The Cannon Byzantium Didn't Buy: The Oldest Case for Early Supplier Involvement

The best suppliers, including the ones whose quotes make you wince at first glance, are not just bringing you goods, but innovations. Procurement has a name for the discipline of acting on that fact -- early supplier involvement, or ESI: pulling suppliers into product development at the concept stage, while the design is still wet. I spent my consulting years at McKinsey around supply chains, and I now lead deployments at LightSource, helping procurement organizations put AI to work on their sourcing; between the two jobs I have watched ESI separate the teams that merely buy well from the teams that build better products.

Outside of work, I am obsessed with Byzantine history. I promise these two facts are related.

Byzantium Ran a Thousand-Year Supply Chain

The Byzantine Empire was the Roman Empire, continued in the East after the West fell, and it lasted for more than eleven centuries -- longer than any other state in the neighborhood before or since. It endured for many reasons, but the big one is geography: Constantinople sits at the narrow water crossing between Europe and Asia, which means that for a thousand years, more or less everything moving between the Mediterranean world and the Black Sea passed beneath its walls and paid for the privilege, typically a customs cut of around 10 percent. Trade paid for the armies, the diplomats, and Hagia Sophia, which went up in five years and stayed the largest cathedral on earth for nearly a millennium. Control the flow, tax the flow, and you can fund a civilization.

That playbook did not retire with the empire, by the way. About a fifth of the world's oil still squeezes through a single strait at Hormuz, and this summer showed what happens when that flow gets pinched -- my colleague Sourav Das Adhikari wrote about the shockwaves that followed. But chokepoints are a different essay. This one is about a sales call.

Hagia Sophia, built in five years and the largest cathedral on earth for nearly a millennium. Trade paid for it. Photo: Arild Vagen, Wikimedia Commons (CC BY-SA 3.0).

In 1452, a Supplier Offered Constantinople the Future

A year before the Ottomans attacked the city, a Hungarian ironmaster named Orban arrived in Constantinople with something to sell. He was a master founder of bronze cannon, and what he offered Emperor Constantine XI was, without exaggeration, the most advanced weapons technology on earth: siege guns bigger than anyone had ever cast. The emperor wanted them. He also could not afford them. The treasury of the thousand-year empire could not meet Orban's salary, and the city could not even supply the raw materials the work required.

So Orban did what good suppliers do when the incumbent passes. He crossed over to Edirne and pitched the competition: Mehmed II, the 21-year-old sultan who was openly preparing to besiege Constantinople. Orban claimed his gun could shatter "the walls of Babylon itself." Mehmed gave him everything he asked for -- generous pay, abundant bronze, whatever he needed -- and about three months later the great bombard existed: a monster roughly 27 feet long that fired stone balls weighing hundreds of pounds. Sixty oxen dragged it to the walls of Constantinople, and Orban cast a whole family of smaller guns to travel with it. The Theodosian Walls had turned away every land army for a thousand years. Against the new technology, they held for 53 days. The city fell on May 29, 1453, and the Roman Empire ended with it.

Historians will remind you that the story leans on a chronicler who loved a good moral, and that Mehmed's real advantage was institutional -- money, materials, foundries, and logistics rather than one man's genius. That is all fair. But as an allegory for what happens when the incumbent sends the innovator across the street to the challenger, the story has never been improved on. A civilization eleven centuries old was brought down, in part, by a vendor it had turned away over price.

The Dardanelles Gun, a 1464 Ottoman bronze bombard cast in the same tradition as Orban's siege gun, at Fort Nelson, UK. Photo: Wikimedia Commons (public domain).

More Than Half of Innovation Arrives Through the Supplier Door

The uncomfortable truth inside Orban's story is that the innovation belonged to the supplier, and that is not the exception in industrial history. It is closer to the rule. McKinsey research has found that more than half of the innovations companies adopt originate outside the organization -- from suppliers, research institutions, and customers. In automotive, suppliers account for somewhere between 70 and 80 percent of a vehicle's value, depending on whose estimate you use; GM itself has said that nearly 70 percent of the value of a car is created outside its own walls. The company logo goes on the finished product, but a great deal of the technology underneath it walked in through the supplier door.

My favorite modern example runs almost beat-for-beat parallel to Orban. In 2007, Steve Jobs needed a scratch-proof glass screen for the first iPhone -- the prototype's plastic face was getting chewed up by the keys in his pocket -- and he needed it in about six months. Corning's CEO, Wendell Weeks, explained that his company had the underlying chemistry, descended from a strengthened-glass project shelved back in the 1960s, but no product and no capacity; developing one normally took close to two years. Jobs' answer, in Walter Isaacson's telling, was to insist: don't be afraid, you can do this. Corning developed Gorilla Glass in about four months, converted a Kentucky plant to make it, and it shipped on the iPhone that June. The supplier had been sitting on the defining feature of the product for forty years, waiting for a buyer who would fund the sprint.

It is happening right now in semiconductors, at the biggest price tag a supplier has ever put on a machine. ASML's High-NA lithography tools cost around $400 million each, and TSMC, the industry's incumbent, has said publicly that they are too expensive to adopt this decade. Intel bought the first one instead, and as of this summer it is shipping processors built with it in high volume. Whether that bet pays off is genuinely uncertain -- deferring an expensive tool is a defensible economic choice, not a blunder. But the shape of the moment is familiar: the supplier brings the future at an eye-watering price, the incumbent passes, and the challenger writes the check. My colleague Spencer Penn has written about the 5,100-supplier machine behind that story. Orban would recognize every character in it.

Early Supplier Involvement Is How You Buy the Cannon

Which brings me back to the bland-sounding idea I opened with. Early supplier involvement just means bringing your key suppliers into product development at the concept stage -- into the design conversations, the make-or-buy calls, the target costing -- instead of handing them a finished drawing and asking for a price. The timing matters because of a rule of thumb every manufacturing engineer knows: by the time a design is locked, somewhere around 70 to 80 percent of the product's eventual cost is already committed. The number is a heuristic, not a law of physics, but the direction is beyond dispute, and it means the most valuable thing a supplier can give you is not a discount on your design. It is a better idea about the design itself, delivered while the design can still change.

The research backs the instinct. Studies of supplier integration into new product development have found double-digit improvements in cost, quality, and cycle time when it is done well, and Toyota and Honda built exactly this into their supplier relationships for decades: setting cost targets together with suppliers and engineering toward them jointly, rather than discovering the gap at quote time. Chrysler ran the same play in 1989 when it gathered its 25 biggest suppliers and asked for their ideas instead of their margins; the program that followed saved billions. And to be honest about the fine print: the benefits do not accrue automatically. ESI done carelessly can leak IP, entrench a sole source, and slow a program down with unmanaged meetings, which is why the practitioners' guides -- CIPS has a good one, and our friends at Dassault Systemes have written one too -- spend as much time on governance as on enthusiasm.

The novice version of procurement treats suppliers as a price list and negotiation as the job. I understand the appeal; haggling produces a number you can put on a slide the same week. But the transformational value in this profession comes from somewhere else entirely: from the deep interplay between procurement, engineering, and suppliers, run early enough that the good ideas land before the design freezes. That interplay is most of what I do at LightSource. The platform's whole premise is connecting those three parties in one system, so that when we deploy with a manufacturer, the supplier with the better idea surfaces during NPI sourcing, next to the quotes, while the answer can still change the product.

Constantine found out what his supplier could build by watching it from the wrong side of the wall. Early supplier involvement exists so you can find out in a conference room instead, a year earlier, while the cannon can still be yours.

Sources

• Donald Nicol -- The Fall of Constantinople (excerpt) -- Orban's offer to Constantine XI, the unaffordable salary, and his move to Mehmed II

• HistoryNet -- The Guns of Constantinople -- the "walls of Babylon" boast and the casting of the great bombard at Edirne

• Wikipedia -- Fall of Constantinople -- siege timeline and the end of the empire, May 29, 1453

• Fast Company -- How Corning's Crash Project for Steve Jobs Helped Define the iPhone -- the six-month glass sprint of 2007

• Corning -- Gorilla Glass FAQs -- the corrected history: 1960s Chemcor chemistry, Gorilla Glass developed in 2007

• ASML -- High-NA EUV reaches new readiness milestone (July 2026) -- Intel in high-volume manufacturing on High-NA EUV

• TrendForce -- Behind TSMC's High-NA EUV deferral -- TSMC's cost-driven decision to defer the ~$400M tools

• McKinsey -- Procuring innovation, wherever it may be -- more than half of adopted innovations originate outside the organization

• McKinsey -- Taking supplier collaboration to the next level -- collaborative buyers outgrow and outearn their peers

• Petersen, Handfield & Ragatz -- A Model of Supplier Integration into New Product Development (JPIM, 2003) -- measured cost, quality, and cycle-time gains from supplier integration

• Liker & Choi -- Building Deep Supplier Relationships (Harvard Business Review, 2004) -- how Toyota and Honda co-develop with suppliers

• BCG -- Buyer-Supplier Collaboration: A Roadmap for Success -- Chrysler's 1989 "brainpower, not margins" supplier program

• NIST -- on design-stage cost commitment -- the 70-80 percent cost-committed-at-design estimate

• AP -- Strait of Hormuz coverage, August 2026 -- the strait that carries about a fifth of the world's oil

• CIPS -- Early Supplier Involvement -- ESI definition, stages, benefits, and risks

• Dassault Systemes -- What is Early Supplier Involvement (ESI)? -- a practitioner's guide to getting started with ESI

Frequently Asked Questions

What is early supplier involvement (ESI)?

Early supplier involvement is the practice of bringing key suppliers into product development at the concept and design stage, rather than after the design is finished. Suppliers contribute manufacturability feedback, alternative materials and technologies, cost insight, and sometimes the defining innovation itself, while the design can still change. It is most common in industries like automotive and consumer electronics, where suppliers create most of the product's value.

Why does early supplier involvement matter so much for cost?

Manufacturing engineers commonly estimate that 70 to 80 percent of a product's eventual cost is committed by the time the design is locked, even though most of the money has not been spent yet. That figure is a long-standing heuristic rather than a precise law, but its direction is well supported. It means supplier ideas delivered during design are worth far more than discounts negotiated afterward.

Who was Orban, and what does his cannon have to do with procurement?

Orban was a Hungarian master founder who offered to build enormous siege cannons for the Byzantine emperor Constantine XI in 1452. The emperor could not afford his salary or the materials, so Orban went to the Ottoman sultan Mehmed II, who funded him generously; his great bombard helped bring down Constantinople's walls in 1453 after they had stood for a thousand years. The story is procurement's oldest cautionary tale about turning away a supplier who is offering innovation rather than just goods.

What share of innovation comes from suppliers?

McKinsey research has found that more than half of the innovations companies adopt originate outside the organization, with suppliers as a leading source. In automotive, suppliers are estimated to create 70 to 80 percent of a vehicle's value. Landmark examples of supplier-delivered innovation include Corning's Gorilla Glass for the first iPhone and ASML's High-NA lithography machines in semiconductors.

What are the risks of early supplier involvement?

The main risks are intellectual-property leakage, over-dependence on a single supplier whose technology gets designed in, and slower development if collaboration is unmanaged -- research shows the benefits do not accrue automatically. Successful ESI programs pair the collaboration with governance: clear IP agreements, trusted suppliers with proven track records, and defined decision rights between procurement, engineering, and the supplier.