Solution Brief

    The Monthly Buy Plan: How Much to Buy for Products You Haven't Sold Yet

    Most of what a retailer buys each period — new products, new suppliers, one-time close-outs — has never been sold before, and for retailers where new items make up 80% or more of procurement, that's the rule, not the exception. The Monthly Buy Plan turns that into next year's purchase plan: how many units to buy, by category and by month. A dollar figure comes along for the ride, but the quantity is what a buyer actually acts on.

    New arrivalsNo sales history neededBudget-drivenPromotions & sales
    The Monthly Buy Plan main table grouped by category, showing a dollar figure and unit count for every category-by-month box across a 12-month plan
    A next-12-month purchase plan, by category and month — a unit count in every box, with a dollar figure alongside it.
    How It Works

    Plan by Category, Not by Item

    Each client sets a main category — and, where useful, a sub-category — once, in Settings. From there the plan computes the dollar and unit target for every category-by-month box automatically, from that category's own seasonal pattern, like the grid above. Every category always drills down into its own sub-category breakdown, and the whole plan — or every SKU behind it — exports straight to a spreadsheet. What it doesn't do is pick products: that's the planner's call. The system decides how much to spend and where; a person still decides what to buy with it, from whatever's in the new catalog.

    The Add Event dialog for a promotion, including a 'Like past promotion' field that borrows a similar past promotion's measured impact
    Planning a promotion: point it at a past one with “Like past promotion” to borrow a real, measured lift instead of guessing.
    Promotions & Holidays

    Promotions and Holidays Do the Heavy Lifting

    A new promotion doesn't start from a guess — it borrows its effect from a similar one that already happened. Point it at last year's comparable sale and the plan carries that same, real lift forward: a 30% Father's Day lift last June becomes a 30% lift this June, before the promotion has even run. Holidays work the same way — whatever actually happened around Black Friday, Diwali, or back-to-school last time is folded into this year's plan automatically.

    The explanation dialog for one month's plan, showing base level, month-of-year pattern, holiday impact, and promotion impact multiplied together into the units planned
    The full breakdown behind one number, in plain language — base level × seasonal pattern × holiday impact × promotion impact.
    Transparency

    Nothing Here Is a Black Box

    Click any box and see exactly how it was built, from four numbers multiplied together:

    • Base — this category's recent and full-history sales, blended into one monthly run rate.
    • Month pattern — how this category normally rises or falls across the year.
    • Holiday — the lift from past holidays, pulled in automatically from sales history.
    • Promotion — the lift borrowed from a similar past promotion, whenever one's scheduled in Events.

    For example, November: 79,911 × 1.00 × 1.22 × 1.00 = 97,382 units — about 10% above what actually sold last year.

    Set a Budget, See It Everywhere at Once

    Change the total and every category and month rescales instantly — no rebuild, no waiting.

    Plan From Day One

    Works before a single unit of a new item has sold.

    Gets Sharper Every Season

    Each promotion and holiday becomes evidence for next year's plan.

    buffers.ai already runs five forecasting models — TimesFM, LightGBM, Prophet, Exponential Smoothing, and Moving Average — to predict units for products with a track record. The Monthly Buy Plan solves a different problem: setting the budget for products that don't have one yet.

    Read next: How buffers.ai forecasts repeat items · Pricing

    See it on your own categories

    Talk to our team about your catalog and your next new-arrival budget, and we'll walk you through a Monthly Buy Plan built on your own data.

    Back to buffers.ai