Somewhere in the vendor agreement you signed to get on shelf, there is a paragraph that reads close to this:

Vendor shall provide Buyer with no less than ninety (90) days’ prior written notice of any price increase, accompanied by documentation substantiating the increase. Buyer reserves the right to accept or reject any proposed increase in its sole discretion. All purchase orders issued prior to the effective date shall be filled at the pre-increase price.

Four sentences. Most brands sign some version of them, file the agreement, and never read past the first one until the day a supplier raises a price and someone asks how quickly it can be passed through.

The answer is: much more slowly than the cost moved, and the gap comes out of your margin. Each sentence in that clause is doing specific work. Here is what each one costs.

“No less than ninety days’ prior written notice”

The ninety days is the part everyone remembers, and it is the smaller half of the delay.

The clock starts when the buyer acknowledges the notice, not when you send it. Notices sent to a departed category manager, or to a portal nobody monitors, do not start anything. Brands routinely discover three weeks in that their notice is sitting unread, and the ninety days begins again from the acknowledgement.

The larger half is upstream. Ask an operator when their cost went up and they will name the month the supplier’s letter arrived. That is rarely the month the cost actually changed. A supplier price change lands on the next purchase order, which is produced, shipped, received, and invoiced over the following weeks. Weighted-average costing then smooths the new landed cost into the old one, so the per-unit number drifts upward across two or three receipts rather than stepping on a single date. By the time the margin report makes the change visible, the cost event is sixty to ninety days behind you.

Stack the two and the real sequence looks like this:

Stage	Typical elapsed
Cost change to first affected receipt	30–60 days
Receipt to visible in margin reporting	15–45 days
Internal decision and packet preparation	15–30 days
Notice period	90 days
Cost event to new price on the invoice	5–7 months

That interval is what you are absorbing, and it is worth pricing. Take a line doing $200K a month at wholesale with COGS at 62%. Landed cost is $124K a month. An eight percent supplier increase adds $9,920 a month. Six months of lag is $59,520 of margin you never recover — the increase is prospective, so nothing you win in month seven claws back months one through six.

The lever with the most give in it is not the notice period, which is contractual. It is the sixty to ninety days before the notice, which is yours.

“Accompanied by documentation substantiating the increase”

This sentence is where most requests die, and the reason is a mismatch between what the brand thinks it is proving and what the buyer is checking.

Buyers accept evidence that maps to the bill of materials of the specific item: a supplier’s price-change letter naming the component and the new unit cost, freight invoices showing a lane rate change, a published index for a traded input like resin or corrugate, customs documentation for a duty change. The test is whether an analyst can trace the number in your request back to a document and forward into the cost of one unit.

Buyers reject general cost narratives. “Our costs have gone up across the board” is not evidence. Neither is a blended percentage across your portfolio, because the buyer is not buying your portfolio. And costs that sit outside the product — overhead, headcount, marketing, a warehouse move — are almost never accepted as justification for an item-level increase, however real they are to your P&L.

The other common failure is arithmetic. A component that represents four percent of the bill of materials cannot support a six percent price increase, and a category analyst with a calculator will find that in a few minutes. Once they do, the credibility of the whole submission is gone and you are starting again next quarter.

What survives review is a per-unit cost bridge. One page, one SKU, current cost to proposed cost, every line traceable:

Item: 12oz jar, 6-pack case            Per case
  Glass jar (6)              $2.94  →   $3.36    +$0.42   supplier letter, dated
  Closure and liner (6)      $0.66  →   $0.66     $0.00
  Ingredient blend           $4.10  →   $4.31    +$0.21   commodity index
  Label and carton           $0.88  →   $0.95    +$0.07   supplier letter
  Inbound freight            $1.12  →   $1.34    +$0.22   lane rate, carrier invoice
  Co-pack labor              $2.30  →   $2.30     $0.00
  ─────────────────────────────────────────────────────
  Landed cost per case      $12.00  →  $12.92    +$0.92   +7.7%
  Requested case price      $19.35  →  $20.27    +$0.92   +4.8%

Note the last two lines. The request passes through the dollar increase rather than the percentage, which holds your gross margin dollars flat and lets your margin rate fall slightly. That is a deliberate concession and worth naming in the cover note, because the alternative — asking for 7.7 percent to preserve the rate — reads as opportunistic and invites a line-by-line fight you will spend two quarters on.

“In its sole discretion”

Read plainly, this sentence says the buyer can decline. In practice a flat rejection is uncommon; what you usually get is a counter, and the counters are predictable. Not this much. Not on this date. Not without something in return.

The something is the part brands leave out. A category manager is measured on category margin dollars, shelf productivity, and hitting a promotional calendar that is already built. A request that improves your economics and does nothing for theirs is a request they have no reason to champion internally, because they also have to defend it upward.

Bring the trade. A promotional commitment for the next two quarters, a new item they have been asking for, agreement to discontinue a slow SKU that is occupying a facing, an improvement in fill rate with a service-level commitment attached, extended terms. Any of these gives the buyer a line to put in their own justification. The submission that gets approved is usually the one that reads as a plan for the category rather than a demand from a vendor.

Timing helps too. Increases land better ahead of a category reset than in the middle of one, and better outside a promotional window than during it. If your notice period can be aimed at a reset date rather than a calendar quarter, aim it.

The order you approach accounts in

The clause governs one relationship. You have several, and the sequence you work them in changes the outcome — which is a decision most brands make by accident, usually by starting wherever the margin pain is loudest.

Two constraints shape it. The first is precedent: whatever you agree with one account becomes the reference point for the next, because category managers talk to brokers, brokers talk to everyone, and your price list circulates further than you think. The second is contractual. Many retail agreements carry a most-favoured-nation clause obliging you not to offer better terms elsewhere, and some require you to extend any more favourable pricing automatically. Under that kind of clause, a concession granted to hold one account propagates to every account carrying the same language, and an increase you soften in one place gets softened everywhere.

So read the clauses before you sequence, and know which of your accounts are linked. Then the usual play is to start somewhere mid-sized, with a buyer relationship that survives a difficult conversation and volume small enough that a rejection is not existential. A completed increase there gives you two things for the next submission: a tested packet, and the ability to say the increase is in market rather than proposed.

Going to the largest account first inverts both. You are negotiating your least practised version of the argument against your most sophisticated counterparty, with your biggest volume exposed, and whatever you concede sets the floor for everyone behind it.

“All purchase orders issued prior to the effective date shall be filled at the pre-increase price”

This is the sentence that costs the most and gets the least attention.

You have just told a buyer, in writing, exactly when your price goes up. A competent buyer responds by ordering ahead. Forward-buying in the notice window is standard practice, it is entirely within the agreement, and it can extend your cost exposure well past the effective date — you are now shipping several months of volume at the old price, against the new cost, and the inventory sitting in their DC suppresses reorders for another cycle after that.

Two defences. The first is contractual: negotiate a cap on pre-increase quantities into the agreement, expressed as a multiple of normal velocity. Something like “orders placed during the notice period shall not exceed 130% of trailing three-month average weekly volume” is a reasonable ask and rarely contested at signing, when nobody is thinking about it. Getting it added later is much harder.

The second is operational. Watch order patterns during the notice window against the trailing average, and plan production and cash for the spike, because a forward-buy you did not forecast becomes a service failure at exactly the moment you are asking for goodwill. The order surge is visible in your own data a week or two before it becomes a problem, if anyone is looking at it.

When you sell through a distributor

Everything above describes one notice window. Selling through a distributor gives you two, in sequence.

You notify the distributor, serve your ninety days, and your price to them changes. The distributor then has its own notice obligation to the retailer, and its own submission to prepare, and its own reasons to time that submission conveniently. Meanwhile the shelf price has not moved and the distributor’s margin has compressed, which they will not absorb indefinitely and may simply decline to pass through if the item is not important to them.

Plan for the full sequence rather than the first leg. A cost change may take three quarters to reach a shelf price. Where you have both direct and distributor business in the same category, the direct accounts will reprice first, which creates a period where your distributor channel is your worst-margin channel. Know that going in, and know what it is costing per month, so the decision to keep serving it is a decision rather than a default.

When the answer is no

Sometimes the increase gets rejected and stays rejected. The options at that point are narrower and worth ranking honestly.

Changing the product is usually first: a pack-size adjustment, a reformulation, a specification change on secondary packaging. This preserves the price point and the shelf position, and it works until it becomes visible to the shopper.

Changing the assortment is next. Not every item needs to be at every account. Pulling the worst-margin SKU from the worst-margin account is often better economics than a portfolio-wide fight, and it can be done quietly.

Accepting the compression is a legitimate third option when the account carries strategic weight — distribution that unlocks other buyers, volume that holds your co-packer minimums, category presence you would pay for in other ways. The condition is that you know the number, review it, and have a date on which you revisit it. Compression you have chosen is a strategy. Compression you have not measured is a slow leak.

What sits underneath all three is cost-to-serve. An account’s real contribution includes deductions, chargebacks, freight allowances, trade spend, and the labour of compliance. Brands regularly find that the account they are fighting to hold a price at was contributing less than the reporting suggested, and that the price increase was never the biggest lever available.

What to instrument before you need it

The clause is not the constraint. The constraint is knowing what your cost is, per item, at the moment it changes, and most brands do not.

Three things make the difference between a five-month lag and a two-month one. Landed cost history per SKU per receipt, so you can see a step change rather than a smoothed average. Alerting on component-level cost movement at the point of receipt, so the finding is not a quarterly surprise. And a notice-window calendar per account, so a submission is aimed at a reset date you already know rather than assembled the week someone notices the margin.

None of that is exotic. It is the same purchase order, receipt, and cost data you already generate, held somewhere that keeps its history instead of overwriting it — which is where a system of record earns its keep, and where a spreadsheet rebuilt each quarter quietly stops being one.

The ninety days in that clause is fixed. The five months around it are not.