Reading the matrix
A forecast category transition matrix records how deals moved between categories over a fixed interval and converts those moves into rates. Rows are the starting category, columns are the ending category, and each cell holds the share of value that traveled that path. The diagonal is the value that stayed put.The matrix answers a question a single accuracy percentage cannot. When commit converts at 70 percent, the missing 30 percent went somewhere specific. Some slipped to the next period, some fell back to best case, and some were lost. Each of those has a different owner.
Build it from stored snapshots
Snapshot every open opportunity at the same time each week, capturing category, amount, close date, and stage. Diff consecutive snapshots at the deal level, then aggregate. Measure in dollars as well as in deal counts, since a single large deal moving categories can swing the rate on its own.
Three cells deserve standing attention:
- Commit to slipped. Value that kept its commit label but moved its close date past the period boundary. ORM finds a deal that slips from one quarter to the next becomes less likely to close even while it stays in commit, so this cell rarely recovers. - Best case to commit. The upgrade path. Watch when in the period it happens, because a cluster of late upgrades usually reflects pressure rather than progress. - Commit to best case. The downgrade path, and the honest one. A team that never downgrades is not being careful, it is avoiding the conversation.
Late movement is worth more than early movement
Transitions carry different weight depending on when they occur. A deal moving from pipeline to best case in week two is normal maturation. The same move in week eleven means an assumption changed late, which is exactly the kind of change that a static forecast absorbs poorly.
ORM groups opportunities with a machine learning model and predicts a close curve for each group, with curves running from 1 to 80 weeks and most of the expectation landing before week 12. Comparing observed transition timing against a group's expected curve shows which deals are moving on schedule and which are drifting, without relying on anyone's recollection of how similar deals behaved.
What it changes on the forecast call
The matrix gives a manager a defensible prior. When a rep upgrades a deal to commit in the final two weeks, the historical rate for that exact move tells you how often it worked out. That reframes the discussion from opinion to evidence.
For related measurement, see win rate and weighted pipeline.
Frequently Asked Questions
How do you build a forecast category transition matrix?
Take weekly snapshots of every open opportunity with its forecast category. For each consecutive pair of snapshots, record the starting category and the ending category of every deal. Count the moves, then divide by the starting population to get rates.
What does the matrix tell you that a conversion rate does not?
A conversion rate says what share of commit closed. The matrix says where the rest of it went, whether it slipped, dropped to best case, or was lost outright. Those paths call for different fixes.
Should you measure transitions weekly or by period?
Weekly for in-period management and by period for calibration. Weekly rates show deterioration early. Period level rates are stable enough to use as forecasting inputs.
What is a healthy commit to commit rate?
High and steady. Deals should enter commit and stay there until they close. Frequent movement in and out of commit means the entry criteria are being applied loosely, and the label stops carrying information.
Put these metrics to work
ORM builds custom revenue forecast models that turn concepts like forecast category transition matrix into prescriptive action for your team.
Schedule a Demo