Forecast with the uncertainty attached

Forecast {{metric}} for the next {{horizon}}. History: {{paste time series}}. Known future events: {{launches, seasonality, pricing changes}}.

1. Describe the series honestly first: trend, seasonality, level shifts, outliers and their causes, and how much history is actually representative of the current business.
2. Produce three scenarios — base, low, high — each with the explicit assumptions that define it, not just a percentage band.
3. Show the simple baseline (last value, or seasonal naive) alongside your forecast, and say whether the extra complexity earns its keep.
4. State the key drivers and how sensitive the forecast is to each — which single assumption moves the number most.
5. Name what the model cannot see: structural change, competitive moves, one-off events.
6. Give the early indicators to watch monthly that would tell us which scenario we're in, with the threshold for each.

No single-number forecasts. If the history is too short or too noisy to forecast, say so.