Planning cycle rebuilt around live evidence (L6)
Annual and quarterly planning stops running on hand built decks. Plans are assembled from live system evidence, assumptions are written as testable statements, and the plan is re-scored monthly against what actually happened.
The steps
- 01
Make assumptions explicit
Tool: Manual
Every plan line carries its assumption in testable form, for example win rate holds at the trailing four quarter average. Untestable assumptions are removed. Owner: finance plus RevOps. DoD: assumption register exists with a source per line.
- 02
Assemble the plan from the warehouse
Tool: Snowflake
Baselines come from modelled data, not from last year's spreadsheet copied forward. The deck becomes a rendering of the model. Pitfall: a parallel spreadsheet that diverges within a month. DoD: plan numbers trace to warehouse queries.
- 03
Re-score monthly
Tool: Claude
Each month, compare assumptions to actuals and mark them holding, broken, or unknown. Broken assumptions trigger a replan conversation, not a note in a deck. Owner: finance partner. DoD: monthly assumption scorecard published.
- 04
Change the meeting, not just the document
Tool: Manual
Planning reviews start with broken assumptions. If the meeting still opens with a status walkthrough, the rebuild did not happen. Owner: leadership. DoD: standing agenda updated and followed for a full quarter.
Tools in this playbook
Next playbooks
Unfamiliar terms are defined in the AI and Revenue Dictionary. Related frameworks live in the framework library.