizations in all chart types.
- ๐กAlerts (Anirudh)
- Migrate to temporal from celery
- Alert notifications in slack (supported from CDP destinations)
- ๐ก Statistical analysis (Anirudh working with Anna)
- Confidence intervals
- Forecasting
- Trend lines
- (stretch) Root cause analysis
- ๐ด AI generated insights (Anirudh)
- Suggested insights section (above all insights maybe)
- Once we have the users data, we can figure out what insights to generate (from say signup events,...) maybe even create entire dashboards
- Related charts
- When users are looking at a funnel step, we can suggest an autogenerated insight that shows dropoffs from the funnel step. Similar to related charts on Amplitude, we can dedicate a section to this on applicable insights
- Suggested insights section (above all insights maybe)
- ๐ก Improve Insight / Dashboard Load Times (Sandy)
- ๐ข Figure out EU database performance issues
- ๐ก Investigate top reasons why dashboard insights are not cached or are recalculated
- ๐ก Look into improving caching
- ๐ก Move cache warming to temporal
- ๐ก Edit Mode for Insight
- ๐ข Improve Exports (Sandy)
- ๐ข Move them to ~~temporal~~ blocking
- ๐ข Better alerting when fail (tell the user why, especially OOM, suggest less columns)
- ๐ด Think about increasing limits or helping the user with pagination
- ๐ด Action Improvements (Sandy)
- ๐ด Saved Filters
- ๐ด Investigate warehouse enabled actions (for insights that support it)
- ๐ด Data warehouse support in insights (Thomas)
- A query using data warehouse and events tables, with complex joins can be implemented using UI only.
- Onboarding improvements (Anna)
- Working together with #team-growth, we should continuously improve new user experience
- Insight tabs / unified "data explorer" (Anna)
- Figure out the UX
Team Web Analytics
OKRs
-
๐ก Make Cookieless mode production-ready
- ๐ข Add batching to redis calls
- Fix associating unidentified users who have allowed cookies
-
๐ข Page reports
- ๐ข Iterate UX based on feedback after general release
- ๐ข Error tracking cross-sell
-
๐ก Marketing analytics
- ๐ก Get ad spend data from 2-5 data warehouse connectors (most likely Google Ads and ~~CSV~~ JSON files, then other ad platforms that we can dogfood)
- ๐ก Build out the 80/20 of marketing analytics insights, e.g. total revenue and spending per channel and per campaign
- Show GSC data via the API or batching with no filters
-
๐ก Query performance
- Add sampling to web vitals
- ๐ข Experiment with pre-aggregating query results
- Improve sessions table join
-
๐ก HogQL virtual fields
- ๐ข Add support for person initial channel type
- Experiment with how to add e.g. CPA (average for channel) as a virtual person property
Retro
Support hero: @robbie-c
High priority
- @robbie-c ๐ข test cookieless on posthog.com
- @robbie-c ๐ด reach out to customers to get them using it
- @robbie-c ๐ด allow custom virtual fields based on hogql expressions
- @lricoy ๐ด support conversion goals with pre-aggregated tables
- @lricoy ๐ข support charts (trends) with pre-aggregated tables
- @lricoy ๐ก finish the backfill
- @jabahamondes ๐ข support self-managed source in marketing analytics
- @jabahamondes ๐ข nativly support google ads source in marketing analytics
- @jabahamondes ๐ข conversion goals in marketing analytics
- @annaszell ๐ด user research for marketing analytics (ask sales "who are some customers that have mature marketing motions (who probably use batch exports also, or significant amounts of revenue)")
stretch goal
- @robbie-c ๐ด total revenue virtual field
- @robbie-c ๐ด max ai -> web analytics filters
- @lricoy ๐ข continue helping @joshsny
This sprint
Support hero: @robbie-c
High priority
- @lricoy finish the backfill
- @lricoy remove the toggle and make it work automatically
- @jabahamondes conversion goal support property aggregations
- @jabahamondes conversion goal UI in the dashboard for exploration
- @jabahamondes iterate on feedback with Brian
- @robbie-c allow custom virtual fields based on hogql expressions
- @robbie-c max ai -> web analytics filters
stretch goal
- @robbie-c total revenue virtual field