> ## Documentation Index
> Fetch the complete documentation index at: https://docs.stockful.app/llms.txt
> Use this file to discover all available pages before exploring further.

# Forecast settings

> Global defaults for lead time, safety stock, velocity and restock strategy, and the storefront traffic switches

Configure defaults for all SKUs in **Settings → Forecasting**. These apply to every variant unless overridden at the SKU level.

<Frame>
  <img src="https://mintcdn.com/ljsstudios/TUCamDUIjkRf5Okj/images/user-guide/forecasting--settings.png?fit=max&auto=format&n=TUCamDUIjkRf5Okj&q=85&s=0b5daa568bb6be489d21bb9ddb1c0231" alt="Forecasting settings" width="1324" height="1592" data-path="images/user-guide/forecasting--settings.png" />
</Frame>

## Replenishment

| Setting          | What it controls                                                                                                                   |
| ---------------- | ---------------------------------------------------------------------------------------------------------------------------------- |
| **Lead time**    | Days it takes to receive stock from your supplier (0–365, default 7). Longer lead time means earlier reorder recommendations.      |
| **Safety stock** | Extra days of buffer stock to keep on hand beyond lead time (0–365, default 3). Protects against demand spikes or shipping delays. |

<Note>
  Where a SKU has enough sales history, Stockful sizes its safety stock **statistically** - from how variable that SKU's demand actually is and your target service level - rather than the flat day-buffer. The day-buffer above is the fallback for new or sparse SKUs. Once a supplier has five or more received purchase orders on record, their measured delivery variability widens the buffer automatically too - an erratic supplier earns more cover than a punctual one.
</Note>

## Velocity calculation

| Setting             | What it controls                                                                                                                      |
| ------------------- | ------------------------------------------------------------------------------------------------------------------------------------- |
| **Velocity period** | Days of sales history to use when calculating how fast items sell (7–365, default 30). A longer period smooths out short-term spikes. |

## Restock strategy

| Setting               | What it controls                                                                                                                                                                                                                                                   |
| --------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| **Restock-to days**   | How many days of supply you want after restocking (1–365, default 30). Higher values mean larger, less frequent orders.                                                                                                                                            |
| **Demand adjustment** | A percentage modifier applied to velocity (-100% to +500%). Use this to account for anticipated seasonal changes - for example, +50% ahead of a busy period. Leave blank for no adjustment.                                                                        |
| **Sales grouping**    | How velocity is calculated for multi-location stores. **Per location** (default) calculates velocity separately at each location. **Combined** aggregates sales across all locations before calculating - useful when stock is fulfilled from a central warehouse. |

<Tip>
  When in doubt, set a slightly longer lead time than your actual average. It's better to reorder a few days early than to run out waiting for a shipment.
</Tip>

## Order on demand

Products set to **continue selling when out of stock** in Shopify are often sold to order rather than kept in stock. Turn this on and Stockful buys them against the orders you have already taken instead of against a forecast: no forecast-based replenishment, no safety stock and no target stock level. The recommendation becomes the quantity customers have ordered that you can't cover from stock on hand or stock already on order.

It's off by default. Plenty of stores set continue selling across the whole catalogue and still stock those products normally, and turning it on there would drop every reorder quantity to zero.

Individual products can opt in or out either way, from [SKU overrides](/user-guide/forecasting/sku-overrides).

## Storefront traffic

The master switch for every [storefront traffic](/user-guide/traffic) surface, plus the opt-ins that let traffic change your reorder and forecast maths. On by default for the insight surfaces; the maths opt-ins are off until you turn them on.

## Low-stock threshold

The low-stock alert fires when a SKU is projected to run out within **X days** at its current sales velocity (default 7). Because it's days-of-cover rather than a fixed unit count, the same threshold works for slow-movers and fast-movers without re-tuning per SKU.

### Units floor (for new SKUs)

Brand-new SKUs have no velocity history yet, so days-of-cover can't be computed. For those, Stockful falls back to a **units floor** - alert when stock drops to or below this many units (default 5). The floor only matters until the SKU has accumulated enough sales history; after that, the days-of-cover comparison takes over automatically.

Both settings live under **Settings → Tracking**.

## Settings hierarchy

When calculating forecasts, Stockful resolves each setting in order:

1. **SKU override** - if set on the variant, use it
2. **Supplier default** - if the variant is linked to a supplier with a default lead time, use that
3. **Global default** - the value from Settings → Forecasting

This lets you set sensible defaults globally, override at the supplier level for lead times, and fine-tune individual SKUs where needed.
