Forecast

The Forecast page answers one question per variant: how many do I need to buy, and when? It’s the working surface most operators spend the most time on — the place where sales history becomes a daily decision about which suppliers to email today.

A forecast here is not a sales report. Sales reports look backwards; the Forecast page looks forwards. The numbers in each row are projections: expected demand, the day you’ll need to reorder, the quantity to buy.

On this page

• What the Forecast page shows
• When to use it
• What you can do here
• Beginner workflow — Average model plus a fixed history window
• How it all fits together
• Key concepts
• Where to next

What the Forecast page shows

Each row in the table is one product variant. Each column is a fact, a metric, or a suggestion. The page comes with a small set of columns visible out of the box; you add the reorder columns yourself from the Columns popover.

The left sidebar lists your saved views — combinations of filters, columns, and sort orders. Switch between views without losing your filter state. The toolbar at the top gives you search, filters, the columns picker, sort, the location picker, forecast settings (model + history period), the Amplifier button, and a “Create PO from selected” action.

When to use it

• Daily. Check which variants have an overdue or imminent reorder dateReorder dateThe day Logistified expects your on-hand inventory to hit the reorder point. Acting earlier is fine; acting later risks a stockout. Read more → . Plan the orders you need to place today.
• Weekly. Review reorder dates in the next 7–14 days; group by supplier; fire purchase orders in batches.
• Drill-down. Open a saved view from the Overview page (or click the ABC chart on the KPIs tab), then click a row to see the full forecast for that variant.

What you can do here

Forecast view table Every column on the table — what it shows, where it comes from, when it's empty.

Forecast models The 12 forecast models and when each one fits.

Automatic model optimization Let Logistified pick the best model per variant.

History period Auto vs. Manual history windows — and which to start with.

Views & filters Save your filter + column combination as a reusable view.

Variant constants Per-variant min / max / target / reorder-point overrides.

Reorder suggestions The Reorder Forecast explained — how the suggestion is built.

Growth amplifiers Nudge a forecast for a campaign or product launch.

Locations & multi-location math How single-location vs. all-locations filters change the math.

Variant detail Per-variant forecast tuning and history drill-down.

Create a PO from the forecast Turn a row (or selected rows) into a purchase order.

Beginner workflow

The Forecast page can look complicated on day one — many columns, many filters, an Auto mode that quietly does its own thing. Here’s a way in:

1. Open the Forecast page.
2. Open the Forecast settings panel. Switch the Model to Average.
3. Switch the History period to Manual, then pick Last 90 days.
4. From the Columns popover, add Actual Reorder Quantity, Reorder Date, and Days of Stock to the visible columns.
5. Sort by Reorder Date ascending.
6. Save as a view named “Beginner — Average 90 d.” Star it for quick access.

Now look at the top rows. Do the suggested quantities and reorder dates roughly match your gut? If yes, you trust the inputs — switch the model back to Auto and let the system pick a better model per variant. If no, drill into a variant detail page; the History tab usually shows why.

Tip

The Average model with a fixed 90-day window has no hidden parameters. The reorder math is fully predictable. Once you trust it, switching to Auto will produce better suggestions for most variants — but starting transparent and switching to magic later builds your judgement faster than starting with magic.

How it all fits together

The Forecast page is the visible layer. Underneath, the math chains together:

1. Logistified watches your Shopify sales and aggregates them into a per-variant sales history.
2. A forecast model — picked automatically or chosen by you — projects demand forward.
3. Lead time, MOQ, pack size, and safety stock come from the variant’s primary supplierPrimary supplierThe supplier whose lead time, MOQ, pack size, and cost feed the reorder math for a variant. One per variant. You can change it any time. Read more → .
4. A simulation walks the forecast horizon forward day by day — draining on-hand at the predicted demand and crediting incoming POs and TOs on their expected arrival dates — to produce the reorder columns: reorder pointReorder pointThe inventory level at which you should reorder. Calculated as the demand you expect during your lead time plus a safety stock cushion. When on-hand drops to this level, it's time to send a PO. Read more → , reorder dateReorder dateThe day Logistified expects your on-hand inventory to hit the reorder point. Acting earlier is fine; acting later risks a stockout. Read more → , reorder quantityReorder quantityThe number of units to order. Starts as a raw target (cover demand until the next reorder), then adjusts for MOQ, pack size, incoming stock, back orders, and EOQ. Read more → , and days of stockDays of stockHow many days your current on-hand inventory will last at your current sales rate, ignoring anything incoming. A pure runway number. Read more → .

The most common operator workflow then runs in the opposite direction: filter to “Reorder within 14 days,” select rows, click “Create PO from selected.” A draft purchase order appears, prefilled with the suggested quantities.

Key concepts

• Reorder forecastReorder forecastThe combined "how many to order" and "when to order it" view for a single variant. Made up of two figures: the reorder quantity (units) and the reorder date (calendar). Read more → — the combined “how many” and “when” view for one variant.
• Reorder pointReorder pointThe inventory level at which you should reorder. Calculated as the demand you expect during your lead time plus a safety stock cushion. When on-hand drops to this level, it's time to send a PO. Read more → — the stock level at which to reorder.
• Reorder dateReorder dateThe day Logistified expects your on-hand inventory to hit the reorder point. Acting earlier is fine; acting later risks a stockout. Read more → — the day on-hand will hit the reorder point.
• Reorder quantityReorder quantityThe number of units to order. Starts as a raw target (cover demand until the next reorder), then adjusts for MOQ, pack size, incoming stock, back orders, and EOQ. Read more → — the suggested order size after MOQ, pack size, and EOQ adjustments.
• History periodHistory periodThe window of past sales the forecast looks at. Either **Auto** (Logistified picks the optimal window per variant) or **Manual** (you pick a fixed range — last 30 days, last 90, last year, max, or a custom range). Read more → — the window of past sales the model looks at.
• Growth amplifierGrowth amplifierA manual nudge to a forecast — "expect 30% more sales between Black Friday and Christmas." Stacks on top of whatever the model predicts. Use it for campaigns, launches, and known events the model hasn't seen yet. Read more → — a manual nudge for a campaign or known event.
• ViewViewA reusable, named saved state — columns, filters, sort, location scope, and forecast settings (model, history period, growth amplifiers, safety stock). Views are the unit of reusable analysis across Logistified: they power the Forecast page, Notifications (which variants to alert on), Reports (which variants to email about), and PO creation (which variants to order from). Most operators build many views — one per real-world question (overdue reorders, slow movers, single-supplier batches) — and switch between them depending on what they're doing. Read more → — a reusable saved filter + setup. Used by Forecast, Notifications, Reports, and PO creation — see Views & filters for the cross-app picture.

Where to next

• The Forecast view table reference — every column documented.
• The Reorder suggestions deep-dive — how the suggestion is built end to end.
• The Reference glossary — every forecast term defined.

Last updated: 19 May 2026