Sprint July 10 to July 21

[thmsobrmlr]

Global Sprint Planning

3 things that might take us down

1. Clickhouse knowledge
2. Incidents across EU and US

Retro: What can we do better next sprint?

Team sprint planning

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

# Team ___

**Support hero:** ___

## Retro

<!-- Grab the high and low priority items from last time and add whether that item was completed or not -->

- 

## Hang over items from previous sprint

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

- Item 1. prioritised/deprioritise

## OKR

1. OKR, status (red/yellow/green) and action points if yellow/red

### High priority

-

### Low priority / side quests

-

[thmsobrmlr]

Team Product Analytics

Support hero: @mariusandra

Michael is off 12-14th

Retro

• Marius: A lot of PR watering. Action matching HogQL bytecode (almost?) ready.
• Michael: Continued progress on PostHog 3000 navigation and support tickets.
• Thomas: Continued progress on Data Exploration and support tickets

Hang over items from previous sprint

• HogQL post-beta shortlist & PoEv2 next steps @mariusandra
• (Proper null support and materialized columns @thmsobrmlr (slack thread, older context))

OKR

See details here.

• 🔵 Objective 1: Port all our insights over to HogQL.
• 🔵 Objective 2: Sleeker insight editor UI

High priority

• Continue with https://github.com/PostHog/posthog/issues/18090 @Twixes
• Continue with https://github.com/PostHog/meta/issues/84 and fixes for outstanding issues

Low priority / side quests

• Get to the bottom of a backend-side caching / query calculation issue where users see empty results @thmsobrmlr

[pauldambra]

Next sprint - Team Noteblobs

Support hero: @pauldambra, then @pauldambra (we're switching to a two week rotation aligned with sprint so one person is always on support + sidequests while the others focus)

Retro

• we met in Cambridge, we laughed, we drank, we pizza'd

Hang over items from previous sprint

• ✅ Strong alignment on Q3 goals and onboarding
  • https://github.com/PostHog/posthog/issues/16287
• ♻️ Notebooks, notebooks, notebooks - (based on Q3 outcomes)
  • added conflict detection, nicer links, and storybook testing as an amuse bouche
• 🪣 Monitor rollout of S3 ingestion
  • up to 100% of traffic in EU and US
  • monitoring realtime playback with PostHog team

Goals

• 🎯 Goal 1 : Notebooks on ProductHunt 🟢
• 🎯 Goal 2 : Next 5 great things for Replay 🟢

High priority

• notebooks recording transcription mode @daibhin (with @benjackwhite)
• session recording user interviews @pauldambra and @daibhin

(Ben is currently travelling to a Serbian rave)

Low priority / side quests

• blob playback on for all teams @pauldambra

[neilkakkar]

Team Feature Success

Support hero: @liyiy then @neilkakkar

Retro

1. Survey customising appearances & emoji / nps rating survey type: @liyiy -> mostly sorted, chose interview survey type over emoji/nps
2. High priority experiments fixes: @neilkakkar - a.k.a make $big_customer and future small customers happy. -> Did a bit of priority juggling, got the most high priority experiment fixes sorted, while also working on moving decide to the read replica, because I realised decide read replica testing might take a bit longer, so front loaded it.

Hang over items from previous sprint

OKRs

1. 🟡 Make it easier for engineers to analyze the success of a feature (shall continue in Q3!)
2. 🔴 Launch pricing for feature flags (we'll launch in July, waiting for comms with existing big customers).

High priority

1. General release of surveys + emoji / nps rating survey type: @liyiy
2. Remaining experiments fixes + decide on read replica: @neilkakkar

Low priority / side quests

[timgl]

Team Pipeline

Support hero: Tiina + Xavier Xavier vacation first week, Tiina vacation second week + week after Harry leaving second week

Retro

• Investigate OOMs/tune RDKafka/Lazy setup for plugins @xvello > RDKafka is rolled back, need to investigate more. OOMs solved by throwing more memory at it. Shipped separate recording capture.
• Exports: shipping retries and backfills. Get Snowflake up to par. Maybe try to identify big users to move over @hazzadous > Shipped retries. We've got a backfill set up. Snowflake, need to tighten up retries etc but we're broadly in a good place.
• Great dashboards for alerts for ingestion @tiina303 > They exist. Worked on a lot incident follow ups, proactively making.
• Luigi fairly busy

Hang over items from previous sprint

OKR

• Defining limits on what users can send us
• Dodgy apps
  • Finish exports so we can deprecate export apps
  • Remove processing apps
• Figure out next 2 year architecture for pipeline
  • Figure out what our service levels should be

High priority

• get inserted_at column out, best-effort deduplication, making exports discoverable, deprecate old apps @tomasfarias
• Incident follow ups: break out onEvent/webhooks into separate consumers. Separate consumer for historical imports. Dashboard runbook follow ups @tiina303

Low priority / side quests

-

[raquelmsmith]

"Team" Growth

Support hero: Me

I will be out the second week, July 17-21. Anyone want to volunteer to help with billing support?

Retro

• [x] Decide on new pricing for product analytics and migrate existing/prior customers who are on old pricing.
  • [x] Make an announcement
• [ ] Mock up a product-specific onboarding flow for eg. Feature flags or Experiments
• CS requests:
  • [x] Handle new startup credits situation
  • [x] Make Enterprise trials a thing
  • Should be done by end of week

OKR

1. Starting to use and paying for new products feels frictionless
   • Status: 🟡
   • Reasoning: We swung too far from plans (where there was only one upgrade to do) to individual product upgrades, so now starting to use an additional product feels unnecessarily friction-ful. We want to keep the benefits of the latter but the simplicity of the former.
   • [ ] KR: We have separate onboarding for each product
   • [ ] KR: People don’t have to “upgrade” to different products, they just turn it on or start using it in some way.
2. Customers can pay a reasonable price for the value they’re receiving
   • Status: 🟡
   • Reasoning: We're charging too much for some products (eg product analytics) and too little for others (eg feature flags).
   • [ ] KR: Billing limits -> billing alerts
   • [x] KR: Make sure rollout of new product analytics pricing goes well
     • Mostly done
   • [ ] KR: Feature Flags pricing is launched, as well as Data Warehouse and CDP, if they are ready for launch

This sprint

High priority

• [ ] Get feature flags pricing ready for launch
• [ ] Mock up & start work on better upgrade flow for each product
• [x] CS: Get startup plan info into pocus https://github.com/PostHog/company-internal/issues/1060

Low priority / side quests

• [ ] Mock up a product-specific onboarding flow for eg. Feature flags or Experiments

[EDsCODE]

Team Data Warehouse

Retro

• pushed out MVP
• Worked through rough edges around columns and table errors
• James: fighting fires, clickhouse cleanup

High priority

Data warehouse

• Make joins work (this week)
• Internal use + 5 happy users for data warehouse
• MVP for person/group modeling

[ellie]

Team Infra

Retro

• Daniel: a lot of incident distractions but helm chart progress!
• Frank: 3 days on one incident, another one last week I don’t remember
• Ellie: Nice to see progress on the charts repo!

Goals

• Overall goal: work on chart splitting issues! @ellie @danielxnj @frankh
• Recovering data from CH @frankh
• Spike upgrading postgres @danielxnj
• Alerting wrap up (almost done) @ellie
• Start moving terraform deployments over @ellie