Nike’s Turnaround May Require a New Manufacturing Strategy
The case for building an advanced manufacturing platform that combines American research and development, regional responsiveness and Nike's global supplier network.
As this article is being written, Nike shares are trading at $35.78.
That is their lowest level in more than a decade and approximately 78% below the company's 2021 high. The decline has erased close to $200 billion in market value. (MarketWatch, Macrotrends)
The market has taken a clear position:
Nike's turnaround is not moving quickly enough.
That does not mean Nike is broken.
The company generated $46.4 billion in fiscal 2026 revenue. It remains one of the most recognizable brands in the world, maintains enormous global distribution and ended the year with approximately $9 billion in cash and short-term investments. (Nike FY2026 Form 10-K)
But Nike is no longer being valued as an untouchable growth company.
It is being valued as a company that has lost momentum and has yet to prove it can regain it.
How Nike Got Here
There was no single decision that created Nike's current position.
The company shifted aggressively toward direct-to-consumer sales, reducing its presence with wholesale partners that had helped distribute and merchandise the brand. It leaned heavily on established franchises while competitors introduced products that felt newer and more specialized.
Innovation slowed in categories where Nike had historically set the pace. Greater China weakened.
Fashion cycles accelerated. Running fragmented across brands such as Hoka, On, Brooks and ASICS. Premium consumers found alternatives in companies such as Vuori and lululemon.
Nike did not suddenly become irrelevant.
The market around it became more competitive while the company became less responsive.
Elliott Hill's turnaround is attempting to repair that. Nike is rebuilding wholesale relationships, reorganizing around sport, reducing its dependence on promotional sales and trying to restore a product pipeline capable of generating visible consumer excitement.
Those are necessary moves.
A better manufacturing system cannot make an undesirable shoe desirable. Automation cannot restore cultural relevance. Robotics cannot rebuild a wholesale relationship or persuade a consumer in Shanghai to want a Nike product.
But the turnaround also raises a less visible question.
Is Nike trying to restore growth using a production architecture designed for a slower and more predictable market?
The Overlooked Part of Nike's Turnaround
Most discussion about Nike's recovery focuses on the parts of the company consumers can see:
Products.
Athletes.
Advertising.
Retail.
Digital commerce.
Brand.
Beneath all of them sits an enormous global production system.
Nike generally commits to product four to five months before it expects to sell it.
At the end of fiscal 2026, the company was carrying approximately $7.5 billion in inventory and had another $4.9 billion in legally binding product-purchase obligations. Nearly all its footwear and apparel was being manufactured by independent contractors, predominantly in Asia. (Nike FY2026 Form 10-K)
That architecture helped Nike become the largest athletic brand in the world.
It provides scale, manufacturing expertise, supplier density and economics that would be extremely difficult to recreate in the United States.
But it also requires Nike to make enormous commitments before it knows exactly what consumers will want.
Months before a product reaches the market, Nike must decide:
How many units?
Which colors?
Which sizes?
Which markets?
Which channels?
If the forecast is wrong, the company can be left with excess inventory and markdown pressure.
If a product unexpectedly succeeds, Nike may not be able to replenish it while demand is still at its peak.
That matters more in a market where consumer attention moves quickly, specialized competitors can capture individual categories and social media can accelerate and exhaust a trend within months.
Manufacturing did not create Nike's current problems.
But manufacturing architecture influences how quickly the company can respond to them.
It determines how far in advance Nike must predict demand, how much inventory risk it accepts, how quickly it can replenish a winning product, how expensive it is to test an idea and how exposed the company remains to tariffs, freight costs and geopolitical disruption.
That is the aspect of Nike's turnaround this article will explore.
Nike Should Not Reshore Manufacturing. It Should Reinvent It.
The proposal is not for Nike to close factories in Vietnam and recreate conventional versions of them in Oregon.
That would almost certainly destroy value.
Nike should instead develop a second manufacturing mode: an advanced manufacturing platform combining American research and development, selective regional production and the scale of its existing global supplier network.
The United States could become the place where Nike develops and proves new manufacturing technologies.
Regional production could allow selected products to be replenished closer to demand.
Successful automation systems could then be deployed throughout Nike's supplier network in Vietnam, Indonesia, China and other manufacturing markets.
Some products may eventually be made closer to American consumers.
Some technologies may create more value when installed inside Asian supplier facilities.
Some experiments will fail completely.
The objective is not reshoring for its own sake.
It is to give Nike more control over how quickly it learns, what it can automate, when it must commit inventory and where a product can be produced economically.
Nike does not need to predict where the cheapest place to manufacture a shoe will be in 2035.
It needs a manufacturing architecture that allows the company to respond when the answer changes.
Nike Built an Extraordinary Manufacturing Architecture
Nike's manufacturing model has historically been one of its greatest strengths.
The company largely designs, develops, markets and distributes products while independent contractors manufacture nearly all of its footwear and apparel outside the United States.
In fiscal 2026, Nike's contract manufacturers operated 95 finished-goods footwear factories across 11 countries. Approximately 52% of Nike Brand footwear was manufactured in Vietnam, 27% in Indonesia and 16% in China. Four footwear manufacturers produced approximately 60% of Nike Brand footwear.
Apparel was more diversified. Contractors operated 321 apparel factories across 34 countries, with approximately 34% of Nike Brand apparel produced in Vietnam, 15% in Cambodia and 12% in China. Nike also reported 205 strategic Tier 2 suppliers providing materials used throughout its products. (Nike FY2026 Form 10-K)
This is not an accidental system.
Asia has developed manufacturing ecosystems that are extremely difficult to replicate. Skilled labor, materials suppliers, component manufacturers, machinery, logistics, ports and decades of accumulated operating knowledge exist in close proximity.
A shoe is not made simply by finding someone willing to assemble it at a lower hourly wage. It exists within an interconnected industrial network.
That is why a simplistic reshoring strategy would almost certainly destroy value.
Nike's supplier network is not an outdated system waiting to be replaced. It is an asset that a new manufacturing architecture should strengthen.
It is also more than a source of production capacity.
It is a distributed base of manufacturing knowledge.
Across decades of production, Nike's contractors have learned how its materials behave, where defects emerge, which operations create bottlenecks, how design decisions affect assembly time and which processes remain difficult to automate.
Much of that knowledge is tacit. It exists inside production teams, engineering routines, tooling decisions and the accumulated experience of solving problems at scale.
That changes the starting point.
Nike would not be attempting to invent automated footwear and apparel manufacturing from nothing. It sits at the center of a network that already understands, in extraordinary detail, how Nike products are made.
The opportunity is to work with strategic suppliers to make more of that knowledge explicit, combine it with Nike's capabilities in product design, materials science, AI, robotics and demand planning, and codify the result into repeatable manufacturing systems.
Research on tacit knowledge transfer suggests that this kind of learning depends on trust, close interaction and the willingness of partners to share what they know. It cannot be treated as a simple extraction exercise. (Cogent Business & Management)
Suppliers would need a reason to participate through co-development agreements, shared productivity gains, equipment support, preferred-partner commitments and clearly defined ownership of intellectual property.
The supplier network is not merely where Nike might eventually install the technology.
It could help Nike invent it.
Tariffs Are a Warning, Not the Strategy
The obvious reason to discuss domestic manufacturing today is tariffs.
But tariffs alone are a bad reason to build a factory.
Trade policy changes. Administrations change. Courts intervene. Countries renegotiate. Exemptions appear and disappear.
Nike experienced this volatility during fiscal 2026. The company recognized a $986 million benefit associated with the expected recovery of IEEPA tariffs after those tariffs had previously increased product costs. Nike allocated $965 million of that benefit to North America. (Nike FY2026 Form 10-K)
That does not mean Nike currently has a recurring $1 billion tariff problem. The particular IEEPA charges were subsequently reversed.
It demonstrates something more important.
Trade policy can move Nike's economics at enormous scale.
Nike itself acknowledges that protectionist measures can increase product costs, changes in American trade policy can materially affect the business, and alternative sources of supply may not be as cost-effective as existing sources.
The right response is not to predict exactly what tariff Vietnam will face in 2031.
The better question is:
How valuable would it be for Nike to have more choices if the economics of global manufacturing change?
Labor costs can rise. Freight can become more expensive. Energy prices can move. Political relationships can deteriorate. Shipping lanes can become less reliable. Environmental requirements can increase costs. New technologies can reduce the percentage of production cost attributable to labor.
Nike does not need to know which variable will matter most over the next decade.
It needs an architecture capable of responding.
The Bigger Economic Opportunity May Be Time
Labor receives most of the attention in reshoring conversations.
For Nike, time may be at least as important.
Nike says it generally orders product at least four to five months before sale, based primarily on advanced wholesale orders and internal orders from its direct-to-consumer operations. It also offers some wholesale customers the opportunity to order five to six months before delivery under its futures ordering program. (Nike FY2026 Form 10-K)
At May 31, 2026, Nike had approximately $4.9 billion in legally binding product-purchase obligations, with $4.7 billion payable during the following 12 months. It was carrying approximately $7.5 billion in inventory.
Think about what those figures represent operationally.
Months before consumers make their decisions, Nike has to make its own.
How many units?
Which colors?
Which styles?
Which sizes?
Which markets?
Which channels?
Nike commits capital, orders production, transports products around the world, distributes them into warehouses and stores, and eventually discovers whether its forecast was correct.
Sometimes a product becomes difficult to replenish.
Sometimes demand disappoints and Nike has too much.
Both outcomes are expensive. Nike explicitly identifies inaccurate forecasting as a source of excess inventory, shortages, discounting, lower margins, reduced cash flow and potential damage to retailer and consumer relationships. (Nike FY2026 Form 10-K)
Manufacturing closer to demand could potentially change that equation for selected products.
Instead of relying entirely on:
Forecast → commit → manufacture → ship → warehouse → sell
parts of the portfolio could move closer to:
Forecast base demand → observe early demand → configure and replenish → sell
That wording matters.
Manufacturing lead time is only one part of the system. A shoe assembled in Oregon may still depend on fabrics, foams, yarns, Air components, tooling and packaging sourced elsewhere. If all of those inputs must be committed months earlier, moving final assembly does not solve the whole problem.
The economic value depends on how far Nike can move the point at which it commits to a specific product, color, size or market.
The most valuable system may hold common materials or semi-finished components and postpone final configuration until demand becomes clearer.
The opportunity is not simply shorter transportation distance.
It is delaying irreversible product decisions.
The Economics of Responsive Manufacturing
Nike's FY2026 cost of sales was approximately $26.5 billion. Average inventory across FY2025 and FY2026 was approximately $7.5 billion.
Dividing average inventory by FY2026 cost of sales produces roughly 103 days of inventory-equivalent on a simple company-wide basis.
That is a JAXONLABS calculation, not a Nike-reported operating metric. It does not reveal inventory by product, lifecycle stage, channel or manufacturing lead time.
It is still useful for understanding scale.
Every 10 days of FY2026 cost of sales represents approximately $726 million of inventory-equivalent.
Twenty days represents approximately $1.45 billion.
Thirty days represents approximately $2.18 billion.
Reducing inventory by $1 billion would not create $1 billion in profit. Nor should these figures be read as a forecast of what advanced manufacturing could release.
But inventory has a cost. So do warehousing, handling, obsolescence and markdowns. Excess inventory can also push a premium brand into promotional behavior that weakens pricing power and marketplace health.
This is why the manufacturing question should not be:
Can Nike manufacture a $10 item in America for less than $10?
It should be:
Can Nike create a production model whose total system economics are superior, even if its direct conversion cost is initially higher?
Imagine an offshore product costs $10 to manufacture and a comparable domestically automated product costs $11.
The domestic product appears to lose.
But what if the offshore product requires Nike to commit four months earlier? What if Nike pays international freight and duties? What if part of the order is eventually marked down? What if a different product sells out but cannot be replenished while consumers still want it?
Manufacturing cost alone does not tell us which system creates more value.
A more complete model would consider:
Expected system value = full-price contribution + recovered contribution from fewer stockouts + strategic flexibility value − conversion and material costs − duties and freight − inventory and warehousing costs − markdown and obsolescence exposure − capacity and capital costs − transition, quality and execution risk.
Nike undoubtedly models many of these variables internally.
The strategic question is what happens when the production system is deliberately redesigned to improve the entire equation.
This is not a new economic concept, even if the technology is changing.
Operations researchers have studied supply chains that combine a lower-cost, long-lead-time production mode with a more expensive, rapid-response mode.
Karen Donohue's research found that the value and coordination of the faster mode depend partly on how much demand information improves between production periods. (Management Science)
Marshall Fisher and Ananth Raman studied how postponing some production until early sales information becomes available can reduce the combined cost of stockouts and excess inventory. Their application involved a skiwear company, not a business with Nike's scale or product complexity, so its results should not be generalized directly. But the work provides strong evidence for the operating principle behind responsive capacity. (Operations Research)
Nike's opportunity is to determine whether automation, product redesign and better demand data can reduce the cost of that second production mode enough to make it viable across selected parts of the portfolio.
Conventional Reshoring Still Probably Fails
Before going further, we should attack the argument ourselves.
If Nike simply replaced Vietnamese labor with American labor while producing the same products in largely the same way, the idea would likely fail.
American wage levels are higher. Domestic textile and footwear supplier networks are less dense. Machinery must be purchased. Factory utilization matters. Materials may still need to be imported. Domestic assembly does not automatically eliminate tariffs on imported components.
Ownership also changes the risk profile.
Nike currently transfers a meaningful amount of factory investment, utilization risk and labor management to independent manufacturing partners. A large owned-factory network would move more fixed assets and operating complexity onto Nike's balance sheet.
The company already had $3.1 billion in operating lease liabilities at May 31, 2026, representing $3.6 billion of undiscounted future lease payments. It reported an 18.7% return on invested capital for FY2026. (Nike FY2026 Form 10-K)
That ROIC is context, not necessarily the hurdle rate for a new manufacturing investment. Any project should be evaluated against Nike's risk-adjusted cost of capital, alternative uses of cash and the option value created by the capability.
The company should also resist turning this into a real-estate strategy.
Manufacturing has specific requirements around power, ceiling height, materials movement, logistics, environmental controls, transportation access and labor. A building does not become a useful manufacturing asset simply because Nike already owns or leases it.
The lesson is straightforward.
Do not start with a giant reshoring commitment.
Start with a disciplined experiment designed to determine whether Nike can change the manufacturing equation.
Adidas Already Ran Part of This Experiment
Nike would not be the first major sportswear company to explore automated regional manufacturing.
Adidas opened highly automated Speedfactory facilities in Ansbach, Germany and Atlanta, Georgia. The premise was similar to the opportunity being discussed today: shorten production cycles, manufacture closer to consumers and respond faster to local demand.
Then Adidas closed them.
Ignoring that fact would make the entire argument easier to dismiss.
But the Speedfactory story is more useful than a simple failure narrative.
Adidas did not conclude that the technology had no value. It decided that Speedfactory technologies could create better economics when deployed within its established Asian supplier network. The company specifically pointed to better use of existing production capacity and greater flexibility in product design. (Adidas Speedfactory announcement)
The factories closed. The manufacturing knowledge survived.
Adidas also continued working to reduce production lead times across its broader supply chain. By 2019, it reported production lead times of 60 days or less for the vast majority of its footwear, apparel and hardware products and connected faster replenishment to higher-quality sales and more productive inventory. (Adidas 2019 Global Operations report)
The wrong conclusion is:
Automated local manufacturing does not work.
The more useful conclusion is:
The location of production and the ownership of manufacturing innovation are separate decisions.
Nike might build a production cell in Oregon and later discover that its best deployment is inside a strategic supplier's factory in Vietnam.
That would not necessarily make the American facility a failure. If it produced defensible technology, better product architectures, shorter changeovers or improved quality across a much larger supplier network, it may have succeeded as a manufacturing research and development platform.
The Speedfactory experience also raises an important warning for Nike: advanced machinery is not valuable simply because it is advanced. Utilization, flexibility, transferability and integration with the broader production system determine whether the economics work.
What Has Changed Since Speedfactory?
Adidas closed its Speedfactories only a few years ago. Nike therefore needs a clear reason to believe the economics could be different now.
Several capabilities are improving.
Robotic vision is becoming more capable. AI systems can interpret more variable physical environments. Industrial automation is becoming more flexible. Additive manufacturing is moving beyond prototyping. Digital twins and connected production systems can capture more operating data. Manufacturing software can coordinate machines, people, maintenance and production planning more effectively.
But there is a reason footwear and apparel have not already been fully automated.
A shirt is surprisingly difficult for a robot.
Fabric bends, folds, stretches and wrinkles. Its shape changes while it is being handled. A recent review in the Annual Review of Control, Robotics, and Autonomous Systems describes deformable textile manipulation as a continuing robotics challenge, particularly when systems must generalize across materials and tasks. (Annual Reviews)
The future of automated apparel probably is not a humanoid robot standing behind every sewing machine.
There may be a better approach.
Instead of forcing robots to imitate every human movement, redesign the process around what machines do well.
The Emerging Manufacturing Stack
Several companies are already working on pieces of this problem.
CreateMe Technologies has developed an autonomous garment-production platform using industrial robotics, specialized fixtures and digitally bonded construction for certain applications. The company claims that its system can operate much faster than manual production and compete with offshore economics. Those are company-reported claims and should be treated as emerging evidence until they are validated at meaningful commercial scale. (CreateMe technology overview)
SoftWear Automation is approaching textile handling through machine vision and automated sewing. The company raised $20 million in 2025 as it worked toward commercializing its Sewbot technology. (SoftWear Automation announcement)
A more direct competitive signal is coming from On.
In 2026, the company opened its second robotic LightSpray production facility near Busan, South Korea. The facility uses 32 robots to manufacture LightSpray footwear, and On has described plans to expand the production model closer to major markets. (Reuters)
The important part is not simply that On bought robots.
It redesigned the footwear upper and the manufacturing process together, using robotic arms to spray a continuous filament onto a form and eliminate many conventional production steps.
That is evidence that the competitive race is already moving beyond better forecasting and cheaper labor. A smaller rival is attempting to turn a proprietary manufacturing process into a product, supply-chain and brand advantage at the same time.
Nike should not assume its scale protects it from that kind of innovation.
Other organizations are developing robotic material handling, autonomous internal logistics, machine tending, automated inspection, connected production software and additive manufacturing.
No single vendor represents Nike's factory of the future.
Together, they suggest that a new manufacturing stack is emerging:
Digital product definitions
Automated knitting and bonding
Molding and additive manufacturing
Robotic handling and assembly
Machine vision and quality inspection
Autonomous internal logistics
AI-assisted scheduling and maintenance
Connected product, demand and production data
Nike is large enough to help define how those pieces fit together.
Nike Has an Advantage Most Brands Do Not
A smaller apparel company may need to wait for the technology to mature.
Nike does not necessarily have to.
More importantly, Nike is already a manufacturer.
Air Manufacturing Innovation, a wholly owned Nike subsidiary, manufactures Air-Sole and other cushioning components at facilities near Beaverton, Oregon; St. Charles, Missouri; and Dong Nai City, Vietnam.
Nike says Air MI holds more than 500 patents and employs more than 100 engineers across 15 engineering specialties, along with PhDs and skilled production workers. Its teams build technology platforms and flexible production lines specifically for Air. (Nike Air Manufacturing Innovation)
Nike is also recruiting engineers to work on automation, robotics, connected systems and Industry 4.0 within Air MI. (Nike Careers)
That materially changes the proposal.
Nike would not be waking up tomorrow and deciding to become a manufacturer.
It already possesses proprietary advanced-manufacturing capability.
The strategic question is how far that capability should extend.
Nike also has an advantage that Air MI alone cannot provide: the breadth of its supplier relationships.
Air MI can contribute proprietary engineering, automation experience and the discipline of operating flexible production lines. Nike's contractors can contribute deep knowledge of cutting, stitching, bonding, molding, material handling, line balancing, quality control and commercial-scale production across hundreds of facilities.
Nike is in a position to connect those capabilities.
Very few brands have the product volume, supplier reach, internal engineering capacity and financial resources required to turn lessons from individual factories into a reusable global manufacturing system.
Product Design and Manufacturing Design Need to Converge
Flyknit provides a useful example.
Instead of constructing a footwear upper through a conventional cut-and-sew process, Nike uses precision knitting to create the upper more directly. Nike says Flyknit produces, on average, 60% less waste than traditional footwear-upper manufacturing. (Nike materials overview)
The sustainability benefit matters.
The broader strategic lesson is that Nike changed how the product is manufactured by changing how the product is designed.
That is exactly the thinking advanced automation requires.
The wrong question may be:
How do we teach a robot to assemble today's Nike shoe?
A better question is:
What would a Nike shoe look like if it were designed from the beginning for highly automated manufacturing?
That could change materials, fasteners, bonding, knitting, molding, component count, assembly sequence, packaging, customization and recyclability.
Nike's work with Zellerfeld on 3D-printed footwear offers another piece of the puzzle. Its Air Works program brought designers to Nike headquarters to develop printed Air Max concepts, and Nike has announced plans to offer the Air Max 1000 through Nike By You. (Nike Air Works)
Additive manufacturing does not currently replace Nike's footwear-production system at anything close to global scale. But software-defined manufacturing can support smaller runs, rapid experimentation, customization and product architectures that conventional processes cannot produce in the same way.
The future factory and the future product may need to be designed together.
That design process should include supplier engineers from the beginning.
They understand which product features create unnecessary operations, where tolerances become difficult to maintain, which materials create handling problems and which assembly steps depend most heavily on human dexterity.
The objective should be to translate that experience into design rules for automation: preferred materials, reduced component counts, standardized interfaces, machine-readable tolerances, reusable process modules and product architectures that can move across compatible production cells.
Nike would not simply be automating the way products are made today.
It would be learning how to design products that belong in a different manufacturing system.
Nike Has Put Important Organizational Pieces Together
Advanced manufacturing is not simply an engineering project.
It requires technology, product design, materials, supply-chain planning, operations, finance and manufacturing to work as one system.
In December 2025, Nike created a new executive vice-president and chief operating officer role under Venkatesh Alagirisamy. His responsibilities brought together Technology, Supply Chain, Planning, Operations, Manufacturing and Sustainability.
Nike said the intention was to integrate technology more deeply across how the company creates, plans, makes, delivers and sells products. (Nike leadership announcement)
In April 2026, Nike announced a broader operations restructuring that included modernization of Air MI facilities in Oregon, Missouri and Vietnam. The stated objectives included greater resilience, streamlined processes and faster innovation. (Nike global operations announcement)
That organizational structure aligns closely with what an advanced-manufacturing platform requires.
AI cannot sit inside an IT function while manufacturing operates somewhere else. Product design cannot be separated from process engineering. Supply-chain planning cannot be disconnected from production capacity.
Manufacturing data needs to inform forecasting. Forecasting needs to shape production. Production needs to influence product design.
Nike has already placed many of those disciplines under the same operational umbrella.
The opportunity is to push the integration further.
Why Develop the Technology Close to Home?
This is where a meaningful American manufacturing investment makes sense.
Not because every Nike product should eventually say Made in USA.
Not because American labor is suddenly cheaper.
Not because Nike should abandon Vietnam.
Nike should develop a significant part of its next manufacturing architecture close to the people building it.
Nike's global headquarters are in Oregon. Much of its product innovation, design capability and executive leadership sit there. Air MI already operates nearby.
If Nike wants to master AI-driven physical manufacturing, proximity could accelerate the learning cycle.
A designer could walk onto a production floor. An engineer could modify a robotic process. A product team could redesign a component around that process. Machine-vision failures could be reviewed by the same group that designed the product. A material could be tested without waiting for another production cycle on the other side of the world.
Research on the relationship between production and innovation suggests that the value of colocation depends on how tightly product development and production knowledge interact. One study of American firms found an association between closer manufacturing and innovation facilities and higher patenting, while also finding that some companies continued innovating after exiting manufacturing. Colocation is not universally necessary, but it can matter when process learning is central to the innovation itself. (Colocation of Production and Innovation)
Nike's initial facility would not necessarily need to be the cheapest in its network.
It would need to be the place where Nike learns how to build cheaper, faster and more flexible production systems.
That is why the first investment should be evaluated partly as manufacturing research and development.
Which Products Should Nike Start With?
The temptation would be to select an iconic sneaker.
That may be the wrong starting point.
An Air Jordan can contain multiple materials, components and assembly requirements. Attempting to automate every operation could turn the pilot into a complex technology showcase that never achieves commercial relevance.
Nike should select products using a disciplined framework.
Responsive manufacturing is most likely to create value when:
Demand is difficult to forecast.
The selling window is relatively short.
Full-price margins are attractive.
Stockouts represent recoverable lost sales.
Markdown exposure is meaningful.
Materials can be staged before final demand is known.
Final product configuration can be postponed.
Equipment can switch between multiple SKUs.
The design is sufficiently automation-ready.
Nike receives usable demand data early enough to respond.
That creates four broad product strategies:
Lower automation readiness | Higher automation readiness | |
Lower demand uncertainty | Conventional global sourcing | Automated production through global suppliers |
Higher demand uncertainty | Limited runs, flexible manual capacity or product postponement | Regional demand-responsive priority |
Potential early candidates could include certain knitted products, footwear uppers, molded or Air components, performance basics, simple garments, personalized products and limited releases designed specifically around automated assembly.
Nike does not need to automate an entire shoe or garment on the first attempt.
The first meaningful breakthrough may be a group of operations that currently dominate labor time, delay changeovers or create quality bottlenecks.
A Hybrid Manufacturing Architecture
The end state should not be complete reshoring.
It should be a hybrid system built for optionality.
1. American advanced-manufacturing hubs
These facilities would operate as both commercial production environments and manufacturing laboratories.
Their job would be to develop and prove robotic assembly, computer vision, automated inspection, materials handling, knitting, bonding, molding, additive manufacturing, predictive maintenance, production software and products designed for automation.
They could produce goods for sale, but their early value would also come from learning.
2. Regional demand-responsive capacity
Once a process is proven, Nike could deploy capacity closer to selected markets where speed creates enough economic value.
The United States accounted for approximately $20.4 billion, or 44%, of Nike's FY2026 revenue. That makes North America an obvious first environment in which to test rapid replenishment, customization, limited production and supply-chain redundancy. (Nike FY2026 Form 10-K)
Regional production would not need to manufacture the entire assortment. It could focus on replenishing selected winners, supporting launches, configuring products closer to demand and providing alternative capacity during disruptions.
Regional capacity also creates its own risks.
Dividing production among multiple markets can reduce economies of scale, duplicate equipment and weaken the inventory-pooling benefits of a centralized network. Research on multi-location production describes this explicitly as a trade-off between pooled inventory and responsive service. (European Journal of Operational Research)
Nike would need to manage regional production as an integrated network, not a collection of isolated factories.
3. Technology-enabled global suppliers
Nike's Asian manufacturing partners should not be treated as an old system that eventually disappears.
They are part of the advantage.
When Nike develops an automation system that genuinely improves manufacturing economics, it can deploy that technology into supplier facilities in Vietnam, Indonesia or elsewhere.
But this should not be framed as Nike collecting supplier knowledge, converting it into proprietary technology and handing the finished system back to factories.
That approach would create resistance and overlook where much of the practical knowledge resides.
The better model is co-development.
Nike could establish long-term programs with a smaller group of strategic manufacturing partners, combining Nike engineers, supplier production specialists, equipment companies and material innovators around defined process problems. Contracts would need to establish how development costs, productivity gains, operating data and intellectual property are shared.
Some systems may be owned by Nike. Some may belong to suppliers. Others may be jointly developed or licensed across selected facilities.
The critical difference is not that Nike owns every machine.
Nike becomes better at how manufacturing works rather than simply becoming a larger owner of factories.
The Real Asset Is Manufacturing Intelligence
A robot is not much of a competitive advantage if Nike and every competitor can buy the same robot.
The advantage exists in the system around it.
At maturity, that system could function like a manufacturing operating system for Nike products.
It would connect digital product definitions, automation-ready design rules, reusable production modules, machine-vision models, quality data, equipment interfaces, scheduling logic and demand signals.
How products are designed for manufacturing.
How production cells are configured.
How machines work together.
How computer vision detects defects.
How quickly a line can change from one product to another.
How much human intervention remains.
How maintenance is predicted.
How demand signals influence production.
How materials are designed around automated processes.
How that knowledge is protected through patents, software, data and trade secrets.
The most valuable output may be a growing library of what Nike has learned: which materials work with which processes, which operations can be combined, how cells should be configured, what failure patterns machines should detect and how a new product should be designed to move through the system.
Every successful production run could make the platform more capable.
Nike's annual report already includes manufacturing techniques and related technologies among the innovations protected through its intellectual-property portfolio.
The potential moat is not:
Nike owns more robots than Adidas.
It is:
Nike knows how to manufacture Nike products through an integrated system competitors cannot easily recreate.
That creates an important governance question.
If Nike depends on suppliers to scale the technology, it must determine which capabilities it needs to own, which can be shared, how suppliers will be compensated for adopting the system, and how proprietary process knowledge will be protected.
The manufacturing model and the ownership model need to be designed together.
Start With the Business Case, Not the Robot
If Nike wanted to test this concept, it should not begin by choosing a robotics company.
Choose a product.
Map every cost and decision required to move it from raw material to customer.
Then determine which parts of the equation could change under a different production architecture.
The model should include:
Conversion cost: How much labor, machine time and overhead does each unit require?
Materials: Which inputs can be sourced or staged locally and which remain dependent on long international lead times?
Commitment point: How late can Nike decide the product, color, size and destination?
Capital intensity: What does the equipment cost, how long will it last and how reusable is it across products?
Capacity economics: What utilization is required, and how much availability should be preserved for rapid response?
Duties and freight: What exposure actually disappears under regional production?
Inventory: How much later can Nike commit capital and how much safety stock is required elsewhere in the network?
Markdowns: Does later production measurably reduce excess inventory and promotional activity?
Stockouts: Can Nike replenish winners quickly enough to capture contribution that would otherwise be lost?
Quality: Does automation improve consistency or create new failure modes?
Product flexibility: How quickly can the line change products, sizes and colors?
Maintenance: What happens when specialized equipment goes down?
Channel economics: Is the product sold through Nike Direct, wholesale or both? Who owns the demand data, who carries the inventory risk and who captures the value of replenishment?
Ownership: Which capabilities must Nike own, and which are better developed with equipment companies or strategic suppliers?
Only after understanding that equation should Nike select the technologies.
A Simple Sensitivity Shows Why the Test Is Worth Running
Nike generated $20.5 billion in North American revenue during FY2026 and reported approximately $11.16 billion in regional cost of sales.
Imagine Nike identifies products representing 10% of North American cost of sales as plausible candidates for a more responsive production model.
That represents approximately $1.1 billion in annual product cost.
Twenty days of cost of sales associated with that pool equals approximately $61 million of inventory-equivalent.
Forty days equals approximately $122 million.
Sixty days equals approximately $183 million.
Those are JAXONLABS sensitivities, not forecasts. They do not estimate how much inventory Nike could actually remove.
They also exclude markdown savings, duties, freight, recovered stockout sales, capital spending, higher domestic conversion costs and new safety-stock requirements.
They simply demonstrate why reducing the commitment period can become financially meaningful even within a small part of Nike's North American portfolio.
The real model must be built at the product and channel level.
Responsive Capacity Should Not Be Judged by Utilization Alone
Automation can become expensive when equipment sits idle.
But a rapid-response system operating at maximum utilization has no room to respond.
That creates a deliberate economic trade-off.
Some unused capacity may function like insurance. It can support an unexpected product winner, a launch, a disruption or a short production run that would otherwise be uneconomic.
Nike should not pursue maximum utilization as an isolated target.
It should optimize the balance between production efficiency, availability and flexibility.
The right question is not whether every machine is running every minute.
It is whether the value produced by the system, including its option value, exceeds the full cost of maintaining it.
What Happens to the People?
A discussion about automated reshoring needs to be honest about employment.
If machines replace a large part of direct labor, American production will not recreate traditional manufacturing employment at the same scale.
Research published in the Journal of International Economics finds that trade uncertainty can encourage both reshoring and automation. It also suggests that automation can increase productivity while shifting demand toward higher-skilled workers and displacing some lower-skilled work. (Journal of International Economics)
The future Nike manufacturing workforce may include more roles in robotics maintenance, process engineering, machine vision, quality assurance, materials science, equipment calibration, production software, data analysis and advanced operations.
Those positions will not automatically go to the same people whose work is displaced. Nor will they necessarily exist in the same locations.
A credible manufacturing strategy therefore needs a workforce strategy alongside it: training, apprenticeships, technical partnerships and an honest plan for managing displacement.
Nike's Air MI organization already offers an early example of engineers and skilled production employees working within the same manufacturing system.
As execution becomes increasingly automated, human value moves toward system design, judgment, exception handling and continuous improvement.
Nike Has the Capacity to Run the Experiment
None of this would be worth proposing if Nike were fighting for survival.
It is not.
Nike ended FY2026 with approximately $9 billion in cash and short-term investments and generated approximately $2.9 billion in operating cash flow. It invested $684 million in property, plant and equipment during the year. (Nike FY2026 Form 10-K)
The company has demonstrated the ability to deploy capital at a scale large enough to matter.
That does not mean it should immediately spend billions on factories.
A staged, multi-year advanced-manufacturing program could be meaningful relative to current capital expenditures while remaining manageable relative to Nike's balance sheet.
The initiative should also preserve ownership flexibility.
The first production environments could be Nike-owned, built as joint ventures, located inside existing Air MI facilities, operated with equipment partners or established as dedicated cells within strategic supplier facilities.
Nike should own the capabilities that create differentiation. It does not necessarily need to own every building or machine used to deploy them.
A Four-Gate Path Forward
Nike should avoid announcing a giant reshoring initiative.
It does not need a slogan.
It needs evidence.
Gate One: Build the manufacturing-intelligence team
Create a dedicated cross-functional program connecting Air MI, product design, materials, robotics, AI, technology, supply-chain planning, finance and selected manufacturing partners.
Its first job should be identifying product architectures where automation and later commitment can create the greatest economic value.
Its second should be mapping the knowledge already distributed across Nike's supplier network: the highest-labor operations, recurring quality problems, difficult materials, slowest changeovers and processes that experienced production teams believe are most ready for redesign.
Selected supplier engineers should be embedded in the program rather than consulted after Nike has already chosen the technology.
Gate Two: Build integrated production cells
Do not start with a giant factory.
Build manufacturing cells that automate narrow groups of operations.
Measure conversion cost, uptime, quality, changeover time, material dependency, human intervention, energy use and capital productivity.
Learn which processes can be reused across multiple products.
Gate Three: Connect manufacturing to demand
Production should not simply become automated.
It should become demand-responsive.
Nike should test whether early sales signals can trigger economically useful production decisions within defined financial, capacity and service constraints.
The objective is a faster loop between what consumers want and what Nike makes.
Gate Four: Scale the manufacturing intelligence
Once a process is economically proven, decide where it belongs.
Keep it in the United States. Install it at a Vietnamese supplier. Build a regional cell in Europe. Combine it with additive manufacturing. Develop it with a strategic equipment partner.
Where suppliers helped create the system, the deployment model should allow them to participate in the value it produces. That may include shared savings, longer-term volume commitments, licensing arrangements or preferred access to new production programs.
The winning deployment is the one that creates the greatest risk-adjusted return, not the one that satisfies a predetermined reshoring target.
Before advancing through each gate, the system should demonstrate:
Competitive total-system economics
Meaningful reduction in decision or replenishment time
Product performance and desirability
Economically productive capacity
Repeatability across products
Reduced supply-chain vulnerability
Transferability across facilities
An acceptable risk-adjusted return on capital
If a process fails those tests, Nike should stop.
The purpose of the program is not to preserve every experiment. It is to discover which capabilities deserve to become platforms.
What This Could Mean for Nike's Turnaround
Advanced manufacturing does not replace the turnaround work Nike is doing now.
A better factory cannot rescue an undesirable shoe.
Robotics cannot restore cultural relevance.
AI cannot rebuild a damaged wholesale relationship.
Automation cannot make a consumer in Shanghai want a Nike product.
Nike still needs to win in sport. It needs visible innovation, better merchandising, stronger storytelling, faster decisions and healthier marketplace relationships.
Manufacturing architecture can support all of those things.
A faster production system makes product teams more responsive.
Smaller production commitments reduce the risk of experimentation.
Rapid replenishment allows Nike to chase selected winners rather than relying entirely on forecasts.
Customization creates new consumer experiences.
Manufacturing intellectual property can become a moat.
Regional capacity can reduce exposure to certain disruptions.
Lower inventory requirements can release working capital.
Closer integration between design and manufacturing can create products competitors struggle to reproduce.
This is not a manufacturing strategy separated from Nike's brand.
It could become part of how Nike innovates.
Nike's Next Platform
Nike has historically been good at turning technologies into systems.
Air was not simply cushioning. It became a platform.
Flyknit was not simply another material. It became a different way to construct footwear.
Nike's digital ecosystem was not simply an online store. It became a consumer platform.
The same thinking can be applied to manufacturing.
Most companies treat manufacturing as the final execution step after product creation.
What if Nike treated manufacturing itself as an innovation platform?
An environment where product design, materials science, AI, robotics, production engineering and demand data continuously interact.
The output would not simply be shoes made by robots.
It would be a company capable of learning faster about how physical products should be designed, made and replenished.
At Nike's scale, small improvements can become significant. A 100-basis-point improvement in gross margin, holding FY2026 revenue constant, would represent approximately $464 million in gross profit.
Nike does not need every experiment to transform the entire company.
It needs a few that become platforms.
The Bigger Strategic Question
Nike's current challenges are real.
So are its advantages.
It has cash, scale, product volume, manufacturing engineers, Air MI, intellectual property, additive-manufacturing experiments and a global supplier network capable of deploying successful new technologies.
That supplier network also represents decades of accumulated knowledge about how Nike products behave in production.
It has also brought Technology, Supply Chain, Planning, Operations and Manufacturing under a single chief operating officer.
Most importantly, Nike has a reason to challenge old assumptions.
Turnarounds create permission to ask questions successful companies often avoid.
The question is not whether Nike should abandon the manufacturing model that helped build it.
That would be reckless.
The question is whether that model should remain Nike's only manufacturing model.
It should not.
Nike should continue using the extraordinary supplier ecosystem it already has. At the same time, it should develop an advanced manufacturing capability closer to home, using AI, robotics, product redesign and automation to explore what the next generation of athletic-product manufacturing can become.
The more ambitious opportunity is to connect those two systems.
Nike has spent decades building one of the world's most capable outsourced manufacturing networks. Its next opportunity is to turn what Nike and that network have learned together into a manufacturing platform neither could create as effectively alone.
That would be more than an automation program.
It would be an attempt to pioneer a new era of footwear and apparel production.
Some production may remain in America.
Some technology may ultimately create more value in Asia.
Some processes may fail.
Others may change Nike's economics.
That is precisely why the company should start now.
Nike does not need to predict the cheapest place to manufacture a shoe in 2035.
It needs the technology, people and manufacturing intelligence to respond when the answer changes.
Nike's next great innovation might be a shoe. But it might also be the system that makes it.
Research Notes
This analysis relies primarily on Nike's FY2026 Form 10-K for financial performance, manufacturing concentration, inventory, purchase commitments, capital allocation, trade risks, leases and geographic results. Financial sensitivities identified as JAXONLABS calculations are directional analyses based on publicly reported figures, not Nike forecasts.
External operations research is used to support the underlying logic of responsive production, forecast updating, inventory commitment and regional manufacturing. Vendor performance claims are identified as company-reported and should not be interpreted as independently validated Nike-scale economics.
Research on tacit knowledge transfer within buyer-supplier relationships informs the discussion of supplier participation, trust and co-development. Reporting on On's LightSpray factory is included as current evidence that a direct competitor is already combining proprietary product design, robotics and regional manufacturing.
