How long should a packaging quote stay valid during a price cycle?

There is no packaging-specific standard, and the widely used thirty-day period is borrowed from general B2B practice rather than derived from anything about board. The useful way to set it is against the cost basis rather than the calendar: a quote should expire when the input price it was built on stops being the price you can buy at. In a year with three announced increases and an implementation lag running into the following year, that is a shorter window than thirty days for board-heavy work and can be longer for jobs where conversion cost dominates.

If an increase is announced for September, is my September cost the new number?

Usually not. There is a documented lag between an announced increase and the price the market actually transacts at. Fastmarkets RISI recognised a net one hundred dollars per ton across the first two rounds of 2026, and Packaging Dive reported in August that implementation of the third round, effective in September, will likely continue into early 2027. Treat the announced figure as the producers' opening position and the recognised figure as the market's answer, because your board invoice follows the second one.

Should we add a price-escalation clause instead of shortening validity?

They solve different problems and most converters need both. A shorter validity limits how long you are exposed on a quote that has not converted yet. An escalation clause covers the period after acceptance, when the job is booked and the board is not yet bought. A clause is also a commercial negotiation with the customer, whereas validity is yours to set unilaterally, which is why validity is the faster of the two to change.

We cannot reprice everything. Where do we start?

Start by listing rather than repricing. The first question is which quotes are outstanding, what board cost each was built on, and which of those are still likely to convert. That list is usually shorter than people expect, and it is the difference between repricing a handful of live jobs and issuing a blanket increase that costs you the accounts you were winning. If producing the list means opening spreadsheets one by one, that is the problem to fix first.

Announced Is Not Recognised: Board Prices and Your Open Quotes

Containerboard producers announced three price increases in five months in 2026, and the third landed while the first two were still being absorbed. But the number a producer announces and the number the market pays are not the same number, and they are separated by months. Fastmarkets RISI has recognised a net one hundred dollars per ton across the first two rounds; the third round, effective on the first of September, is expected to keep implementing into early 2027.

That gap is where a converter's real exposure sits, and it is not in the quotes you are about to send. It is in the ones already out. Every quotation currently outstanding was priced on a board cost from the day it was built, and your packaging quote desk has almost certainly not been asked which ones those are. This post is about that question rather than about the price of board, which you already watch.

Three Increases in Five Months

The sequence is unusual enough that the independent converters' own association put it in writing. AICC, The Independent Packaging Association, published a statement on 11 August 2026 opposing the third round, and the statement is the clearest public account of the year's shape: three price increase announcements by the majority producers in five months, most recently Packaging Corporation of America's one hundred and forty dollars per ton on liner and medium effective 1 September, International Paper's eighty dollars, and Smurfit Westrock's hundred. As AICC noted, those three announcements all took place within three days of each other.

Georgia-Pacific, ND Paper and White Birch Paper joined afterwards at a hundred to a hundred and ten dollars per ton, and Cascades matched the hundred and forty on medium, according to Packaging Dive's August pricing brief. About ten percent of US containerboard capacity was permanently retired between February 2025 and March 2026, which is the supply tightness producers cite. AICC, for its part, calls the increases unjustified. We are not going to adjudicate that, and a converter does not need it adjudicated to price a job. What matters commercially is that the announcements arrive in steps, close together, and while the previous step is still being absorbed.

What the Index Has Actually Conceded

This is the part that changes how you should quote, and it is the part that never makes the headline. The announced figure is the producers' opening position. The recognised figure is what the published index concedes the market is actually paying, and in 2026 the two have been a long way apart.

Third-round announcements, and separately what the index has recognised

PCA, announced for 1 September$140

Cascades, medium, announced$140

Smurfit Westrock, announced$100

Georgia-Pacific, ND Paper, White Birch, announcedplotted at the midpoint of the announced range$100 to $110

International Paper, announced$80

Recognised so far in 2026, cumulative across the two earlier rounds, not comparable to the bars abovea whole year of recognised movement, not a third-round figure$100

Announcements per AICC, 11 August 2026, and Packaging Dive, 25 August 2026. Recognised net year-to-date figure per Fastmarkets RISI via Packaging Dive, 25 August 2026. The five upper bars are single-round announcements for September. The accented bar is a different quantity: the cumulative movement the index has conceded across the two earlier rounds of the year, shown for scale rather than for comparison.

AICC put the mechanism plainly: producers announce large increases "with the hopes of ultimately obtaining a meaningful, but lesser rise in the transacted prices. In essence, they are testing the resolve of the market." The association also observed that the March and June increases were, as of August, still "not yet fully absorbed into the marketplace, as stated by the suppliers themselves, and confirmed by many converters", and that the two published indexes in the market "have not agreed, been in parallel, or even close to each other all year."

So a converter pricing work this autumn is working without a single authoritative input cost. Not because they are disorganised, but because the market has not produced one. Meanwhile box demand is not helping: the Bank of America Securities box report released on 11 August expects flat growth over the following two quarters, down from half a percent expected in June.

Your Exposure Is the Quotes Already Out

A price cycle does not primarily threaten the jobs you have not quoted. Those you can price at today's cost, and every converter does. It threatens the quotations sitting in customers' inboxes, priced weeks ago, still nominally open, and increasingly likely to be accepted precisely because they are now cheap. In a rising market your stale quotes get more likely to convert, not less: a buyer holding a July number and facing September pricing everywhere else has an obvious reason to act on it. The selection runs against you automatically.

The uncomfortable question, then, is not what board costs. It is: how long would it take you to produce a list of every open quotation, the board cost each was built on, and which are still live? For most converters the honest answer is hours, because that information is not in one place. It is in the estimator's spreadsheets, in sent email, and in whichever of those two the customer is actually holding.

We know the shape of that answer because we used to give it. Before we wrote any of this software we ran a packaging business, and a price move meant one person reconstructing the open quote book from sent mail and a spreadsheet, then guessing at which quotes the customer still considered live. The reconstruction was slow enough that the practical decision was usually made without it: reprice everything and annoy the accounts you were winning, or reprice nothing and absorb the difference. Both are bad, and both are what you choose when the list takes a day to produce. That is the failure this argument comes from, and it is ours rather than a hypothetical. If you are weighing whether the fix is a full packaging ERP or a faster quote desk, it is a useful diagnostic: the converter who cannot list their open quotes by cost basis does not have a production-control problem.

Thirty Days Is a Convention, Not a Decision

Most quotes carry a validity period, and it is usually thirty days. Search for where that number comes from and you find general B2B and contracting advice, which is refreshingly honest about the number's status: it is a convention rather than a law, but a widely followed one. We could not find packaging-specific guidance on the question at all, only that general advice. The number appears to be inherited rather than chosen.

But a validity period is a risk decision, and it should follow the cost basis rather than the calendar. The right window ends when the input price the quote was built on stops being the price you can buy at. In a year with three announced steps and an implementation lag measured in quarters, that is not a constant. It is shorter on board-heavy corrugated work than on a job where diecutting, finishing and assembly carry most of the cost. The same footer line cannot be correct for a commodity brown box and for a rigid box with foil, and thirty days applied to both is wrong in two directions at once.

What a Quote Has to Carry

Which brings the argument to where we are partisan, because this is a software problem and we build the software. A quote is usually treated as a document: a PDF, sent, filed, and then effectively gone. If a quotation is only a document, every price cycle becomes an archaeology exercise across spreadsheets and inboxes.

If a quotation is a record, they become a filter. That means the price basis is an attribute of the quote rather than a memory: the version of the pricing rules it used, and the cost inputs those rules read at the time. In Packative One, pricing runs on versioned playbooks and every calculation keeps a step-by-step audit trail, so a quote can answer what it was priced on rather than only what it priced. Expiry is a property of the record too, with automation attached, rather than a sentence in a footer nobody acts on. And because quotations hang off the customer and the case in the account record, "what is outstanding on this account, priced when" is a question the system answers rather than a person.

That is the whole claim, and it is deliberately narrow. None of it changes what board costs. What it changes is the time between an announcement landing and you knowing your exposure to it, which in a year with three announcements is not a small thing. The same argument runs underneath why response speed decides packaging RFQs, and a price cycle is the least forgiving version of that test.

Three increases in one year is unusual but not unprecedented, and AICC notes it also happened in 1994 and in 2010. What is different now is that the market's answer to an announcement takes months to arrive and two published indexes disagree about it the whole time. Quoting into that with a thirty-day footer and a folder of spreadsheets is a choice, and it is worth making deliberately. If you would rather it were a filtered list, that is what a quote desk built on versioned pricing is for.

자주 묻는 질문

: There is no packaging-specific standard, and the widely used thirty-day period is borrowed from general B2B practice rather than derived from anything about board. The useful way to set it is against the cost basis rather than the calendar: a quote should expire when the input price it was built on stops being the price you can buy at. In a year with three announced increases and an implementation lag running into the following year, that is a shorter window than thirty days for board-heavy work and can be longer for jobs where conversion cost dominates.
: Usually not. There is a documented lag between an announced increase and the price the market actually transacts at. Fastmarkets RISI recognised a net one hundred dollars per ton across the first two rounds of 2026, and Packaging Dive reported in August that implementation of the third round, effective in September, will likely continue into early 2027. Treat the announced figure as the producers' opening position and the recognised figure as the market's answer, because your board invoice follows the second one.
: They solve different problems and most converters need both. A shorter validity limits how long you are exposed on a quote that has not converted yet. An escalation clause covers the period after acceptance, when the job is booked and the board is not yet bought. A clause is also a commercial negotiation with the customer, whereas validity is yours to set unilaterally, which is why validity is the faster of the two to change.
: Start by listing rather than repricing. The first question is which quotes are outstanding, what board cost each was built on, and which of those are still likely to convert. That list is usually shorter than people expect, and it is the difference between repricing a handful of live jobs and issuing a blanket increase that costs you the accounts you were winning. If producing the list means opening spreadsheets one by one, that is the problem to fix first.