Product Growth Experiments

A disciplined experiment programme for sign-up, onboarding, sharing and upgrade flows: trustworthy assignment and metrics, a scored backlog, sound statistics and a decision log the team learns from.

Free · Mac (Apple silicon) and Windows · v1.0.5 · Other platforms

Then open the Marketplace in the app, search for Product Growth Experiments, and click Install. It goes into your own files, and works offline from then on.

What it installs.

1

Pillar

1

Area

50

Projects

203

Starter tasks

🧩 Product Management

Product Growth Experiments

Running a backlog of in-product experiments on sign-up, onboarding, sharing and upgrade flows, with a disciplined process for learning from results.

  1. Growth experiment charter and decision rights
  2. Funnel baseline from sign-up to upgrade
  3. Exposure and assignment logging check
  4. Primary and guardrail metric definitions
  5. Choosing an experimentation platform
  6. Traffic and minimum detectable effect sheet
  7. One-page experiment brief format
  8. Growth experiment backlog set-up
  9. AA test of the assignment pipeline
  10. Pre-launch experiment QA checklist
  11. Experiment results log
  12. Weekly growth experiment stand-up
  13. Fortnightly backlog scoring session
  14. Live experiment health checks
  15. Experiment collision calendar
  16. Monthly experiment readout
  17. Concluded experiment decision log
  18. Searchable library of past experiment learnings
  19. Experiment programme scorecard
  20. Quarterly experimentation retrospective
  21. Reading confidence intervals and statistical power
  22. Peeking, early stopping and sequential tests
  23. Diagnosing sample ratio mismatch
  24. Writing falsifiable growth hypotheses
  25. Segment cuts without fishing for wins
  26. Bayesian and frequentist readouts compared
  27. Clean experiment code and variant removal
  28. Estimating the annual impact of a win honestly
  29. Sign-up form friction experiments
  30. Onboarding sequence experiments
  31. Invite and sharing flow experiments
  32. Upgrade and checkout flow experiments
  33. Choosing a test design for each idea
  34. Server-side versus client-side testing decision
  35. Pruning untestable ideas from the backlog
  36. Dark pattern and user trust guardrails
  37. First end-to-end live split test
  38. Two-week growth experiment sprint
  39. Launch review for a high-stakes checkout test
  40. Sharing a failed experiment at the all-hands
  41. Annual experimentation programme review
  42. Experiments in a low-traffic B2B product
  43. First 90 days as an engineer on a growth team
  44. Growth testing without a data analyst
  45. Restarting a stalled experimentation programme
  46. Testing through a redesign or replatform
  47. Long-term holdout for cumulative impact
  48. Variance reduction with pre-experiment data
  49. Interference and cluster tests in sharing features
  50. Novelty effects and long-run follow-up readings

Where it starts.

Each project arrives with its purpose, milestones and first tasks written out. Here is how the first three read.

Growth experiment charter and decision rights

This is a team plan, best installed into a shared workspace so everyone with a role in it can see and own their tasks. Growth teams that start testing without agreed rules end up arguing after every result about who can launch a test, what counts as a win and who makes the ship call.

First tasks

  • List every flow the team expects to test in the next six months
  • Draft approval rules for low-risk and high-risk tests
  • Name who makes the ship, iterate or kill call on mixed results
  • Circulate the charter draft to the product, engineering, design and data leads

Funnel baseline from sign-up to upgrade

Without a baseline, nobody knows whether a 2 point lift in sign-up completion is meaningful or ordinary week-to-week noise. Pulling eight to twelve weeks of conversion rates for each step from landing page to paid upgrade, with their weekly spread, shows where the biggest drop-offs are and how far a test must move a number before it counts.

First tasks

  • Write down every step from landing page to paid upgrade in order
  • Pull weekly conversion for each step over the last twelve weeks
  • Note the highest and lowest week beside each average
  • Save the queries next to the baseline with their date range
  • Refresh the funnel baseline with the latest twelve weeks @recurring(quarterly)

Exposure and assignment logging check

An experiment can only be analysed if every user's assignment and the moment they actually saw the change are logged, and the conversion events fire identically in both variants. Checking this before the first test catches the commonest silent failure: counting users who were assigned but never reached the changed screen, which dilutes every effect toward zero.

First tasks

  • Trigger a test assignment and find its event in the raw data
  • Confirm the exposure event fires only when the changed screen renders
  • Compare conversion events between variants in a staging build
  • Log each gap with an engineer and a fix date

Make it yours.

Install Product Growth Experiments, then tell Aurum the date you're aiming for. It checks whether the plan fits the weeks you actually have.

Free · Mac (Apple silicon) and Windows · v1.0.5 · Other platforms

Blueprints are general information to help you organise your own work. They are not medical, financial, legal or other professional advice. See the Terms.