A response to Eric Miscoll’s ‘The De-EMSing of the EMS Industry’ (EMSNOW, May 2026)
By Richard Nilsen, CEO and Chairman, Tepcomp Group Oy
Eric Miscoll’s recent piece on the The De-EMSing of the EMS Industry’ makes a sharp argument about how the top tier is evolving. Flex, Celestica, Foxconn, Jabil, and Sanmina are moving beyond pure manufacturing services into hybrid models that combine manufacturing with platforms, IP, and vertical integration. The strategic logic for them is exactly as Eric describes: higher margins, stickier relationships, less commoditisation exposure, greater strategic relevance to OEMs. The framing is well-observed and the read on the top tier is correct.
Where the article leaves a question open is what happens to the rest of the industry. In introducing the EMSNOW/in4ma EMS & ODM Global 100 list earlier this year, Eric himself noted that “the pureplay EMS segment appears more fragmented and is dominated by midsized companies.” Dieter Weiss’s own data on that segment tells a difficult story. The 2026 in4ma annual survey carries the headline title “A difficult year ended, and we are facing another challenging year.” The 2025 half-year survey subtitle was sharper still: “After two years of artificially inflated demand, we are returning back to reality with many problems.”
The question this leaves is not what mid-sized EMS evolution looks like as a parallel to the top-tier story. It is something more uncomfortable. The conventional way of running a mid-sized EMS at scale is no longer financially sustainable, and the segment is increasingly showing it. The question is what the structural response looks like.
The difficulty is structural, not cyclical
Over the past two decades, I have led and operated several EMS companies through different positions in the European market. The pattern across these companies has been consistent. At scale, the conventional EMS operating model produces financial outcomes that are structurally unsustainable, and incremental operational improvement does not fix the structural problem. Better procurement, leaner production, smarter sales, more disciplined cost management. All of it helps at the margin, none of it changes the fundamental economics of the model.
That observation used to be controversial. The 2025 and 2026 in4ma data is now making it harder to dispute. European EMS revenue declined 2.9% in 2025. Austria fell 15.2%, France fell 13.5%, Germany fell 7.0%. In Germany, removing two outlier companies, the remaining 127 firms declined 12.0%, and this is the second consecutive year of double-digit decline for the German segment, following 14.6% in 2024. A two-year analysis of 129 German EMS companies found that only 16% increased revenues across the period, while 34% saw revenues fall by more than 30%. France recorded 11 insolvencies in 2025; Germany recorded 5. Across the surveyed companies in Europe, 3,445 employees were laid off, with 94% of those layoffs in Western Europe.
The size correlation in the data is striking. In the 2025 half-year survey, EMS companies with revenues above 50 million euros declined 6.6%, while those below 50 million euros declined 12.7%, roughly twice the impact. The smallest reporting group (under 2 million euros) declined 24.2%. Book-to-bill ratios for the first half of 2025 were below 1.0 across all revenue groups except the largest. The smaller the company, the worse the position.
This is not the picture of a tough patch that will resolve through better operations and the next demand cycle. Two consecutive years of double-digit decline in the largest markets, a third of German companies losing more than 30% of revenue over two years, insolvencies happening in numbers, and a clear correlation between company size and severity of difficulty. This is the picture of a segment where the underlying model is failing, not where individual operators are managing poorly.
The top-tier path is not available to the segment
The De-EMSing path Eric describes requires three things that the top tier has and the mid-sized segment does not.
Capital for proprietary platforms and meaningful IP investment. Building differentiated proprietary technologies, vertical integration, or platform-based offerings requires sustained investment over years before returns materialise. The top-tier players have balance sheets that absorb this kind of investment without compromising core operations. Mid-sized EMS, particularly in Europe, typically do not.
Scale to monetise vertical integration and platforms. A proprietary subassembly or platform technology produces commercial returns through the number of customers who adopt it. The top-tier players have customer bases broad enough that even modest adoption produces meaningful revenue. For a mid-sized EMS with a smaller customer base in specialised markets, the same proprietary investment may never reach the scale where it pays back.
Operational tolerance for running hybrid manufacturer-product companies. Operating a manufacturing services business while simultaneously running a product or platform business introduces complexity that the largest companies can absorb through organisational depth. For a mid-sized EMS, attempting the hybrid model on inadequate capital and inadequate organisational depth is genuinely dangerous. The result is often a company that is neither competitive on its core manufacturing nor differentiated enough on its proprietary work to justify the investment.
None of this is criticism of the top-tier path. For the companies Eric names, it is the right evolution, and they are well-positioned to execute it. The point is structural: the path is not available to the segment that dominates the industry by company count and treating it as “the next logical stage of maturity” for the industry as a whole reads the segment from the top tier’s vantage point.
And the conventional model cannot be fixed by doing it better
The more uncomfortable point is what this leaves. If the top-tier path is structurally unavailable and the conventional mid-sized model is financially unsustainable, then the segment is in a position where incremental improvement is not enough. The model itself must change.
Several forces are squeezing the conventional model simultaneously. Working capital intensity has historically absorbed 20 to 35% of revenue under conventional structures, tying up capital that produces no return. Fixed equipment costs do not flex with demand cycles, leaving operators exposed in downturns. Headcount intensity grows proportionally with revenue under relay-based coordination structures, making scale expensive rather than profitable. Margin compression comes from both Chinese price competition at the commodity end and from increasing European customer leverage at the strategic end. And customer expectations are shifting toward self-service experiences and structured commercial commitments that conventional EMS structures cannot meet at any operational cost.
Operators in the segment know something is wrong. The conventional response is to treat the symptoms as cyclical or operational: cut costs in downturns, chase volume in upturns, hope the next cycle will be better. The structural argument is that the cycles are not the problem; the model is the problem, and it has been masked for some years by various supports (favourable financing conditions, supplier flexibility, customer tolerance for opaque commercial structures) that are now eroding simultaneously.
The other form of De-EMSing
The architecture I want to describe did not emerge as an analytical answer to a theoretical question. It emerged from working the problem across multiple companies over more than two decades, watching what works and what does not when an operator tries to build a financially sustainable mid-sized EMS at scale. This is something most operators in the segment rarely have the opportunity to do across enough different situations to draw structural conclusions. The people who run a single company through a single cycle tend to see their experience as situational rather than as data.
What I have observed is that the responses operators commonly try do not solve the structural problem. Doing the conventional model better (leaner production, smarter sourcing, more disciplined cost management) produces marginal improvement but not the financial profile the segment needs. Becoming a smaller version of the top tier fails for the structural reasons already discussed. Fragments of the right answer are visible in different companies: some have got the supplier conversation right but not the digital platform, others have got customer commitments right but not the financial architecture, others have invested in services without changing the underlying cost structure. What none of them, in my experience, has done is bring the elements together into one coherent operating model.
The architecture below is the synthesis. Five elements that address the structural problem coherently, because the problems they address are interconnected. Addressing them individually produces partial improvement; addressing them together produces the financial profile that conventional EMS cannot produce, no matter how well it is run. The architecture is operating-model-redesigning rather than capability-additive. The company moves beyond conventional EMS by changing how it operates, not by adding what it does.
The five elements are:
Asset-light financial structure. Equipment moves to OPEX in all cases. Revenue-tied arrangements with equipment providers where they are available, conventional operational leasing where they are not. The fixed cost base reduces to staff, where capability is properly capacity-related and tightly controlled. Operational leverage starts working with the company through demand cycles rather than against it.
Consumption-based working capital with strategic suppliers. Components paid for as consumed, not held in inventory at the EMS’s expense. This requires supplier partners commercially structured to bear the inventory risk, in exchange for the market position the partnership gives them. For a mid-sized EMS, this is the largest single change to working capital intensity and the hardest single conversation to execute. It is also the most powerful, once it is in place.
Customer commitment tiers with consumption-based billing. Customer relationships structured around commitment, with three tiers. Strategic customers commit to annual volumes with adjustable monthly schedules, in return for meaningful rebates, capacity priority, and consumption-based billing terms that align cash flow with actual usage. Middle tier and transactional customers receive proportional terms. Above the committed base, remaining capacity is sold at premium pricing in periods of high demand. The customer-side terms mirror the supplier-side terms, so working capital flows symmetrically.
Paid services on top of manufacturing. Beyond manufacturing itself, a service portfolio with its own commercial logic. Portal-based self-service for routine customer work. Expert account engagement for consultative work that benefits from human judgement. Paid provenance and compliance documentation for customers in regulated end markets, where regulation is making this data competitively necessary. Intelligence-led commercial counsel. Lifecycle commitment with structured customer handover. These services scale on top of the underlying platform with better unit economics than additional manufacturing volume.
Digital platform absorbing the relay roles. A governed data layer connecting ERP, CRM, quoting, and shop-floor systems eliminates the coordination roles that absorb capacity in conventional EMS and grow proportionally with revenue. Internally this produces meaningfully lower headcount intensity at the same revenue compared to conventional EMS, with the margin improvement compounding through growth. Externally it powers self-service customer experiences and the paid documentation services on top.
Each element supports the others in specific ways. The asset-light arrangement reduces capital intensity but only works if customers and suppliers participate in consumption-based terms. The customer commitment tiers only produce predictable revenue if the digital platform can deliver on the capacity priority. The paid services only have margin headroom if the manufacturing platform itself is lean enough not to consume the value. This is what integration by design means in practice.
And the result is also De-EMSing in Eric’s sense. The company moves beyond conventional EMS. It is structurally different from contract manufacturing with administrative overhead. It produces revenue with different cash flow dynamics and a financial profile materially different from the EMS norm. It just achieves this by redesigning how it operates, not by adding new product or platform layers on top.
Why the moment matters
Two things are converging that make this conversation possible now in a way it has not been before.
The first is that the unsustainability is becoming visible at segment scale. Until recently, individual companies in difficulty could be explained as isolated cases: bad management, unlucky customer concentration, failed acquisitions, generational transitions handled poorly. With one third of German EMS losing more than 30% of revenue over two years, 16 insolvencies across France and Germany in 2025, and the smaller-is-worse pattern documented across multiple in4ma surveys, the individual-story explanation no longer works. The pattern is the data, and the data is making the structural reading harder to avoid.
The second is that the underlying enablers are now mature enough to actually execute the response. On the supplier side, large component suppliers with broad portfolios are increasingly willing to engage on consumption-based commercial terms in exchange for European market position. On the digital side, the platforms required to absorb the relay roles and power self-service customer experiences are now feasible to build in months rather than years. PCBWay and JLCPCB have proven the model at industrial scale. On the regulatory side, CSRD, CBAM, and supply chain due diligence regulations are pushing accountability for upstream provenance toward the manufacturer, producing a real paid service opportunity that did not exist five years ago. And on the customer side, expectations have shifted toward the kind of self-service-plus-expertise model that conventional EMS gives neither well.
Dieter’s own outlook in the 2026 annual survey is unambiguous: “a major breakthrough for the European EMS industry is not in sight. Asian markets continue to grow at more than 20% annually, underscoring the need for more strategic and intelligent transformation within Europe’s EMS sector.” The industry’s most respected data source is explicitly calling for transformation. The question for operators is what they intend to do about it.
And a more uncomfortable observation
Transformation of this depth is hard to undertake voluntarily. The conventional model is embedded in how the organisation works, how customers expect to be served, how suppliers want to engage, and how staff understand their roles. Telling a stable mid-sized EMS that the way it has operated for decades is structurally unsustainable, in a year when revenues are flat but profits are intact, is not a conversation most operators can win internally.
What unlocks the conversation in practice is difficulty. When a company has been through serious financial pressure, has lost customers, has had to lay people off, the organisation becomes capable of engaging with structural questions in a way that stable years make nearly impossible. The conventional response is to use such moments for restoration: get back to where we were before. The strategic response is to use them for transformation: become what the segment actually needs to become.
My own vantage point on this is direct. Tepcomp came through a difficult funding period that produced exactly this kind of reset. The organisation, customers, suppliers, and stakeholders all understood that the previous structure was not coming back. That created the conditions in which a meaningfully different operating model could be put in place, with cooperation from people who would not have engaged with the same conversation during stable years. I would not wish a crisis on any company, but having been through one, the conditions it produced are exactly the conditions that make transformation tractable.
The uncomfortable implication is at segment level. With many mid-sized EMS companies currently in difficulty simultaneously (visible in Dieter’s data, in the insolvency counts, in the layoff numbers), the segment is at a moment where collective recognition of the structural problem is more available than it has been before. The companies whose situations forced the question are now able to engage with the answer. The companies that look comparatively healthy are the ones for whom the structural change will be hardest to undertake, because their current performance lets them postpone the question. As one operator in the segment to others: the postponement is not safety, it is a slower version of the same problem.
Two paths through De-EMSing
The De-EMSing of the EMS industry, looked at across the full company size distribution, is moving in two directions at once. The top tier is adding capability: platforms, IP, vertical integration, hybrid manufacturer-product models. The mid-sized segment is, or has to be, redesigning operating models: asset-light, consumption-based, customer-committed, service-augmented, digitally enabled. Both move companies beyond conventional EMS. They serve different segments with different starting positions and different competitive contexts.
This matters for how the industry conversation is framed. The top-tier evolution is well-covered and well-understood. The mid-sized evolution is less covered, less recognised, and arguably the more urgent of the two, because it is the response to a documented financial sustainability problem affecting the segment that dominates the industry by company count. The companies executing this redesign are not consolation operators settling for a smaller version of the top-tier story. They are doing what their segment actually needs to do.
In4ma’s own positioning notes that the work “gives selected benchmarks to disillusion management that they are the best, but encourages them to optimize or change their strategy.” The data is doing the first half of that job effectively. The second half, encouraging mid-sized operators to actually change their strategy rather than to optimize within the old one, is the conversation the industry conversation has not yet fully engaged with.
That is the conversation worth having, and the segment dominated by mid-sized companies deserves to be central to it.
Richard Nilsen is CEO and Chairman of Tepcomp Group Oy, a Finnish-Estonian EMS rebuilding around the operating-model-redesign path described in this article. He has led and operated multiple EMS companies in the European market over the past two decades. The views expressed are his own and reflect a mid-sized European EMS operating perspective.
