School season stationery stock: one spike, no reorder

One spike, and no second attempt

Most inventory planning assumes you can correct. A reorder point exists because demand and lead time both wobble, so you top up. That machinery is worth having and it does almost nothing for a school season. Selling through in eight weeks while replenishment takes six is not replenishment. It is a second buy that arrives after the buyer has gone.

So treat the season as one decision with a supply-side deadline. Count backwards from the earliest reopening date you sell into: shelf date, inbound appointment, transit, dispatch, production or import, and the supplier’s own slot, which in winter is contested because every other buyer is counting back from the same week. That chain usually runs four to five months. A June reopening is a January decision. An April session start is an autumn decision, which is why this is a September conversation.

One consequence gets resisted. With no second buy, being short and being long are not symmetric, and the answer is not a bigger number. It is a narrower list bought deeper. Buying the tail shallow is the right instinct for an untested line in an ordinary year and the wrong one here, because a tail you cannot reorder into fragments the buy across lines that each go short in the same fortnight.

Back to school is not one date

The dates are checkable and they are far apart. CBSE runs an April to March session and 2026-27 opened on 1 April 2026. Kerala’s schools reopened on 1 June 2026, the state marking it with Praveshanolsavam. Maharashtra’s state board schools resumed on 15 June 2026, and Vidarbha’s a week later on 22 June 2026. That is close to twelve weeks of spread, and one of those splits sits inside a single state.

West Bengal is on neither pattern. The West Bengal Board of Secondary Education issued its Annual Academic Calendar 2026 on 29 December 2025, under memo number D.S.(Aca)/940/A/25/6, and names its sessions by a single calendar year rather than a pair. A board publishing the whole year’s calendar in the last week of December is running a session that opens in January, which puts stock on shelf in Bengal a quarter before the CBSE cohort needs it.

Maharashtra also shows the second risk: a date moves. The state has announced an intent to align its board with an April start, while the calendar actually issued resumed classes in June. When an announcement and the issued calendar disagree, the issued calendar is what your stock has to meet.

The operating change is small and it is most of the value here. Stop holding a national back-to-school date. Hold a cohort per start date, each with its own on-shelf deadline, and split the buy across cohorts using where demand sat last year at city grain. The method for building that from your own history is the demand calendar. What it cannot give you is the asymmetry below.

Split the buy before you size it

Stationery does not expire. That sounds like relief and it is the category’s specific trap, because it removes the forcing function a dated product hands you for free. Nothing makes you decide. The stock sits, and eleven months later it is still in the room while you plan the next buy.

So make the split yourself, before sizing anything. Sort every line into stock that is still sellable next season and stock that is not. The carryable side is plain paper, pencils, erasers, sharpeners, geometry boxes, plain notebooks and files. The non-carryable side is larger than most teams admit: anything printed with a year, a syllabus or class marking, a licensed character, a school-specific cover, an exam-year planner, and any pack whose artwork you already know you will change.

Size the two sides differently, because their recovery rates are not close. Overbuying the carryable side costs storage and eleven months of capital, and you can price both. Overbuying the non-carryable side costs most of the value, because next season it is ordinary dead stock with no occasion attached. Cover on the dated side should be tighter than instinct suggests, and on the undated side looser than an aggregate rule allows. One service level across the category gets both wrong in opposite directions.

The cash shape of a one-spike year

The gap between paying for stock and being paid for it is the general constraint, and we have set out the cash model for a festive build. A single spike has a different shape. The trough is deeper, because no second peak refills the account before the year closes, and money leaves in the winter quarter for a season that settles in the monsoon. Settlement lands after the spike has passed, so the worst cash week usually falls weeks after the best sales week.

Two things follow. Model the trough at the level of the buy rather than the year, and stress only the downside. And watch what the accounts do to your reading of it. Unsold carryable stock stays on the balance sheet at cost and flatters gross margin until somebody provisions it, which is its own mechanic. A business holding three seasons of undated stock with no provision policy looks healthier than it is, right up to the year it does not.

Decide the carry rule in advance

Write the rule before the season opens, because the version written in July is cheaper than the version negotiated in September. It needs one date and three numbers. The date is when stock stops counting as in-season. The numbers are the weeks of cover at which the dated side goes to markdown, the carrying cost you will accept to hold undated stock for eleven months, and the recovery rate below which you clear rather than carry.

Timing the cut is a solved problem, and the finding that a first markdown is almost always too small applies here too, so use that rule and the clean routes for dated stock that will not carry. Measure age honestly, because a season that closed in July and a purchase order raised in January are two different clocks, and only one of them says anything about recovery.

The thing not to do is the default. Holding everything because none of it spoils is a decision to fund next year’s buy with this year’s mistakes, and it shows up as a catalogue that gets a little worse each season. What the reopening actually pulls is set out in the category guide.