How to calculate open-to-buy (and why the number goes stale by week three)

There are two ways people ask about open-to-buy. The first is the honest beginner question: what is the formula, and how do I fill in the cells. That part is easy, and we will do it in one paragraph. The second question is the one that actually matters, and almost nobody asks it out loud: once I have the number, how long is it true for.

The uncomfortable answer is that a calculated open-to-buy is a snapshot of a plan, and the plan starts decaying the moment the season opens. You did not do the math wrong. The math was fine. The world just moved, and your number did not move with it. Most brands treat OTB as a figure you compute once and defend, when the whole point of it is to be a dial you keep turning.

So let us do both. First the calculation, so nobody has to guess at the mechanics. Then the harder and more valuable thing: why the number you just computed is already drifting, and what it takes to keep it honest all season.

Open-to-buy is the amount of new inventory, in cost or retail dollars, that you are still allowed to bring in for a given period without blowing your inventory plan. The formula is a plan for where your stock should end up, worked backwards into how much room you have left to buy.

Start with your planned end-of-period inventory: the stock you want on hand when the period closes. Add your planned sales for the period, because you have to cover what you expect to sell. Add your planned markdowns, because marked-down units leave inventory too. That total is the stock you need to have moved through the business. Now subtract what is already committed: your beginning inventory plus any orders already placed and on the way. What is left is your open-to-buy.

Written out: OTB = (planned sales + planned markdowns + planned end inventory) minus (beginning inventory + inventory on order). If planned sales are 800, planned markdowns 120, planned end inventory 400, beginning inventory 500 and on-order 300, your open-to-buy is 520. That is the number you take to your suppliers. It really is that simple, and that is exactly the trap.

You can compute this in retail dollars or in cost dollars, and both are common. Retail is easier to reconcile against a sales plan; cost is what your cash and your margin actually run on. It does not matter much which you pick, as long as you are consistent, because the mechanics are identical either way. The formula is not where brands go wrong. What trips people up is treating the output as a fact when every term that produced it is an assumption, and assumptions have a shelf life.

It also helps to remember what the number is for. Open-to-buy is a spending limit, not a target. It tells you the most you can bring in without breaking the inventory plan, given everything you already have and everything already on the way. Used that way, it is a brake and an accelerator at once: it stops you over-committing on a category, and it tells you exactly how much room you still have to chase one that is working. That dual job is precisely why a stale number is dangerous. A brake that reads last month's speed is worse than no brake at all.

Look again at what fed that 520. Planned sales is a forecast. Planned markdowns is a forecast of a forecast, since it depends on how much of the buy sells at full price. Planned end inventory is a target you set before you knew how the season would open. Beginning inventory is real, and on-order is real, but everything else is an assumption you froze in a spreadsheet.

Now run the season for three weeks. Outerwear is pacing twelve points ahead of plan and dresses eight points behind. The full-price sell-through you assumed for markdown math has not held, so your planned markdowns are wrong, which means your end-inventory target is wrong, which means the 520 is wrong. Not by a rounding error. By enough that buying against it will either stock you out of the runners or bury you in the laggards.

This is the part the formula never tells you. A calculated OTB is only as accurate as its least accurate input, and its least accurate inputs all describe a future that is now visibly different from what you assumed. The number did not get worse because you were careless. It got worse because it stopped being updated while the business kept trading.

Three weeks is not long. It is barely past the first read on newness. And yet the standard cadence in most planning organizations is to recompute OTB monthly at best, quarterly at worst, which means the number is stale for most of the time it is being used to place orders.

There is a second, quieter problem in the arithmetic. The formula produces one number, and one number carries no sense of how sure you were of it. The 520 you get from a category you know cold and the 520 you get from a new launch you are guessing at look exactly alike on the page. When the guess turns out wrong, nothing in the calculation warns you. You placed the order at the same confidence you placed the safe one, and you find out the difference at markdown. A single point estimate is a formula with the risk deleted, and the risk is the part you most needed to see.

The fix is not a cleverer formula. The formula is correct. The fix is to stop treating the inputs as constants. Put open-to-buy on the same live model as your demand forecast and your merchandise financial plan, so that when actual sell-through, returns and channel mix come in, the planned sales, planned markdowns and end-inventory target all re-forecast, and the OTB re-derives itself from the updated numbers.

When you do that, the calculation runs continuously instead of once. The moment outerwear pulls ahead, its planned sales climb, its markdown assumption falls, and its open-to-buy expands, while dresses do the reverse. Money is released from the category the market rejected and made available to the one it is pulling, and it happens while there is still lead time to place and receive the order.

The guardrails you set stay in force through all of it. The working-capital ceiling, the markdown headroom, the MOQ and lead-time floors are limits the recompute plans within, not rules a buyer has to hold in their head at the end of a long day. The agent stages the reallocations, the buyer reviews the set, and the number that goes to suppliers is one derived from this week's reality, not January's.

The formula is a five-minute lesson. Keeping the answer true for thirteen weeks is the entire job, and it is the part a spreadsheet was never going to do.

So learn the arithmetic, because you should know where the number comes from. Then stop pretending the number is fixed. An open-to-buy that never updates is not a control. It is a photograph of a decision you made before the season told you anything. The brands that hold full-price sell-through are the ones whose OTB moves the week demand does, not the ones with the tidiest opening spreadsheet.