Welcome to the 120th edition of Deep Tech Catalyst, the educational channel from The Scenarionist where science meets venture!
This week, I sat down with Roman Wolff, Industry Expert in Manufacturing Scale-Up, to unpack how an experienced manufacturing and engineering operator thinks about moving Deep Tech companies from the lab to industrial scale.
Key takeaways from the episode:
🧪 Scale-Up Starts When the Equipment Becomes Real
Moving from grams to kilos is an important technical step, but true scale-up begins when the equipment, operating logic, and process assumptions start to resemble the future plant.
🤝 Vendors Are Strategic Partners, Not Just Suppliers
Equipment vendors help founders understand what can actually scale, what data is needed, what performance can be guaranteed, and how to reduce unnecessary equipment risk.
📐 FEL Turns Technology Into an Investment Case
FEL1, FEL2, and FEL3 create the engineering discipline that moves a project from process concept to defined scope, cost estimate, site selection, vendor quotes, and final investment decision.
⏳ Industrial Scale-Up Takes Longer Than Most Startups Expect
For large projects, the journey from early scale-up planning to mechanical completion can take 4 to 5 years, with engineering, permitting, financing, construction, and ramp-up all shaping the real timeline.
👷 The Company Has to Be Built Before the Plant Starts
By the time the facility is mechanically complete, the operating organization must already exist. Manufacturing leadership, supply chain, maintenance, logistics, HR, and training all need to be ready before startup.
Why Scale-Up Is Not Just a Bigger Lab
In Deep Tech manufacturing, the term scale-up is often used as soon as a company moves from gram-scale experiments to kilogram-scale production.
That transition is meaningful. It allows the team to handle more material, generate larger samples for customers, and learn how the process behaves outside the smallest laboratory conditions.
From an engineering and manufacturing perspective, however, this stage is still often closer to an expanded lab environment than to true industrial scale-up.
The distinction matters because producing more material does not automatically mean that the company is learning what it needs to know to design, finance, and operate a commercial plant.
From grams and kilos to real manufacturing questions
At the early stage, the central question is usually whether the chemistry works. The company is focused on proving the core technical principle, generating data, and producing enough material for evaluation.
As the process moves toward industrial relevance, the questions become more practical and more operational.
The company has to understand the role of recycle streams. It has to evaluate reliability and determine how long equipment can run under realistic conditions. It has to consider materials of construction. It has to assess product stability, shelf life, handling, and storage.
These are the kinds of considerations that determine whether a process can become manufacturable.
They are also the factors that may remain invisible when the company is working only with gram or kilo quantities in a laboratory setting.
In that environment, many streams can be handled manually. Small losses may not matter. Equipment uptime may not be a central concern. Materials compatibility may not yet create a constraint. The process can look successful while still being far from industrially proven.
The shift toward scale-up begins when those hidden issues start to surface.
Why scale is defined by equipment, not size
A useful way to think about scale-up is that it is defined less by volume and more by equipment similarity.
The important question is not only how much product the company can make. It is whether the company is using equipment that can provide relevant information for the future plant.
This is why engagement with equipment suppliers becomes an important milestone.
Vendors understand the unit operations, equipment limits, sizing logic, and performance guarantees associated with their systems.
Their input helps the company understand what is commercially available, what the smallest scalable unit might be, and what assumptions need to be tested before larger capital decisions are made.
This also helps reduce equipment risk.
In a first-of-a-kind (FOAK) plant, the technology may already introduce enough uncertainty. The chemistry may be new, the process integration may be new, and the commercial model may still be developing. Adding custom or unproven equipment can increase the risk profile significantly.
The preferred approach is to keep the newness concentrated in the chemistry or process innovation while relying, as much as possible, on standard equipment.
That does not remove the need for engineering work. Standard equipment still has to be selected, configured, tested, and integrated into the broader process. But it gives the company a more credible path toward scale because the equipment has an industrial reference point and a supplier who can support it.
The first signs that a process is becoming industrial
A process starts becoming industrial when the team begins to describe the full plant, not only the core chemistry.
That includes the feedstock coming in, the product going out, and every intermediate stream that has to be handled along the way.
A practical way to frame the problem is to ask what a truck will bring into the site and what a truck will take away from it.
That question changes the level of analysis.
It forces the team to define feedstock specifications, product specifications, byproducts, recycle streams, waste streams, storage requirements, and logistics.
It also reveals practical issues that may have been invisible in lab work: how to load a catalyst, how to separate solids from liquids, how to move material between steps, how to manage recycles, and how to keep the process running reliably.
These details can look secondary from a research perspective. From a manufacturing perspective, they are central. They influence the process design, the cost estimate, the pilot strategy, the vendor conversations, and ultimately the investment case.
A company becomes more credible when it can show that it understands this transition.
The move from lab work to scale-up is therefore less about declaring that more material has been produced and more about demonstrating that the process is being translated into an industrial system.
Designing the Scale-Up Path Backward from the Commercial Plant
A practical scale-up roadmap starts from the end state.
Before deciding what the next pilot should look like, the company needs to form an initial view of the commercial plant it is eventually trying to build.
That view will still be incomplete and uncertain, but it gives structure to the work. It allows the team to understand which questions must be answered, which equipment must be tested, and which assumptions will drive the techno-economic model.
At this stage, the company may already have a product concept, a potential customer, and some form of commercial signal. A customer may have indicated interest in buying a defined volume at a certain price, or may have provided enough information to begin estimating demand. That creates the basis for early economics.
The first task is to turn that commercial signal into a manufacturing problem.
The company has to understand what product specification it must reach, what feedstock it can actually buy, and what the plant must do in order to convert one into the other. The result is an early picture of the commercial facility, built backward from the market requirement.
Starting with the product specification and the feedstock specification
The product specification is one of the first anchors of the scale-up process.
If there is a real customer or a credible market pull, the customer defines what the final product must be. The product has to meet a specification, and that specification becomes the target for the process design.
At the same time, the company has to look carefully at the feedstock.
In the lab, materials are often purchased in forms that are convenient for research. They may be high purity, available in small quantities, or selected because they allow the chemistry to be tested cleanly.
A commercial plant operates under a different logic. The feedstock has to be something the company can actually purchase at scale, with real specifications, real availability, and real supply constraints.
This is an important shift in thinking.
The company is no longer asking only whether the chemistry works with ideal inputs. It is asking whether the process can work with materials that can be supplied to an industrial facility. That feedstock specification becomes part of the economic and technical model.
If the feedstock is unrealistic, the rest of the analysis becomes fragile. If the product specification is unclear, the process has no stable endpoint.
Both sides of the equation need to be defined early enough to guide the design.
Mapping each step of the future plant before building the pilot
Once the company has an initial view of the commercial plant, each step in the process needs to be assigned to a type of equipment.
At the beginning, there may be more than one possible equipment choice for a given step. That is expected at an early stage. The important point is that every step is being connected to an industrial unit operation.
This process also defines the scale-up path.
When the company speaks with equipment suppliers, the vendors can help identify the commercial-scale equipment and the smallest scalable version that can be purchased for pilot work.
That smallest scalable unit often determines the size and structure of the pilot.
The design should be guided by the questions that must be answered for the commercial plant.
If a certain equipment type is likely to be used at commercial scale, the setup should generate the data needed to validate that choice.
If recycles will matter in the plant, the pilot should help the company understand them.
If reliability, stability, or materials of construction are important, the pilot should create conditions where those questions can be studied.
This is also where early techno-economic model begins to become more grounded, and it becomes a working tool for improving the design.
At this stage, the company may have enough information to prepare a funding proposal for the next phase of work.
That funding is usually meant to support the pilot, the first level of engineering work, and the company’s operating runway while those activities are being completed.
This is still an early investment case. It is not yet the funding case for the full commercial plant. There are too many uncertainties. The process still needs to be tested, the design still needs to mature, and the risks need to be reduced step by step.
How FEL Turns Technology into an Investment Case
Front-end loading, often described through FEL1, FEL2, and FEL3, is the process that progressively turns a technical concept into a Final Investment Decision (FID).
In the early stages of a Deep Tech company, many parts of the industrial plan are still fluid. The chemistry may work. The product may have customer interest. The company may have an initial view of the commercial plant.
However, the level of engineering definition is still too low to support a major capital decision.
The FEL process creates structure around that uncertainty.
Each phase increases the maturity of the design, reduces the range of cost uncertainty, and forces the company to make decisions that were previously open.
The work moves from a high-level view of the process to a detailed definition of what will be built, where it will be built, which equipment will be used, and how much the project is expected to cost.
This matters because the investment case has to show that the process has been translated into a defined project.
FEL1: fixing the process logic and learning while it is still cheap
FEL1 is the stage where the company fixes the basic process logic. At this point, the block flow diagram is defined.
The company knows the main steps required to move from feedstock to product. It may still be converging on the exact scope, and many details remain open, but the overall route is becoming clear.
The cost estimate is still relatively broad because the design is still early. The purpose of FEL1 is not to create final precision. It is to define the process architecture well enough to understand what must be studied, tested, and improved.
This is also one of the best stages for learning. Changes are still relatively inexpensive.
The company can update assumptions, improve the flow diagram, revisit alternatives, and use feedback from vendors, engineering partners, and pilot work to strengthen the design.
The goal is to create enough clarity to move into the next phase with a more disciplined scope.
For Deep Tech companies, this phase should also interact closely with the pilot strategy. FEL1 and pilot planning can inform each other. As the company learns more from equipment suppliers and early testing, the engineering view can be refined.
That cycle is valuable because it helps identify problems before they become embedded in a more mature design.
FEL2: converging on scope and completing the technology work
FEL2 is where the process moves from block flow logic to a more defined process flow diagram.
At this stage, the company is no longer only defining the major steps. It is beginning to decide how those steps will be performed.
This is where scope convergence becomes critical. The company needs to define the type of equipment that will perform that operation.
FEL2 is also the stage where alternatives are reviewed and major choices are narrowed.
The company should be locking in site selection, or at least moving toward a level of site definition that allows the next phase of design to proceed. It should also be completing the pilot and technology development work that is required for the project.
This point is especially important.
By the end of FEL2, the company should not expect to continue fundamental technology development for that specific plant design. The pilot should have reduced the key technical risks. The process data needed for the project should be available.
The major questions that could change the design should have been answered.
FEL2 therefore acts as a gate between technology development and detailed project definition. It is where the company proves that the process is mature enough to support a more precise engineering effort.
FEL3: locking the design before the Final Investment Decision (FID)
FEL3 is the stage where the project becomes highly defined. At this point, the company should know what it is building.
The work is focused on the details required to support execution.
This includes piping and instrumentation diagrams, plot plans, 3D models, vendor selections, and vendor quotes. The scope is fully locked. The design has moved from a conceptual or semi-defined process into a project that can be costed, reviewed, and prepared for execution.
At FEL3, the company should not be using engineering to solve unresolved technology questions. It should be using engineering to finalize a project whose technology basis has already been validated.
That distinction is central to capital efficiency.
Late-stage design changes are expensive. If a company discovers during FEL3 that the pilot data is insufficient, the site assumptions are unstable, or a core unit operation has not been proven, the schedule and budget can be significantly affected.
The earlier stages exist to prevent that situation.
FEL3 creates the technical and cost basis required for the investment decision. It gives the stakeholders enough definition to understand the scope, the capital requirement, the project risks, and the path to construction.
The final investment decision (FID)
The final investment decision is often discussed alongside the engineering process, but it is fundamentally a financial milestone.
Engineering creates the conditions for FID. It provides the defined scope, mature cost estimate, vendor support, site information, and project design required to make a capital decision. However, the decision itself is about committing capital.
At FID, the company or its financial backers decide whether to fund the project and move into execution.
For a large industrial plant, that may involve not only the direct capital cost of construction, but also fees, interest, company operating costs during the construction period, and the capital needed to support the organization until the plant is producing.
This is why the quality of the FEL process matters so much. For a Deep Tech company, this is the point where technical progress and financial readiness begin to converge.
The Real Timeline of Industrial Scale-Up
Industrial scale-up requires a timeline that is often longer than early commercial discussions suggest.
Once a company has customer interest, an initial product specification, and a view of the process, it can be tempting to think of the next step as a direct move toward production.
In practice, the path from early scale-up planning to a commercial plant includes several parallel workstreams: engineering, piloting, site selection, permitting, financing, construction, hiring, and commissioning.
The duration depends on the size and complexity of the project. A lower-capex project may move faster. A larger first-of-a-kind plant will usually require more time, more definition, and more capital discipline.
For a large industrial project, the path to mechanical completion can easily extend across 4 to 5 years if the sequence runs well.
That timeline includes the engineering phases, the final investment decision, detailed engineering, construction, and the early work required to prepare the organization that will operate the plant.
Why a commercial plant can take 4 to 5 years to reach mechanical completion
The front-end engineering process alone can take more than a year and a half for a large project.
FEL1 may take 3 or 4 months. FEL2 may take roughly 6 months. FEL3 may take around 12 months, assuming the required information is available and the project does not have major unresolved issues.
Those timeframes are indicative, and they vary by project, but they give a realistic sense of the planning horizon.
After FEL3, the company reaches the point where it can move toward a final investment decision. That decision can take additional time, potentially several months, because it involves financing, risk review, commercial agreements, and the final commitment of capital.
Once the investment decision is made, detailed engineering and construction begin.
For a large project, that phase can take approximately 2 years, assuming there are no unusual constraints around long-lead equipment or other major delivery issues.
Taken together, the industrial timeline becomes substantial.
The company may spend around 18 months or more reaching the end of FEL3, additional time moving through the FID, and roughly 24 months on detailed engineering and construction.
This is how a project can become a four-to-five-year journey from the decision to pursue scale-up to mechanical completion.
How permitting, site selection, engineering, and construction move together
Several activities have to move in parallel, and delays in one area can affect the rest of the timeline.
Site selection is one of the most important examples. The company needs a defined site to complete the design properly. Without a site, many engineering assumptions remain unstable. Layout, utilities, logistics, permitting requirements, and other site-specific design assumptions all depend on where the plant will be built.
Permitting also affects the schedule. A company typically needs enough engineering definition before it can apply for permits. That means permitting often begins around FEL3, when the project has enough detail to support the application. The permitting process then runs alongside the later project work and should ideally be sufficiently advanced by the time the company is ready to proceed after the FID.
Construction timing also depends on the equipment profile. If the project requires equipment with long delivery times, the schedule may extend. If there are unusual components, custom systems, or supply chain constraints, the construction plan has to absorb those realities. In a FOAK plant, these dependencies need to be managed carefully because the project is already carrying technology and execution risk.
The overall timeline therefore has to be treated as an integrated plan.
Engineering is connected to site selection. Site selection is connected to permitting. Permitting is connected to the investment decision. The investment decision is connected to procurement and construction. Each phase creates the conditions for the next one.
Why first product and design-rate production are different milestones
Mechanical completion does not mean the plant is already producing at full commercial performance.
Once construction is complete, the company still has to start up the plant, commission the systems, bring materials into the facility, and begin operating the process under real conditions.
For FOAK technologies, this ramp-up period is particularly important because unexpected issues often appear only when the plant is running.
The first product may come relatively soon after startup. In a successful case, the first truck leaving the plant could happen perhaps a month after startup.
However, reaching design-rate production can take longer.
A reasonable planning assumption may be 3 to 6 months after mechanical completion, depending on the complexity of the process and the issues that appear during commissioning and early operations.
This distinction matters because early production and stable production are different operating states.
A plant may demonstrate that it can reach the intended design rate early, but then experience equipment failures or reliability problems. These issues can be highly specific.
Gaskets may leak because the selected material was wrong. Certain components may behave differently under continuous operation than they did in the pilot.
These are normal learning points in early plant operation.
The pilot reduces risk, but it cannot reveal every issue that will appear in the full-scale facility. Some problems only emerge when equipment is connected at scale, materials are moving continuously, and the plant is operating as an integrated system.
This is why the timeline should include a realistic ramp-up period.
The company needs time to troubleshoot, modify, repair, optimize, and stabilize the plant. From an investment and customer perspective, the relevant milestone is not only first production. It is the ability to operate reliably and approach the intended production rate.
For Deep Tech companies, this has direct implications for cash planning.
The company needs enough capital not only to design and build the plant, but also to operate through commissioning and ramp-up. The period between mechanical completion and stable production can still consume cash. If the plan assumes immediate full-rate production, the financing model may become too optimistic.
A realistic industrial roadmap therefore separates the major milestones clearly: engineering completion, FID, mechanical completion, first product, and design-rate production.
Each milestone represents a different level of readiness, and each requires its own assumptions about time, capital, and execution risk.
Building the Operating Company Before the Plant Starts
The construction phase is not only a period for building the physical plant. It is also the period in which the company has to build the organization that will run it.
By the time mechanical completion arrives, the plant cannot be treated as a finished asset waiting for a team to appear around it.
The company needs the people, systems, procedures, supply chain, logistics, and operating structure already in place. Otherwise, the project may reach physical completion without being ready to operate.
This is one of the most important differences between a technology company preparing for scale-up and an industrial company preparing for production.
A plant is not only equipment. It is an operating system.
Why offtake agreements matter before the investment decision
Before the final investment decision, the company should ideally have its offtake agreements in place.
This is especially important for large capital projects. If a project requires hundreds of millions, or even more than a billion dollars, the investors or lenders backing the plant will want to know who is buying the output, under what terms, and for how long.
A potential buyer’s interest is useful, but it is usually not enough to support a major plant investment. The financial backer will likely want a committed customer, often under strong contractual terms, for a meaningful share of the plant’s production.
That agreement becomes part of the financing logic.
It shows that the project is connected to demand. It reduces uncertainty around revenue. It gives investors more confidence that the plant, once completed, will have a market for its output.
Hiring the manufacturing leader early
The operating organization should begin forming before the plant is finished.
One of the first important hires is the manufacturing leader, whether that role is called site director, plant manager, or another equivalent title.
This person should join early enough to influence the engineering and design process, not only to manage the facility after construction.
That timing matters.
An experienced manufacturing leader can provide practical input that improves the plant design. They can identify operational issues, maintenance concerns, staffing needs, safety considerations, commissioning requirements, and design choices that may look acceptable on paper but create problems in day-to-day operation.
This is particularly valuable in a FOAK plant.
Engineering teams can design the system, but operating leaders understand what it takes to run the system continuously. They bring the perspective of reliability, maintainability, procedures, staffing, and plant discipline.
Hiring this person early also creates a bridge between the project team and the future operating team. The plant is then being designed not only as a capital project, but as a facility that people will have to operate safely and reliably.
Why the ramp-up exposes what the pilot could not reveal
Even with a strong pilot, good engineering, experienced vendors, and a capable operating team, the first commercial plant will still teach the company new lessons.
FOAK technologies tend to fail in unexpected ways. The pilot can reduce risk, but it cannot perfectly reproduce all the conditions of the full plant.
When the equipment is larger, the streams are connected, the process runs continuously, and real production targets apply, new issues can appear.
Some of those issues may be mechanical. Other issues may involve reliability, maintenance, startup procedures, operator training, or supply logistics.
This is why ramp-up should be treated as part of the scale-up plan.
The company should expect a period of troubleshooting and stabilization after mechanical completion. First product may come relatively early, but stable production at design rate takes more time.
The operating company therefore has to be ready before the plant starts.
It needs the people who can identify problems, fix them, document them, and keep improving the system. It needs the procedures and supply chain needed to support continuous operation. It needs enough capital runway to move through commissioning and early production without assuming that everything will work perfectly on day one.
The broader lesson is that scale-up is not only an engineering sequence. It is an organizational buildout.
The company has to mature at the same time as the plant. By the time the facility is mechanically complete, the business must already have become capable of operating it.
