# Forecasting

> Build forecasts that are useful for planning — baselines, features, evaluation, and communication of uncertainty to decision-makers. Use when predicting demand, revenue, capacity, or other operational time-based quantities.

- Skill: `itsual/forecasting` (Agent Skill)
- Install (CLI): `npx skillmds@latest add itsual/forecasting`
- Raw SKILL.md: https://api.skillmd.com/api/skills/itsual/forecasting/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Data & Analytics
- Author: itsual (https://skillmd.com/u/itsual)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/itsual/forecasting

---


# Forecasting

## Overview

Forecasting supports planning under uncertainty. A useful forecast is accurate enough, timely, and packaged so decisions can use the range — not only a single point.

## When to Use

- Demand and supply planning
- Revenue and financial forecasting support
- Capacity and staffing forecasts
- Seasonal business planning

## Core Practices

- Establish naive/seasonal baselines first
- Incorporate known future drivers (promos, launches, holidays) carefully
- Evaluate with time-aware CV and business-relevant errors
- Provide prediction intervals or scenarios
- Monitor forecast bias over time and recalibrate
- Align forecast grain with planning decisions

## Principles

- Point forecasts hide risk — show uncertainty
- Bias can hurt more than variance for some operations
- Process changes invalidate historical patterns
- Stakeholder trust depends on honest error communication

## Verification

- [ ] Baselines are beaten or the gap is explained
- [ ] Evaluation is time-respecting
- [ ] Uncertainty is communicated to users of the forecast