forecasting-readout

Forecasting Readout

Purpose

Turn a metric projection into a decision-ready forecast with explicit trend basis, assumptions, and uncertainty bounds - not just a point estimate.

When to use

Use this skill when:

• asked "what will revenue/users/orders be next quarter/month/year?"
• projecting a metric for planning, budgeting, or target-setting
• presenting a forecast to a stakeholder who will use it to make a decision
• evaluating whether a current trend leads to hitting or missing a target

When not to use

Do not use this skill when:

• there is insufficient history to support any projection (fewer than 3-4 comparable periods)
• the metric is driven primarily by an upcoming event or decision with no historical analog
• the question is why a metric changed, not where it is going (use root-cause-analysis)

Required thinking discipline

• Never produce a point estimate alone. A single number without a range implies false precision and misleads decision-makers.
• State the basis for the projection explicitly - what pattern does the forecast extrapolate?
• Distinguish extrapolation from causally grounded projection. Trend extrapolation assumes "what has been true will continue." That assumption needs to be named.
• Separate trend from seasonality. Projecting November revenue in July requires handling the seasonal pattern explicitly.
• A forecast is a decision input, not a commitment. State what would need to change for the forecast to be wrong.
• Evidence constraint: Every conclusion must cite specific data — a number, a rate, a segment, or a timeframe. Do not speculate without evidential basis. If data is insufficient, state what is missing rather than asserting an unsupported inference.

Workflow

1. Define the metric, entity, and forecast horizon (e.g., net revenue, all markets, next 90 days).
2. Characterize the historical trend basis:
   • Flat (no directional trend)
   • Linear growth or decline
   • Accelerating or decelerating growth
   • Mean-reverting or cyclical
   • Event-driven (spikes around campaigns, holidays, product launches)
3. Identify and separate seasonality: does the metric have known weekly, monthly, or annual periodic patterns? State how seasonality is handled (carried forward, averaged, ignored).
4. List key assumptions the forecast depends on:
   • No major product, pricing, or acquisition strategy changes
   • External environment remains consistent
   • Seasonality pattern from prior years applies
   • Any specific operational assumptions (new market launch, campaign planned)
5. Produce the point estimate plus a scenario range:
   • Base case: continuation of recent trend with normal seasonality
   • Upside case: trend continues at the favorable end of recent variance
   • Downside case: trend continues at the unfavorable end, or a known risk materializes
6. State conditions that would invalidate the forecast - what event or change would require revisiting the projection?