Sprint 1.41.0 2/2 - Oct 17 to Oct 28

[timgl]

Global Sprint Planning

3 things that might take us down

Retro: Status of Outcomes from Previous Sprint

1. Curation: 🎨 Make it easy to quickly create an AARRR-inspired dashboard. @pauldambra > We did research, probably not very valuable. We'll do more customer interviews.
2. Exploration: 🧭 Build v0.1 of the "data explorer" / "universal search". @mariusandra > Fail, tunnel vision trying to wrap up. Carry over to next sprint.
3. 🇪🇺 cloud needs productionalization + billing @ellie > in cloud and has billing!
4. Digitalocean single click + hobby deploy - make sure that works - testing if neccessary @fuziontech > it works, CI e2e tests are in review so should be wrapped up.
5. MVP for app metrics (how many events delivered, error rates) @macobo > so-so, main blocker is frontend experience but have a solution to this. Will get customers using this next week.
6. Ensure robustness of async handlers (jobs and scheduled tasks) @hazzadous (pair: @yakkomajuri) > rolling out to test teams for half of jobs. The rest of async handlers will roll out next week
7. Enabling Session Recording billing via BillingService @benjackwhite > Not on track b/c illness, EU cloud got in the way, but very close. End of the week we'll have this connected to dev.
8. Event filtering in Session Recordings & fuzzy filtering of events (MVP) @alexkim205 > This is done!
9. Revisit metahog (direct sql access) and set up for MVP deployment on self-hosted @EDsCODE > in progress, should be done for end of the week.
10. Cleanup and growth tasks for experiments @liyiy > on track to be completed. Switched OKRs so not implementing some of the additional growth tasks.

Retro: What can we do better next sprint?

1. EU cloud, we didn't have a clear owner. Lots of confusion between teams.

Support hero this sprint

Week 1: Tiina
Week 2: Marius

Plan: Proposed Goals for Next Sprint

Each goal should have a single owner. Owner can only be an engineer.

1. "I don't need PostHog to build all the features, I just need it to work" (citation needed) @Twixes
   why? User interviews highlighted above quote, so we're fixing it
2. Exploration: 🧭 Build v0.2 of the "data explorer" / "universal search". @mariusandra
   why? Hard to explore between related data.
3. Billing - Paying customer on the new BillingService @benjackwhite
   why? Billing is still our top prio
4. Event Summaries - Migrating our underlying CH table to surface key stats more easily @alexkim205
   why? Unblocker for higher performance and follow up features around querying and listing
5. 🚢 upgrade to Kubernetes v1.23 @danielxnj
   why? We don't want to be behind, our current version will come out of long term support
6. 🚨 sort incident handling (setup pagerduty to work similarly to incident.io with /incident, retro process, etc) @ellie
   why? we want to handle incidents better
7. Upgrade ClickHouse to >=v22.3.13 as our version has a memory leak causing issues @ellie
   why? memory leak almost took us down
8. Offsite the first week. Make hackathon stuff live second week Thinking of targeting "developer integrated featureset for feature flags". Code linking/error logging/CI @EDsCODE
   why? do some innovation
9. Insert service for CH @fuziontech
   why? So we can use clickhouse cloud for direct sql access
10. Removing any dependencies on Postgres availability from the pipeline @hazzadous
    Why: So our pipeline can be reliable even when external services are down
11. App metrics v1 @macobo
    Why: This is a requirement for CDPs and will give us and users more confidence in the pipeline
12. CDP design/architecture document @lharries
    Why: so we can have a sense of what to build for our CDP and how to build it

Team sprint planning

For your team sprint planning copy this template into a comment below for each team.

Team ___

## Retro

<!-- Talk about what went well, what didn't go well and any actions to improve next time -->

- 

## Hang over items from previous sprint

<!-- For each item, decide to re-prioritise (and add below) or deprioritise -->

- Item 1. prioritised/deprioritise

## Planning

<!-- Each item should have a single owner. Owner can only be an engineer. -->

### High priority

-

### Bugs

-

### Low priority / side quests

-

[mariusandra]

Team Product Analytics

Retro

• Marius: apps took all the attention, went too tunnel vision and ignored bugs
• Paul: text dashboards took longer, but a few quality of life dashboard bugs squashed along the way
• Michael: nailed actually working on product analytics
• Annika: great analysis

Hang over items from previous sprint

• Curation: 🎨 Make it easy to quickly create an AARRR-inspired dashboard. @pauldambra
  • Data now shows it's not going to be useful.
  • Text cards on dashboards solve some of the curation help.
  • More research needed if there's anything we do.
• Exploration: 🧭 Build v0.1 of the "data explorer" / "universal search". @mariusandra
  • V0.0.1 built

Planning

Paul off most of the sprint

High priority

• "I don't need PostHog to build all the features, I just need it to work" (citation needed) @Twixes
• Exploration: 🧭 Build v0.2 of the "data explorer" / "universal search". @mariusandra

Bugs

• TBD

Low priority / side quests

Media cards on dashboards

• why? builds on text card work

[benjackwhite]

Team Record-revenue-ings 💸

Retro

• @kappa90 basically doing billing 😅
  • Pairing is successful and having multiple people on a big topic makes a lot of sense
  • Should keep this up for items from the megaissue
  • Growth sprint planning and efforts are a little unclear atm. Assuming just to be included in SR work for now.
• @annikaschmid
  • Successfully used SR to actually learn things 🥳
  • Clear goals meant that things were pretty straightforward
• @benjackwhite
  • Billing is dragging along due to learning by doing but it's great to be pairing on this with Em
  • Jumping into actual SR feature improvements felt super good even though not the goal
  • I feel little bit distant from Alex partly because of being sick and partly because we are just out of sync
  • Next sprint want to invest more pairing time and also general time with Alex!
• @alexkim205
  • Worked on linking product analytics product to session recordings
  • Should continue working on interlinking as well as responding to incoming feedback
  • Also felt a little out of sync from team due to offsite and OOO schedules. This sprint should have more face-to-face time!

Hang over items from previous sprint

• Item 1. Billing! - I think should continue to be our high prio for next sprint
  • Get it on dev by end of week

Planning

High priority

1. Billing - Paying customer on the new BillingService

• why? Billing is still our top prio

2. Event Summaries - Migrating our underlying CH table to surface key stats more easily

• why? Unblocker for higher performance and follow up features around querying and listing

Bugs

Low priority / side quests

• Add simplified link to the given SessionRecording

[ellie]

🚀 Team Infrastructure 🚀

Retro

• @danielxnj: The onboarding process was very clear and well-organized, and everything was ready for me for my first day. I even got to see our infrastructure during my first hour at work.
• @ellie: EU launch went well! Infra has held up 🙌 Definitely need to handle communication better next time though (eg marketing not being sure on launch timings)
• @guidoiaquinti: enjoying a bank holiday today

Hang over items from previous sprint

N/A

Planning

@danielxnj has joined us! woo!

• 🚢 upgrade to Kubernetes v1.23 @danielxnj
• 🚨 sort incident handling (setup pagerduty to work similarly to incident.io with /incident, retro process, etc) @ellie
• Upgrade ClickHouse to >=v22.3.13 as our version has a memory leak causing issues @ellie
• Plan and explore "Migrate postgres out of heroku" goal @ellie
• Plan and explore "Regular synthetic load testing on production" goal @guidoiaquinti
• Plan and explore "Progress made on SOC2" goal @danielxnj

Also a note we've now got our board sorted out!

[EDsCODE]

Team Experiments

Retro

• Experimentation feature requests from Support Hero #10426 is done
• metahog revamp should be done by friday
• fix: change timestamps to be timezone aware upon parsing #11935 done

Hang over items from previous sprint

• NA

Planning

• Offsite the first week. Thinking of targeting "developer integrated featureset for feature flags"
  • code linking
  • error logging
  • CI
• Second week can be used to polish and deploy what was done at the offsite

[yakkomajuri]

Team Pipeline

Planning

• Removing any dependencies on Postgres availability from the pipeline @hazzadous
  • Why: So our pipeline can be reliable even when external services are down
• App metrics v1 @macobo
  • Why: This is a requirement for CDPs and will give us and users more confidence in the pipeline
• CDP design/architecture document @lharries
  • Why: so we can have a sense of what to build for our CDP and how to build it

[benjackwhite]

(For next sprint)

Team Session IPA

Retro

• @kappa90 morning cuddles with Ben are ❤️ - we should carry on with this more especially for close work

• @benjackwhite more cuddles with Alex wanted

  • Billing stuff is an iceberg, started to feel like we made the wrong call
  • Strong validation chat with Simon put me back on track of this was the right call

• @alexkim205 Not super involved with billing but good keeping in sync via Ben through the sprint

  • Little siloed working on completely different things to the other guys
  • Working on one goal together with someone please (cuddles)

• @annikaschmid

  • Not loads of progress due to other commitments
  • Super valuable and relevant user interviews reinforcing our decisions so far (+1 from Ben)

• @benjackwhite and @alexkim205 should work closely on something next sprint

Hang over items from previous sprint

• Releasing billing to new customers (maybe done before end of sprint)
• Supporting and monitoring the new billing
• Event summary migration stuff

Planning

High priority

• Saving groups of recordings for easy future finding / playback
  • Saving filter configurations for future usage
  • Saving specific recordings to a list
• Supporting Billing rollout

Bugs

• Fix URL linking to a recording from PersonsModal

Low priority / side quests

• Flamegraphs via Events summary

[mariusandra]

Last sprint

Team Retro

• We're in the "getting the crew together" stage of this team.

• One person out for almost 2 weeks when there are 3 of us is felt... especially when one of the other two is support hero. How did we not see this coming?

• Paul - bugs and leftovers for 2 days out of 10

• Marius - 1 week support, other week tables and support

• Annika - great offsite

• Michael - implemented cool new features

Goal retro

1. "I don't need PostHog to build all the features, I just need it to work" (citation needed) @Twixes

• why? User interviews highlighted above quote, so we're fixing it
• Status: Michael fixed what didn't exist or didn't work for him:
  • new math selector
  • min/max/avg count by user
  • in progress: adding avg/min/max count by group and property selector

2. Exploration: 🧭 Build v0.2 of the "data explorer" / "universal search". @mariusandra

• why? Hard to explore between related data.
• Status:
  • Solid progress on tables, didn't get to exploring actual data. See this sprint.

Next sprint

OKR: Nail Data Exploration.
why? Hard to explore between related data.

• Michael "doing a refactor of the SQL queries"
• Marius "getting the tables just right everywhere"