The Supplier Who Knew We Couldn't Leave

What happened

Corrugated box prices rose 12.5% in 2021 and another 15.9% in 2022, and the brand's plant passed the increases through both years without anyone renegotiating anything. Two years of quiet absorption had taught the supplier that a letter with a number on it was enough. In January 2023 it gave sixty days' notice of an 18% rise at renewal, worth $432,000 a year on a $2.4 million spend, or about 14% of operating profit. The brand stopped drafting a furious reply and counted what it could actually do instead: eighteen days of boxes on hand and tooling sitting in that plant. It signed at 6%, on a thirty-month term, with the price reset quarterly against the published index inside a band.

The case, start to finish

The brand could not walk away and the plant knew it. So it stopped negotiating about price and started negotiating about everything else.

Two years of quiet, which was not neutral

An anonymized composite: a mid-size US e-commerce brand buying custom-printed corrugated shipping boxes from one regional plant. The brand's volumes and the value it assigned to each concession are illustrative. The index moves around them are documented.

Boxes run about $2.4M a year, 1.6 million units at roughly $1.50 each, against $3.1M of operating profit. In 2021 the published index for corrugated boxes rose 12.5% and the plant passed through two mid-year increases. In 2022 the index rose another 15.9% and the brand absorbed that too. Nobody renegotiated anything; the contract rolled every twelve months and the increases felt like weather.

That silence was information. Two years of absorbing had taught the plant precisely how this account responds to a letter with a number in it, which is to say that a letter and a number were the entire cost of raising the price here.

Counting what you can actually do

In January 2023 the plant gave sixty days' notice of an 18% increase at renewal: $432,000 a year, about 14% of the brand's operating profit, priced ahead of even 2022's index move and against a year that had not happened yet.

The instinct is to draft a threatening reply. Instead the brand counted its BATNA, the best thing it could actually do if it walked away, and the count came up short. Eighteen days of boxes on hand. Cutting dies and print plates sitting on the plant's floor. A shortlist of alternative plants close enough that inbound freight would not eat the saving, each needing six to ten weeks and about $40,000 of new tooling to qualify. Corrugated is bulky and light, so freight is a large share of what a box costs delivered, and that single physical fact turns a national market into a local one.

Conclusion: it could not leave, and the plant knew it could not leave. A bluff is not leverage here, it is a costume. This plant knows the account's volume to the pallet, holds the dies, and quotes every brand in the region, including whichever competitor gets named.

Trading what is cheap for what is expensive

So the brand stopped negotiating about price, over which it had no leverage, and started negotiating about the things a box plant genuinely cares about: machine loading, run length, changeovers, forecast quality and how long the commitment lasts. Inside a plant the enemy is idle machine time and job changeovers, and none of those four things costs a customer much to give.

The deal signed in March 2023 was +6%, $144,000 a year, on a thirty-month term. On the illustrative numbers, the longer term was worth about $168,000 a year because it de-risks the plant's own containerboard buying. Retiring one of three box sizes meant longer runs and fewer changeovers on the folder-gluer, about $72,000. A rolling thirteen-week forecast plus accepting mid-week production slots let the plant schedule the account into gaps rather than around them, about $48,000. Together, $288,000 of a $432,000 ask, closed without once threatening to leave.

The price also reset quarterly against the published index inside a ±2% deadband. That clause cuts both ways and will raise the price in a rising market. It also handed the brand something asking would never have produced: when the index finished 2023 down 2.6%, the price came down with it, the first decrease in the seven-year history of the relationship.

The line item that does nothing this year

In April 2023 the brand moved 10% of its volume to a second plant and paid roughly $40,000 for a duplicate set of dies and plates. That 10% saved nothing. It is too small to replace the first plant if the first plant goes down, and running two setups costs money.

It is also the only item in the whole exercise that is about the next negotiation rather than this one. By the 2025 renewal the index was up 6.6% on the year and the plant wanted more. But the conversation opened with a proposal rather than a notice, because a qualified second source was running real volume and both parties knew the dies existed in two buildings.

Leverage is bought months before it is used. Tooling, qualification and inventory cover all have to be paid for while they still look like a pointless expense, because on the day the notice lands it is far too late for any of them to help. And in the meantime, when you cannot credibly leave, go looking for the things that are cheap for you to give and genuinely valuable inside the other side's operation. They are usually sitting there unpriced, because nobody on either side thought to put them on the table.

Timeline

• 2021 The BLS index for corrugated boxes rises 12.5% over the year (339.0 to 381.4). The brand's plant passes through two mid-year increases. Nobody renegotiates anything; the contract is a rolling 12 months and the increases feel like weather.
• 2022 The index rises another 15.9% (381.4 to 442.1). The brand absorbs it again. Two years of absorbing has quietly taught the plant that a letter and a number is the entire cost of raising this account.
• Jan 2023 — Month 0 The plant gives 60 days’ notice of an 18% increase at renewal: on a $2.4M annual spend, $432,000 a year, or about 14% of the brand’s operating profit. The ask is ahead of even 2022’s index move, and it is priced off a year that has not happened yet.
• Feb 2023 — Month 1 The brand stops drafting a threatening reply and counts what it can actually do instead: 18 days of boxes on hand, cutting dies and print plates sitting on the plant’s floor, and a short list of alternative plants close enough that inbound freight does not eat the saving. Conclusion: it cannot leave, and the plant knows it cannot leave.
• Mar 2023 — Month 2 It signs at +6%, or $144,000 a year, on a 30-month term, with the price reset quarterly against the published BLS corrugated index inside a ±2% deadband. It also retires one of its three box sizes, commits to a rolling 13-week forecast, and accepts mid-week production slots.
• Apr 2023 — Month 3 It moves 10% of volume to a second plant and pays roughly $40,000 for a duplicate set of dies and plates. That 10% buys nothing this year. It is the only line item in the whole exercise that is about the NEXT negotiation.
• Q4 2023 The corrugated index finishes the year down 2.6% (442.1 to 430.5). The quarterly reset moves the price down with it, the first decrease in the seven-year history of the relationship, and one the brand would never have received by asking.
• 2025 renewal The index is up 6.6% on the year (430.8 to 459.0) and the plant wants more. This time the conversation opens with a proposal rather than a notice, because a qualified second source is running 10% of the volume and both parties know the dies exist in two buildings.

You're in the owner's chair

The notice is on your desk: 18% in 60 days, on the boxes every order ships in. You have 18 days of stock, your cutting dies are on their floor, and the nearest alternative plant needs six to ten weeks and $40,000 of new tooling. What do you do?

• Tell them you have a competing quote and will move the business
• Accept it and put the 18% into your own prices
• Trade everything but price; quietly qualify a second plant

What each non-price concession was worth

• Term: 12 months → 30 months: 168 $ thousands a year
• One box size retired — longer runs, fewer changeovers: 72 $ thousands a year
• 13-week rolling forecast + mid-week production slots: 48 $ thousands a year

Illustrative only, on a $2.4M annual box spend. Together these closed $288,000 of a $432,000 ask, taking it from +18% to +6%, without the brand once threatening to leave, because it could not.

Business model

A direct-to-consumer brand that ships physical goods: every order that leaves the building leaves inside a printed corrugated box. The box is simultaneously packaging, protection, and the brand’s only physical touchpoint with the customer. That is why it is custom-printed, and why the printing is what makes the supplier hard to replace.

Revenue model

Orders on the brand’s own storefront and two marketplaces. Nothing in this case touches the revenue line; it is entirely a story about a single input cost and who gets to set it. That is the point: the brand’s marketing team spent 2022 arguing about a 2% conversion-rate test while a supplier quietly took 18% of a bigger number.

Cost structure

Boxes are about $2.4M a year (1.6 million units at roughly $1.50 each) against a $3.1M operating profit. Corrugated is bulky and light, so freight is a large share of what a box costs delivered; that is what makes the brand’s realistic alternatives a short list of nearby plants rather than a national market. Inside the plant, the cost drivers are containerboard, machine time, and changeovers between jobs.

Strategic challenge

A weak BATNA, which is the best thing you can do if you walk away. The brand’s was genuinely poor: 18 days of stock, tooling in someone else’s building, and a replacement that takes weeks and money to qualify. Two years of absorbing increases without comment had also removed any signal that the account would ever push back.

Key decision

Stop negotiating about price, which the brand had no leverage over, and start negotiating about the things a box plant actually cares about: machine loading, run length, changeovers, forecast quality, and how long the commitment lasts. Then spend $40,000 on a second source so the next renewal starts from somewhere else.

What worked

Trading what was cheap to give for what was expensive to buy. A 30-month term de-risks the plant’s own containerboard purchasing and was worth about $168,000 a year of the ask. Cutting three box sizes to two meant longer runs and fewer changeovers on the folder-gluer: about $72,000. A 13-week rolling forecast plus accepting mid-week slots let the plant schedule the account into gaps instead of around it: about $48,000. Together, $288,000 of a $432,000 ask, and the index clause turned 2023’s falling market into the brand’s first-ever decrease.

What failed

The two years before the notice. Absorbing 12.5% and then 15.9% without opening a conversation was not patience, it was information: it told the plant exactly how this account responds to a price letter. The brand also discovered its second source too late to help in 2023: qualification takes weeks and dies take money, and both have to be bought before you need them, when they look like a pointless expense.

Risk factors

A single plant holding the tooling for every SKU; inventory cover measured in days rather than weeks; an index clause that cuts both ways and will raise the price in a rising market; a second source at 10% that is too small to actually replace the first if the plant goes down; and containerboard allocation in a tight market, where the goodwill you spent on a bluff is the thing that decides who gets shipped.

Lesson summary

A weak BATNA does not stop you negotiating. It stops you negotiating about price. When you cannot credibly leave, the move is to find the things that cost you little and are worth real money inside the other side’s operation, such as term, forecast, SKU count and scheduling, and trade those instead. Then buy your next BATNA while you still have time, because leverage is bought months before it is used.

Key data

• $2.4M (1.6M boxes at ~$1.50) Annual corrugated spend
• +18% — $432,000 a year Supplier's renewal ask
• About 14% of a year’s operating profit What that ask was worth
• 18 days Boxes on hand when the notice arrived
• ~$40,000 of dies and print plates Cost of a second source
• +6% — $144,000 a year, on a 30-month term What was signed
• $288,000 a year Closed by non-price trades
• +12.5% (2021), +15.9% (2022), −2.6% (2023) BLS corrugated box index

Sources & basis

The business in this story is a stand-in, not a company you can look up. This case is an illustrative composite: the operator, the people and most of the dollar figures represent a pattern rather than reporting one firm's history. What the list below cites is the other half, the documented industry data and public reporting the composite was assembled from, including any real company whose published figures the case draws on by name. The mechanism and the arithmetic are real even where the business is not.

1. BLS producer price index, corrugated and solid fiber box manufacturing (PCU322211322211) — annual averages 2019–2025, BLS public API View source ↗
2. Composite pattern — the brand, its volumes, and the value assigned to each concession are illustrative; the index moves are not.