Why Most Greenfield Factories Start Strong… and Then Struggle

A greenfield factory launch has a particular kind of electricity. Day One begins with machines that are technically perfect, teams that are still hopeful, and leaders who are suddenly everywhere—on the line, in the cutting room, on the phone to the buyer. Pilot runs are rehearsed, the first boards look neat, and the first shipment often goes out with a quiet sense of pride.

Then the applause ends. Somewhere between Month Three and Month Six, a surprising number of greenfields begin to wobble. Output starts fluctuating. Changeovers stretch. Supervisors start firefighting. The same factory that looked “ready” in the first 60–90 days begins to feel unstable. Not because it started badly, but because early success was often momentum-driven. Sustained performance is a different job: it requires system design, not just setup.

The early wins are real, but they are attention-driven

New factories often perform better than expected at the beginning for a reason that doesn’t show up in capex schedules: the organisation is in high-touch mode. Senior leaders are on the floor, so decisions that typically take days collapse into hours. If method engineering is unclear, the industrial engineer is right there. If a shipment is at risk, everyone knows before lunch.

The first order book helps as well. Greenfields rarely start with the hardest mix they intend to run long-term. Early styles are controlled—simpler constructions, fewer options, predictable fabric behaviour. Volumes are kept within what the early team can absorb, and external support quietly fills capability gaps.

Manufacturing research treats this as a distinct lifecycle phase rather than “business as usual”. In a survey-based study across 147 manufacturing companies, ramp-up recommendations are organised around preparation, conducting ramp-up, and a deliberate transfer into normal production, with collaboration, integration, and system robustness highlighted as central.  In apparel language, that transfer is the moment when extraordinary attention must turn into ordinary discipline. If the handover is premature, the factory starts scaling on a false baseline.

This is how the first illusion forms. Early efficiency looks decent and the factory ships on time, so everyone concludes stability has arrived. Buyers build trust quickly. Internal teams assume the learning curve is behind them. Momentum is mistaken for maturity.

The wobble begins quietly, then becomes structural

Greenfield struggle rarely arrives as a dramatic collapse. It arrives as variance. Some days hit plan, some days miss it, with believable explanations each time. The floor stays busy, but output stops behaving predictably. Those early fluctuations are not “noise”; they are the system telling you it still depends on attention.

The turning point is usually commercial. The factory moves from a protected start-up environment into a real operating environment. Order volume increases because early shipments created confidence. The style mix gets harder: more options, tighter windows, more volatility in size ratios and trims. The workforce expands quickly, and training begins to lag behind hiring. Under pressure, the factory reaches for the quickest lever it knows—moving people across lines—trading short-term coverage for long-term instability.

Apparel learning dynamics make this worse. An apparel learning-curve study describes how rapid product changes demand operator learning to reach steady-state performance, and how frequent product-type changes adversely affect production performance.  A greenfield is the most extreme version of that condition: operators are learning new styles while supervisors are learning, in real time, how to run the factory itself. If you don’t build a disciplined way to stabilise after each change—in style, manpower, or plan—variance hardens into behaviour.

When complexity enters, the factory is tested beyond its initial design

Many greenfields are managed as physical projects more than operating systems. Buildings get built. Machines get installed. Recruitment gets completed. The factory “exists”. But existence is not capability.

When complexity arrives—more styles, tighter shipping windows—the factory has to coordinate at a higher level. Planning must become precise. Pre-production has to become non-negotiable. Supervisors must manage flow, not just push pieces. If those disciplines are not designed and embedded, performance starts to slide even when “nothing is wrong” in the familiar sense.

That slide is visible in public greenfield programmes too. In Ethiopia, a World Bank case study of PVH Corp.’s supplier-led commitment in Hawassa Industrial Park describes spectacular early milestones: construction completed in less than a year and first exports soon after inauguration.  Yet the same body of work is explicit that the harder work is productivity, management capability, and the broader system that supports manufacturing at speed. 

A separate government-supported programme report on Ethiopia’s industrial parks makes the underlying point even sharper: investors underestimated skills shortages; factory productivity was described as around a quarter of the global average; and many factories were operating below capacity, with skills constraints undermining competitiveness.  You can build the shell quickly. You cannot shortcut the operating system.

The breakpoints that turn early success into later struggle

In practice, greenfields break in predictable places.

The first breakpoint is that “systems” exist as documentation, not behaviour. SOPs may be written and boards may be installed, yet routines are not embedded into daily work with clear ownership. Ramp-up performance measurement literature describes early ramp-up phases that can become dominated by trial-and-error decision making, with repeated iterations and unnecessary repetitions.  When a factory runs on heroic judgement, performance depends on who is present—not on what the system guarantees.

The second breakpoint is team architecture. Greenfields are staffed with people who can push to first shipment, but sustained performance depends on mid-layer depth: supervisors, line chiefs, planners, and industrial engineers who can run a stable rhythm without constant escalation. This is not a “people” story; it is an output story. In Bangladesh, Better Work reports that production lines managed by trained supervisors were 2.5 percentage points more efficient than comparable lines managed by supervisors without the training, and 4.3 percentage points more efficient than lines managed exclusively by male supervisors.  The lesson for greenfields is direct: if you don’t industrialise the supervisory layer, the factory’s performance remains attention-dependent.

The third breakpoint is pre-production discipline. During launch, readiness is treated like a ritual—methods are checked, inputs are chased, line loading is controlled. As the order book expands, readiness becomes reactive. Styles are loaded because the plan says they must be loaded, not because the factory is truly prepared. Instability enters through incomplete method engineering, missing attachments, late trims, and inconsistent feeding. The line may still “run,” but it stops behaving.

The fourth breakpoint is the learning curve that keeps resetting—often because people do. In a greenfield, workforce churn is not an HR statistic; it is a production constraint. A study by the International Growth Centre, working with a large garment firm in that industrial park, reported turnover so high that more than 40% of workers had left after just 12 weeks, while noting that workers were hardly productive in their first eight weeks because they needed training.  In plain terms: the factory is running on a training treadmill. You cannot stabilise output if your workforce is constantly restarting from near-zero.

I’ve seen a consistent mistake at this point. The factory interprets the wobble as a capacity issue and tries to solve it with expansion—more lines, more operators, more running hours. Output lifts briefly, then collapses back. The system never got the stabilisation phase it needed; it just got more surface area on which to fail.

The leadership shift that happens too early, and the missing phase most factories ignore

Greenfields fail more often because of timing than intent. During setup, leadership is deeply involved, decision cycles are short, and standards are enforced by proximity. During scale, leadership pulls back—sometimes because it must, sometimes because early success has convinced everyone the factory can now run. But the factory is often left to operate independently before it has earned operational independence.

This is when early success becomes a hidden liability. The business over-commits to buyers, assuming early results will hold under load. The next wave of orders arrives with more complexity than the system can absorb. And because the factory is now “live,” it pays for instability in the harshest currency: delivery reliability, overtime, expediting, and trust.

The conceptual mistake is that many factories think in two phases—setup and scale—when there is a missing phase that determines whether scale will stick: stabilisation. The ramp-up literature’s insistence on a structured “transfer to production” is a reminder that handover is not an administrative milestone (“go-live complete”), but a capability milestone (“the system is robust without extraordinary support”).  Stabilisation is where routines become habit, managers stop rescuing and start governing, and performance becomes predictable enough that commitments stop being gambles.

Strong greenfield factories do not behave differently because they have better infrastructure or more resources. They behave differently because they recognise that setup is only the beginning, not the outcome.

They treat stabilisation as a deliberate phase, not an accidental one. Instead of accelerating order flow immediately after go-live, they control it. The focus shifts from how much the factory can produce to how consistently it can deliver. Output is not allowed to fluctuate beyond a narrow band, even if it means slowing down growth in the short term.

 There is also a visible investment in the layer that most factories overlook—the mid-level operating team. Supervisors, line leaders, and industrial engineers are not just deployed; they are developed to hold the system together. The factory does not depend on a few strong individuals at the top. It builds depth where execution actually happens.

Pre-production is treated as a gate, not a formality. Styles do not enter the line unless they are fully ready—methods defined, inputs aligned, and risks anticipated. This discipline alone eliminates a large portion of the instability that typically shows up later on the floor.

Most importantly, these factories do not separate thinking from doing. The same teams that design the system stay involved in making it work. They remain close to execution until the system begins to behave on its own. Stabilisation is not handed over. It is built, layer by layer.

This is where most greenfield journeys diverge. Not in how they start, but in how they transition from setup to sustained performance.

Greenfield success is not difficult to create. With the right investment, the right people, and the right intent, most factories can reach their first shipment and even their first few months of performance with reasonable confidence.

What is difficult is building a factory that continues to perform when the initial energy fades, when complexity increases, and when leadership attention is no longer constant.

That transition—from a factory that is being driven to a factory that can drive itself—is where most greenfield projects struggle.

Because that transition is not about infrastructure. It is about system design, execution discipline, and the ability to translate intent into repeatable performance on the floor. 

This is also where the gap between traditional consulting and real impact becomes visible. Diagnosing problems is not enough. Designing frameworks is not enough. The real work lies in embedding those systems into day-to-day operations—until they hold, without external push.

Factories do not struggle because they start wrong. They struggle because they scale before they stabilise.

And stabilisation, in the end, is not a concept. It is execution.