forecast

Forecast generates a weighted sales forecast with best, likely, and worst-case scenarios, a commit versus upside breakdown, and gap analysis. It is intended for preparing a quarterly forecast call, assessing gap-to-quota from a pipeline CSV, deciding which deals to commit versus call as upside, or checking pipeline coverage against a number. It works in a standalone mode from uploaded or described data and becomes supercharged when a CRM is connected, in which case it can pull pipeline automatically, use real historical win rates by stage, segment, and deal size, factor activity signals into risk scoring, and track forecast changes over time.

To run, it needs pipeline data provided in one of three ways: a CSV export from a CRM such as Salesforce or HubSpot with at minimum deal name, amount, stage, and close date (and optionally owner, last activity date, created date, and account), a pasted list of deals, or a described territory. It also needs targets, namely the quota, the timeline or period end, and how much has already closed this period. Applying stage-weighted probabilities, it produces weighted values and scenario projections.

The output is a structured markdown report containing a summary of quota, closed to date, open pipeline, weighted forecast, gap to quota, and coverage ratio; forecast scenarios for best, likely, and worst case; a pipeline-by-stage table with counts, values, probabilities, and weighted values; a commit versus upside split; risk flags for issues like passed close dates, no recent activity, or deals still early with imminent close dates; a gap analysis with options to close the gap; and a recommendations checklist. Default stage probabilities are provided, for example Closed Won 100%, Negotiation or Contract 80%, Proposal or Quote 60%, Evaluation or Demo 40%, Discovery or Qualification 20%, and Prospecting or Lead 10%, and users can supply custom probabilities. Practical tips stress honest commits, keeping close dates current, treating roughly 3x pipeline coverage as healthy and below 2x as risky, and using activity recency as a risk signal.