Planning horizon

Planning horizon is the time window your plan covers — how many weeks, months, or quarters into the future you're forecasting demand and committing supply.

It's set by the longest lead time you have to commit to a decision today. If you place ocean-freight orders today that arrive in 16 weeks, your planning horizon has to be at least 16 weeks; otherwise you're committing to inventory blind.

The right planning horizon is rarely a free choice. It's a function of supply chain physics, not preference.

Different horizons for different decisions

Most operations run multiple horizons in parallel:

• Strategic (12–24 months) — annual budget, range planning, supplier negotiations, capacity decisions
• Tactical (3–12 months) — buying for long-lead inventory, marketing calendar, headcount
• Operational (4–12 weeks) — replenishment, allocation, promo top-ups
• Execution (0–4 weeks) — daily order pulls, dispatch, channel rebalancing

Each horizon answers different questions and tolerates different levels of forecast accuracy. Annual budget can be ±15% and still be useful; this week's allocation can't.

Why the horizon has to be at least as long as your lead time

If your lead time is 14 weeks and your planning horizon is only 12 weeks, the inventory arriving in week 14 was committed before your plan even reached that week. You bought it on instinct or last year's number, not on plan.

Rule of thumb:

Minimum planning horizon = Longest lead time + Cycle time + Safety buffer

For a brand with a 14-week ocean lead time, a 4-week production cycle, and a 2-week safety buffer, the minimum useful horizon is 20 weeks. Plan shorter than that and you're flying blind on the early part of the cycle.

Where the horizon breaks

Three failure modes:

• Too short. You can't see far enough ahead to make long-lead commitments. You're constantly placing emergency orders.
• Too long. You're forecasting 24 months ahead with the same precision as 4 weeks. Forecast accuracy at 24 months is not useful for daily operations — it's strategic input only.
• Mixed timescales in one plan. A single forecast that mixes the next 4 weeks (high accuracy needed) with months 18–24 (low accuracy tolerable) ends up either too noisy or too smooth. Better to run separate horizons with different accuracy expectations.

Forecast accuracy declines with horizon

Predictable consequence: the further out you forecast, the less accurate the forecast.

Typical accuracy curve:

This isn't a planning failure — it's a property of the underlying demand. Plans built on 52-week forecasts can't be granular at the SKU level; they have to be category-level or higher to be useful.

Rolling vs fixed horizons

Two ways to handle the time window:

• Fixed horizon. Plan a calendar year, freeze the plan, execute against it.
• Rolling horizon. Maintain N periods ahead at all times. Each month you drop the oldest period and add a new one at the end.

Rolling is almost always better operationally. Fixed plans go stale. Rolling plans stay current and force the discipline of continuous re-planning.

The most common cadence: monthly rolling 18-month horizon, with weekly updates to the next 4–6 weeks.

Connecting horizon to working capital

Longer planning horizons usually mean more working capital tied up in inventory — because you're committing earlier to inventory that won't sell for months.

The trade-off is real. You can't shorten the horizon below your supplier's lead time without exposing yourself to stockouts. You can shorten the horizon by shortening the lead time (different supplier, faster freight, regional sourcing). That's the lever to actually pull, not the horizon itself.