Forecast accuracy retro
Compare last quarter's forecast against actuals — surface where the model missed and which reps over- or under-called.
Eluu — forecast-accuracy-retro
Shown in preview 
Runtime ~3 min per run
Tokens ~12K per run
Owner a revops colleague
Works with
Pick whichever tool your team already uses.
- CRM Salesforce
HubSpot 
Pipedrive - Messaging Slack
Teams - Spreadsheet Google Sheets
Excel
What it does
Forecast accuracy is rarely measured against the locked snapshot from the start of the quarter — most teams just compare actuals to the latest internal number, which has already drifted. This recipe pulls last quarter’s commit, best-case, and pipeline buckets as they were locked at quarter-start, joins them to the actuals, and surfaces where the model missed and which reps over- or under-called.
How it works
- Pull forecast + actuals. The colleague reads the locked forecast snapshot from quarter start, then loads every deal closed in the quarter alongside its rep-level commit / best-case / pipeline values.
- Compute variance + bias. Hit rates per bucket get computed, per-rep over/under-call bias is scored, and variance is sliced by segment (SMB / Mid / Ent). The 5 deals that swung the result most are identified.
- Ship the retro. A per-rep bias table and the 5 swing deals are rendered into a clean retro thread, posted to #forecast-retro for the next forecast call.
Setup
- Connect a CRM (Salesforce, HubSpot, or Pipedrive) with forecast snapshots enabled.
- Connect a spreadsheet destination (Google Sheets or Excel) for the variance compute.
- Connect a messaging channel (Slack or Teams) for the retro post.
Variations
- Run a per-region retro instead of company-wide.
- Swap the channel post for a coaching DM to each high-bias rep with their personal numbers.
- Run a year-end rollup that compares the four quarterly retros side-by-side.
