FPP Amsterdam Speaker Interview: Building the plane while flying it – David Ziskind on smarter facilities for novel proteins

Veteran facility designer David Ziskind has helped dozens of novel protein companies move from lab bench to factory floor. At The Future of Protein Production Amsterdam 2025, he’ll share why scaling isn’t about gleaming tanks or futuristic control rooms, but about making deliberate choices under real-world constraints

When David Ziskind talks about manufacturing, he doesn’t start with gleaming stainless-steel tanks or futuristic control rooms. He starts with constraints: capital, time, risk. As Managing Partner of Mach Global Advisors – and a veteran of designing facilities for dozens of novel protein companies – he has seen both sides of the scaling equation.

“I talk about minimum viable facilities in terms of core elements,” he begins. “What’s absolutely essential to produce your product safely and at the quality you need, versus what would just be nice to have.”

That framing – what he calls being “scrappy yet focused” – has become a survival principle for startups moving from lab to pilot to commercial scale. And it’s a theme Ziskind will explore when he takes the stage at The Future of Protein Production Amsterdam 2025 on 29-30 October 2025.

The bigger-beaker fallacy

Ziskind sees the same mistakes crop up repeatedly. The first is assuming that scaling is as simple as multiplying equipment size. “I call it the ‘bigger beaker’ fallacy. Companies think if it works in the lab, they just need larger versions of everything. But biological processes don’t scale linearly – you often need completely different equipment behavior at commercial scale.”

Another blind spot is underestimating how much of a budget goes into the facility itself, not just the shiny bioreactors. “Fermentation tanks and downstream skids might look like the big ticket items, but the building, utilities, and sanitary design usually cost more. And that’s before you factor in timelines. Companies try to skip front-end engineering to save money, then burn through months and millions fixing preventable problems.”

In his experience, proper front end engineering design (FEED) consistently saves around 30% on total project costs in addition to improved project execution times. But convincing a cash-strapped startup to spend US$150,000 on upfront planning can be tough. “They think they need to just start building,” he says, “but that decision often determines whether you’ll ever reach commercial production.”

“Fermentation tanks might look like the big-ticket items, but the building, utilities, and sanitary design usually cost more”

In-house versus outsourcing

The question of whether to build facilities or lean on contract manufacturers has no one-size-fits-all answer. Ziskind approaches it with a simple framework: identify the part of your process that defines your competitive edge, and keep that in-house.

“I often recommend a hybrid approach – make the cake mix that matters, then let someone else bake it. One Series B company was determined to build its own plant. The numbers showed it would eat up 80% of their funding before a single product launch. Instead, we designed a hybrid where they produced their substrate in-house and outsourced downstream processing. That cut their capital requirement by 60% and got them to market a year faster.”

The decision hinges on risk tolerance, speed, and internal expertise. “Building a plant while developing your core technology is like building a plane while flying it,” he says. “It’s possible, but you’d better be deliberate.”

Designing for uncertainty

If there’s a unifying thread in Ziskind’s philosophy, it’s planning for uncertainty. “Don’t size equipment for a single-point estimate. If your viscosity could be X or 3X, design for 3X. It’s much cheaper than retrofitting later.”

He calls this “intelligent oversizing” – especially for utilities such as electrical feeds, water, and compressed air. The incremental cost upfront is modest; the cost of ripping and replacing later can cripple a business.

Data gathering is another lever. “I helped one company spend US$50,000 on targeted experiments. That eliminated US$2 million of construction uncertainty and changed investor perception overnight. Six months later, they closed their Series B oversubscribed.”

“Hope is not a strategy. The companies that win make informed trade-offs, plan deliberately, and turn unknowns into knowns before they build”

Future-proofing without overbuilding

Talk of ‘future-proofing’ facilities can sound like an invitation to overspend. But Ziskind cautions that it’s really about designing for adaptability, not predicting the future.

“Flexible utilities, modular equipment, and planning for the pivot – that’s the recipe. Maybe you oversize your electrical backbone, or you design cleanrooms that can be subdivided differently. But you can’t future-proof everything. The smart companies pick where flexibility matters most based on their technology and market risks.”

With funding conditions tighter than in previous years, startups often ask where they must spend versus where they can defer. Ziskind’s rule: never cut corners on design, food safety, or front-end engineering. “Scrimp there and you’ll pay multiples later.”

What can wait? “Automation can come later. Start with manual minimum viable processes if needed. Premium finishes, redundant systems – those can be deferred. Even some utilities can be leased as a service initially, which raises operating costs but preserves capital for what really matters.”

Creative solutions are increasingly common. Auctions and canceled orders can provide equipment at 40-60% discounts. Leasing and shared infrastructure deals are gaining traction. “I’ve seen clients convert warehouses into food-grade plants using a ‘box-within-a-box’ approach for a quarter of the cost of greenfield construction.”

Collaboration with caution

Shared production facilities can accelerate progress, but Ziskind warns they’re not for everyone. “Collaboration works when companies can share infrastructure without giving away their secret sauce. I’ve seen successful shared fermentation sites where upstream systems are common but downstream purification is separate.”

But the risks are real: intellectual property exposure, mismatched priorities, conflicting regulatory timelines. “I’ve watched partnerships implode because one company optimized for cost while another needed speed. You need clear governance, predefined capacity allocation, and exit strategies that don’t strand anyone.”

Some of the most decisive design choices are invisible on opening day. Sanitary design, for instance, can dictate operating costs for years. “Bad drain placement or hard-to-clean surfaces can add hours to changeovers and create food safety risks. I’ve seen plants where cleaning, not production, was the bottleneck.”

Automation is another balancing act. Too much, too early, locks companies into costly maintenance and rigid processes. Too little, and labor bottlenecks crush throughput. The sweet spot, he says, is semi-automation that can be upgraded later.

‍A tale of two startups

Ziskind recalls two nearly identical precision fermentation startups. Company A spent US$150,000 on front-end engineering; Company B skipped it.

“Eighteen months later, Company A was on budget and six months from commercial launch. Company B was 300% over budget, still debugging basics, and had to spend over US$1 million fixing design flaws. The kicker? Their layout made it impossible to add a purification step they suddenly needed. Company A had built that flexibility in from the start.”

Asked for his single piece of advice to today’s startups, Ziskind doesn’t hesitate. “Plan deliberately, but don’t wait for perfection. The companies that win make informed trade-offs aligned with their strategy. And they turn unknowns into knowns through smart testing. Hope is not a strategy.”

“Building a plant while developing your core technology is like building a plane while flying it”

Looking ahead five to 10 years, he predicts more modular, data-driven approaches, tighter integration of digital tools, and greater collaboration across the sector. “The shift is from just building capacity to building adaptive capacity. The technology will evolve, but the fundamentals won’t change: deliberate planning, informed trade-offs, and focusing on what creates real advantage.”

At its core, his message is simple but hard-earned: building the future of protein production isn’t about building perfect facilities. It’s about building the right ones – for today, and for the uncertainty of tomorrow.

David Ziskind is one of more than 100 speakers taking to the stage at The Future of Protein Production Amsterdam on 29/30 October 2025. To join him and more than 1,000 other attendees, book your conference ticket today and use the code, 'PPTI10', for an extra 10% discount on the current rate. Click here