build-forecast-scenarios — Claude Code command
build-forecast-scenarios builds base, upside and downside revenue scenarios with signal-based adjustments, then explains the gap against the previous forecast. Useful ahead of a quarterly plan review.
Apache-2.0 ★ 450 updated 2026-04-03
.claude/commands/build-forecast-scenarios.mdWhat build-forecast-scenarios does
Five steps run in order: gather bookings, pipeline, product usage, renewal and macro inputs; recalibrate conversion, seasonality and growth; generate the three scenarios plus stress tests; write variance commentary against the prior outlook and targets; map GTM and finance moves to each case. The output is a forecast workbook with confidence bands, a variance memo and an action register.
Flags set the horizon (--horizon), cadence (--granularity as weekly, monthly or cohort), confidence band (--confidence at 70, 80 or 90) and --include-renewals true when CS signals belong in the mix. The optional --drivers flag takes JSON or CSV with macro, product or usage adjustments. Skip it and the model works from what the intake step collected.
Some steps hand off to the revenue-forecasting-analyst and revenue-intelligence-director subagents and to the cohort-analysis and exec-briefing-kit skills. The analyst owns the modelling, the director shapes the executive narrative.
What it helps with
- build base, upside and downside revenue scenarios for a board deck
- explain why this forecast moved against the last one
- tie hiring and budget decisions to a named scenario
- fold renewal signals into the model
- stress-test assumptions against macro drivers
Example prompts for the agent
- Build Q4 revenue scenarios with weekly steps:
/revenue-analytics:build-forecast-scenarios --horizon q4 --granularity weekly --confidence 80 --include-renewals true - Rerun the forecast with my driver file and tell me what moved:
/revenue-analytics:build-forecast-scenarios --horizon quarter --granularity monthly --drivers drivers.csv - Give me the downside case and what it means for hiring:
/revenue-analytics:build-forecast-scenarios --horizon fiscal_year --granularity cohort --confidence 90
How to install build-forecast-scenarios
Manually: put the files in place
.claude/commands/build-forecast-scenarios.mdCall it with /build-forecast-scenarios. For all projects use ~/.claude/commands/
Manually: put the files in place
~/.codex/prompts/build-forecast-scenarios.mdCall it with /prompts:build-forecast-scenarios. Codex marks this format deprecated in favour of skills
Manually: put the files in place
.github/prompts/build-forecast-scenarios.prompt.mdManually: put the files in place
.gemini/commands/build-forecast-scenarios.tomlGemini stores commands as TOML: the prompt text goes into the prompt field
Questions about build-forecast-scenarios
How do I install build-forecast-scenarios in Claude Code?
The command file lives in the gtmagents/gtm-agents repo. Put it in your project at .claude/commands/build-forecast-scenarios.md, then call /revenue-analytics:build-forecast-scenarios from a Claude session.
What does build-forecast-scenarios do?
It produces base, upside and downside revenue scenarios with signal-based adjustments, adds a memo on what changed versus the prior forecast, and lists actions tied to each scenario.
Does the command rely on other agents or skills?
Yes. Several steps route to the revenue-forecasting-analyst and revenue-intelligence-director subagents, plus the cohort-analysis and exec-briefing-kit skills.
Source text
build-forecast-scenarios.md · Apache-2.0
--- name: build-forecast-scenarios description: Generates base/upside/downside revenue forecasts with signal-based adjustments. usage: /revenue-analytics:build-forecast-scenarios --horizon q4 --granularity weekly --confidence 80 --include-renewals true --- # Command: build-forecast-scenarios ## Inputs - **horizon** – timeframe (month, quarter, fiscal_year). - **granularity** – cadence for projections (weekly, monthly, cohort). - **confidence** – confidence band (70/80/90) for intervals. - **include-renewals** – true/false to layer in CS signals. - **drivers** – optional JSON/CSV with macro/product/usage adjustments. ## Workflow 1. **Signal Intake** – collect bookings actuals, pipeline, product usage, renewal, and macro drivers. 2. **Model Calibration** – update conversion rates, seasonality factors, and growth coefficients. 3. **Scenario Generation** – produce base/upside/downside plus stress tests tied to driver toggles. 4. **Variance Commentary** – explain changes vs prior forecast and targets. 5. **Action Mapping** – recommend GTM + finance moves aligned to each scenario. ## Outputs - Forecast workbook (base/upside/downside) with drivers and confidence bands. - Variance memo explaining deltas vs prior outlook. - Action register tied to scenarios (investment, enablement, hiring, demand gen). ## Agent/Skill Invocations - `revenue-forecasting-analyst` – owns modeling + scenario analysis. - `revenue-intelligence-director` – crafts executive narrative. - `cohort-analysis` skill – benchmarks cohorts to validate assumptions. - `exec-briefing-kit` skill – packages summary for leadership/board. ---
© gtmagents · Apache-2.0 · gtmagents/gtm-agents