# What is the Best Forecasting Method for Seasonal Products?

{% hint style="info" %}
**Quick Answer:** For seasonal products, the most effective forecasting methods combine seasonal decomposition with trend adjustment — separating the underlying growth trend from predictable seasonal peaks. Brands with 2+ years of history benefit most from statistical models; those with limited history should blend historical seasonality curves with forward-looking sales targets.
{% endhint %}

**Seasonal forecasting** is the process of predicting demand that follows predictable, recurring patterns tied to the time of year — holiday shopping, back-to-school, summer peaks, weather-driven demand, or annual promotional events like Black Friday.

Seasonal products require different forecasting approaches than stable, year-round products because a simple average of recent sales will systematically over-predict in off-peak periods and under-predict during peaks.

## The Core Challenge with Seasonal Forecasting

A simple trailing average doesn't work for seasonal items. If you sell 1,000 units in December and 100 units in February, a 6-month average of 550 units would lead you to overstock in spring and understock heading into the next holiday season.

Effective seasonal forecasting requires:

1. **Isolating the seasonal pattern** from the underlying trend
2. **Applying a seasonal index** to adjust the baseline forecast up or down by period
3. **Adjusting for year-over-year growth or decline** in overall demand
4. **Incorporating known future events** — promotions, new channels, distribution changes

## Forecasting Methods for Seasonal Products

{% tabs %}
{% tab title="Seasonal Decomposition" %}
**How it works:** Splits the time series into trend, seasonality, and noise components. Applies the historical seasonal index (ratio of that period's demand to average annual demand) to a forward-looking trend line.

**Best for:** Products with 2+ years of consistent sales history and clear seasonal patterns.

**Strength:** Captures the shape of the season accurately.

**Limitation:** Requires sufficient history; struggles with rapidly growing or declining products.
{% endtab %}

{% tab title="Year-over-Year Growth" %}
**How it works:** Takes last year's actual sales by week or month and applies a growth rate multiplier.

**Best for:** Stable businesses with minimal structural change year-over-year.

**Strength:** Simple to understand and audit; naturally captures seasonal shape.

**Limitation:** Propagates last year's errors (stockouts, promotions) forward without adjustment.
{% endtab %}

{% tab title="Top-Down with Seasonal Curve" %}
**How it works:** Sets a total season target (e.g., total units or revenue for the quarter), then distributes it across weeks using a historical or planned seasonal curve.

**Best for:** Brands doing financial planning first, then breaking it down to operations.

**Strength:** Keeps forecast aligned to financial targets.

**Limitation:** Accuracy depends on the quality of the top-line target.
{% endtab %}
{% endtabs %}

## Key Inputs for Seasonal Forecasting

* [ ] At least 1–2 prior seasons of sales history at the SKU level
* [ ] Knowledge of which periods were affected by stockouts (to exclude from the baseline)
* [ ] Promotional calendar — which weeks had campaigns that inflated or shifted demand
* [ ] Any planned changes for the upcoming season (new channels, products, distribution)
* [ ] Supplier lead times — seasonal items often have longer or inflexible lead times

## When to Order for Seasonal Products

Seasonal forecasting must connect directly to ordering timelines. For holiday or seasonal peaks:

| Lead Time                         | Latest Order Date Before Peak              |
| --------------------------------- | ------------------------------------------ |
| 2 weeks                           | 3–4 weeks before peak (with safety buffer) |
| 4–6 weeks                         | 6–8 weeks before peak                      |
| 90+ days (overseas manufacturing) | 4–5 months before peak                     |

{% hint style="warning" %}
Missing the order window for a seasonal peak often means missing the season entirely. Build seasonal replenishment plans early and tie them to firm supplier lead times.
{% endhint %}

## How Moselle Handles Seasonal Forecasting

Moselle detects seasonal patterns in your sales history and incorporates them into SKU-level forecasts. You can review and adjust forecast values for specific weeks or months before finalizing — which is especially important heading into high-stakes seasonal periods.

For BFCM, holiday, and other major demand events, Moselle lets you layer planned promotional uplifts on top of the statistical forecast.

{% content-ref url="../../planning-and-execution/forecasting/refine-your-forecast" %}
[refine-your-forecast](https://learn.moselle.io/planning-and-execution/forecasting/refine-your-forecast)
{% endcontent-ref %}

{% content-ref url="../planning-and-operations/how-to-plan-inventory-for-bfcm" %}
[how-to-plan-inventory-for-bfcm](https://learn.moselle.io/faq/planning-and-operations/how-to-plan-inventory-for-bfcm)
{% endcontent-ref %}

## Frequently Asked Questions

### What if I only have one year of sales history for a seasonal product?

**Answer:** One year of history can establish a seasonal pattern, but it won't account for year-over-year growth or anomalies. Supplement with top-down targets and apply manual judgment for the first couple of seasons.

### How do I forecast a new seasonal product with no history?

**Answer:** Use a comparable product's seasonal curve as a proxy. Apply the same seasonal index to your sales target for the new product, and adjust as early sell-in or pre-order data arrives.

### Should I forecast by week or by month for seasonal items?

**Answer:** Weekly forecasting gives you better visibility into the shape of the peak and supports more precise ordering windows. Monthly forecasting is sufficient for financial planning but may miss the timing nuances that matter for operations.