codexkit-sales-forecast-analyzer

# Sales Forecast Analyzer ## When to Use - Forecasting revenue from historical sales or active pipeline. - Preparing weekly, monthly, or quarterly sales reviews. - Comparing committed, best-case, and upside scenarios. - Explaining forecast movement to founders, finance, RevOps, or sales leadership. ## Procedure ### Step 1 - Normalize Inputs Separate actuals, pipeline, assumptions, and qualitative signals. Do not mix closed revenue with open pipeline. ### Step 2 - Segment The Pipeline Group opportunities by stage, close date, owner, segment, product, and confidence where available. ### Step 3 - Apply Forecast Logic Choose the simplest defensible method: - historical trend for stable recurring sales - stage-weighted pipeline for active opportunities - rep commit for manager-reviewed forecast - scenario range when inputs are uncertain ### Step 4 - Explain Drivers Identify the movement drivers: - new pipeline - slipped deals - closed-won / closed-lost - expansion / contraction - conversion rate change - average deal size change ### Step 5 - Produce Scenarios Provide Base, Upside, and Downside scenarios with assumptions and confidence. Flag data quality gaps. ## Inputs | Input | Required | Format | |-------|----------|--------| | Historical sales | Recommended | Period, revenue, bookings, units | | Pipeline | Recommended | Deal, amount, stage, probability, close date | | Sales cycle assumptions | Optional | Win rate, stage duration, seasonality | | Forecast horizon | Yes | Month, quarter, year | | Business context | Optional | Promotions, market changes, hiring, capacity | ## Output ```markdown ## Sales Forecast - [Period] ### Executive Summary [Forecast number, confidence, main movement drivers] ### Scenario Forecast | Scenario | Forecast | Assumptions | Confidence | |----------|----------|-------------|------------| ### Pipeline Movement | Driver | Impact | Notes | |--------|--------|-------| ### Risks And Watch Items - [Risk] - [mitigation] ### Data Quality Notes - [Missing fields, stale opportunities, probability caveats] ``` ## Quality Criteria - [ ] Forecast method is stated and fits the available data. - [ ] Actuals, pipeline, and assumptions are clearly separated. - [ ] Scenario assumptions are visible and testable. - [ ] Data quality gaps are not hidden. - [ ] Recommendations are operational, not just numerical. ## Verification (4C) | Check | Question | |-------|----------| | **Correctness** | Are formulas, stage weights, win rates, dates, and totals calculated consistently? | | **Completeness** | Are actuals, open pipeline, assumptions, scenarios, and risks all covered? | | **Context-fit** | Does the method match the sales motion, cycle length, and data maturity? | | **Consequence** | What decision could be distorted if this forecast is overconfident? | ## Edge Cases - **Sparse history** - Use scenario ranges and clearly mark confidence as low. - **Stale pipeline** - Flag opportunities with old next steps or close dates before including them. - **Enterprise deal concentration** - Show forecast with and without the largest deals. - **Seasonal business** - Avoid straight-line forecasts unless seasonality is explicitly addressed. ## Examples > **Prompt:** "Use this opportunity export to forecast Q3 bookings. Show commit, base, and upside scenarios and explain slipped-deal risk." > **Good pattern:** "Base forecast is $1.2M because 64% of weighted pipeline is in late-stage deals with close dates before quarter end. Confidence is medium because 3 of 8 largest deals have stale next steps." ## Definition of Done - [ ] Forecast number and range are defensible from inputs. - [ ] Assumptions are explicit. - [ ] Risks and confidence are visible. - [ ] A sales leader can act on next steps. ## Changelog - v1.0.0 - Initial release