AI Implementation

Freight Costs Are Moving Weekly: Keeping UAE Supplier Quotes Accurate During the Hormuz Disruption

Maersk gave six days of notice on a USD 500 per container surcharge into the UAE. If your quote validity is thirty days, most of your open quotes were already priced on a cost you could no longer buy at. How to keep landed cost, quote validity and margin in line while freight moves weekly.

September 29, 202611 min readBearingNorthAI

Cargo ships and stacked shipping containers at a port terminal during the day — Photo by Andy Li on Unsplash Source

On 9 September 2026, Maersk published a notice that most UAE importers will have seen as a routine line in an inbox. An Emergency Operational Cost Recovery surcharge of USD 500 per container, applied to dry, reefer and special equipment moving from the world excluding Far East Asia into the UAE, effective 15 September for most origins and 9 October for regulated trades. Six days of notice. If your standard quote validity is 30 days, roughly two thirds of the quotes sitting in your buyers' inboxes that morning were already priced on a cost you could no longer buy at.

That is the real shape of the problem for UAE suppliers, traders and distributors right now. It is not that freight is expensive, although it is. It is that the landed cost underneath a quote has a shorter shelf life than the quote itself. Every week the gap between the number in your price list and the number on your supplier invoice moves, and every quote issued against a stale number is either a lost deal or a lost margin. Which one depends purely on which direction the cost moved.

This is a pricing and process problem, not a shipping problem. You cannot fix the Strait of Hormuz. You can fix how fast your own quotes reflect what a container actually costs you today.

What actually changed in the cost of landing a container in the UAE

The disruption around the Strait of Hormuz has been running since the end of February 2026, and the cost effects have been unusually large even by the standards of recent supply chain shocks. Freightos, tracking the route in its running assessment of the Hormuz shipping impact, reported container rates from Shanghai to Jebel Ali moving from under USD 2,000 to above USD 8,000 per container since the start of the conflict, emergency fuel surcharges of USD 200 to USD 500 per container across lanes, and port waiting times of seven to ten days at some UAE ports.

Insurance moved on a similar scale. The National reported on 17 July 2026 that war risk premiums for Gulf transits had reached 3 to 10 percent of hull value against a pre-war baseline of 0.25 percent, which turns a roughly USD 250,000 cost on a USD 100 million vessel into USD 3 million to USD 10 million for a single voyage. Marcus Baker of Marsh, speaking to AGBI about whether Gulf cover was still obtainable, made the point that matters most for anyone pricing goods: war rates have been on a roller coaster mirroring the oil price. Not a step change that you absorb once and move on from. A moving number.

The volume picture explains why routing costs keep shifting too. Jebel Ali handled 3.14 million TEU in the first half of 2026 against 7.77 million in the same period of 2025, and fell from tenth to thirty-second in the global port rankings in six months, according to Splash247 reporting in August. DP World has moved around 500,000 TEU through alternative Gulf routes since March and built bonded corridors from the Gulf of Oman ports inland. That is a genuine mitigation, and it is also a different cost base, with trucking legs and handling charges that did not exist in your price library a year ago. The operators running those corridors are the same ones we looked at in our piece on how AI is already being used across UAE ports and maritime logistics, and their own systems are adapting faster than most of their customers' pricing is.

Costs are not only going up, which is the harder problem

If freight only ever rose, this would be simple. You would add a buffer, quote high, and take the occasional lost bid as the cost of safety. The reason that does not work in 2026 is that the direction keeps changing.

In April 2026, at the worst of it, the S&P Global UAE PMI recorded the sharpest rise in overall input costs since July 2024, driven by oil and transport, with non-oil firms raising their own selling prices at the fastest rate in almost fifteen years. By August the picture had turned. The headline PMI reached 55.3, the strongest non-oil growth since December 2024, and input cost inflation fell to its lowest level since February, even though firms still reported higher prices for energy, fuel, cement, steel and chemicals. S&P Global's David Owen noted that UAE businesses were actively building resilience by switching to domestic suppliers.

Read those two readings together and the commercial risk becomes obvious. A quote built at April assumptions and honoured in September is priced above the market and loses to a competitor who repriced. A quote built at September assumptions and honoured after the next surcharge notice is priced below cost and wins business you would rather not have won. Both failures come from the same root cause, which is a quote whose cost assumptions were fixed at a point in time and never revisited.

Meanwhile the underlying trade volume is not shrinking. UAE non-oil foreign trade reached AED 1.937 trillion in the first half of 2026, up 13.1 percent year on year, with non-oil exports at a record AED 452.8 billion. There is more to quote for, not less. The constraint is the speed at which a firm can put an accurate number in front of a buyer.

Where the margin actually leaks

In most UAE trading and supply firms, the leak is in one of five places, and none of them are dramatic enough to show up in a board pack until the year end margin comes in short.

The freight, duty and handling line in the price library is updated on a schedule, usually monthly or quarterly, rather than when the cost changes. A surcharge announced with six days of notice will never be caught by a monthly refresh.
Quote validity is longer than the notice period on the charges underneath it. Thirty days of validity against six days of surcharge notice is a structural mismatch, not bad luck.
Landed cost is baked into a single unit price, so when one component moves, the only way to revise is to rebuild the whole quote. Rebuilds are slow, so they do not happen.
Re-quoting is manual, so it is rationed. The estimator revises the three largest open quotes and leaves the other forty, which are collectively worth more.
Nobody records which cost assumption a given quote was built on. When the buyer comes back six weeks later to place the order, there is no way to tell whether the number is still safe without redoing the work.

The fifth one is the most expensive and the least visible. A quote without a recorded assumption set is not really a quote, it is a guess with a letterhead. When it converts, you find out what it was worth at the invoice stage.

What a live price library changes

The fix is structural and fairly ordinary. Separate the cost of the goods from the cost of getting the goods here, hold both in one place that is updated as costs change rather than on a calendar, and make revision cheap enough that every open quote gets revised rather than just the big ones.

That is the job an agentic quoting system does. In practical terms, a tool like ZentraBid reads the tender or enquiry documents, matches the requested line items against the firm's own price library, and produces a priced draft for an estimator to review, with the freight, duty, insurance and handling components held as separate, dated lines rather than absorbed into a unit price. When the freight line moves, every open quote carrying that line can be reissued in minutes, with a clear record of what changed and why. If the terms retrieval augmented generation and agentic are still doing heavy lifting in the proposals you receive, our explainer on what RAG and agentic AI actually mean for UAE businesses sets out the five questions that separate a working system from a rebranded chatbot.

The capacity argument is the same one we made about tender volume in how UAE trading firms can answer more tenders without adding estimators, but the failure mode here is different and worth stating plainly. That article was about bids you never got to. This one is about bids you did send, at a price that was already wrong when it arrived.

Setting a validity period you can actually honour

Three changes to how a quote is written will do more for margin in the next quarter than any amount of negotiation.

First, shorten validity and say why. Seven to fourteen days on anything with an ocean freight component is defensible in the current market, and buyers in the UAE are living through the same disruption. A stated reason converts what looks like a hardening of terms into evidence that you know your own cost base.

Second, name the assumption on the face of the quote. A single line stating the freight rate, war risk allowance and surcharge position the price is built on, with the date it was set, does two things. It makes a later revision a factual conversation rather than a renegotiation, and it gives your own team a record to check against when the order lands weeks later.

Third, write a surcharge pass-through clause covering charges announced by the carrier after the quote date, and itemise those charges so the pass-through is auditable. Carriers give days of notice. Your contract terms need to acknowledge that reality rather than pretend the number is fixed.

A thirty day plan

Week one: pull the last sixty days of supplier invoices and compare the landed cost actually paid against the landed cost assumed in the quote that won each order. The distribution of that gap is your real exposure, and it is usually wider than anyone expects.
Week two: split landed cost into named components in the price library. Goods, ocean freight, war risk and insurance, surcharges, inland leg, duty, handling. If these live in one blended unit price today, this is the single highest value change on the list.
Week three: set validity periods by component volatility rather than by habit, and add the assumption line and pass-through clause to the quote template.
Week four: instrument the revision path. Whether that is an agentic quoting tool or a disciplined manual process, the test is simple. When a carrier announces a surcharge on a Tuesday, how many open quotes carrying that lane are reissued by Thursday?

If the answer to that last question is a number you would not want to say out loud, that is the gap worth closing first.

Measuring whether it worked

Three numbers tell you most of what you need. Realised gross margin against quoted gross margin, measured per order rather than in aggregate, because aggregate hides the losses under the wins. Median hours from cost change to quote reissue. And win rate on revised quotes against win rate on quotes that were never revised, which tells you whether repricing is costing you deals or simply protecting the ones worth having.

Set those baselines before you change anything, because after go-live you will not be able to separate the effect of the system from the effect of the market. Our guide to measuring ROI from an AI implementation sets out how to establish that baseline and how to keep the attribution honest when several things are moving at once.

The point

The Hormuz disruption will resolve on its own timetable, and nothing a UAE supplier does will change that. What the last seven months have exposed is a quieter weakness that was always there and simply did not cost much when freight was stable. Most firms price from a library that updates slower than their costs do, and issue quotes that outlive the assumptions behind them.

Firms that fix this will keep the margin when the market turns quiet again. Firms that do not will go on discovering the size of the gap one invoice at a time.

Research sources used

FAQ

Common questions.

How short should quote validity be for UAE imports right now?

For anything with an ocean freight component, seven to fourteen days is defensible in the current market. Carriers have been announcing surcharges with under a week of notice, as Maersk did on 9 September 2026 for a charge effective 15 September, so a thirty day validity leaves you exposed to charges you cannot yet see. The more important change is to state the cost assumption the price is built on, so a later revision is a factual conversation rather than a renegotiation.

Is freight into the UAE still rising, or has it peaked?

Both, depending on the week and the lane. S&P Global's UAE PMI for April 2026 recorded the sharpest input cost rise since July 2024, while the August 2026 reading showed input cost inflation at its lowest since February, with firms still reporting higher prices for energy, fuel, cement, steel and chemicals. That two way movement is exactly why a static price library is dangerous. A quote priced at the peak loses the deal, and one priced at the trough loses the margin.

Can an AI system price a quote without an estimator reviewing it?

It should not, and a serious vendor will not claim otherwise. The value is in the preparation: reading the enquiry, matching line items against the price library, pulling current cost components and producing a priced draft with the assumptions shown. The estimator then reviews and approves. What changes is not whether a person decides the price, but how many quotes one person can get through and how quickly an open quote can be revised when a cost moves.

What is the first thing to fix if we only have time for one change?

Split landed cost into named components in your price library rather than holding it as a single blended unit price. Almost every other improvement depends on it. Once freight, war risk, surcharges, inland leg, duty and handling are separate dated lines, you can revise one component without rebuilding the quote, you can write an auditable pass-through clause, and you can tell after the fact which assumption was wrong.

Does this apply to a firm with fewer than fifty staff?

It applies more, not less. A larger firm can absorb a handful of underpriced orders across a wide book. A firm of twenty to fifty people running on a small number of large orders cannot, and typically has one or two people doing all the quoting, which means manual re-quoting gets rationed to the biggest files. The smaller the estimating team, the larger the share of open quotes that never get revised at all.