Lizzy Manzi spent seven years navigating changing requirements at SpaceX, where she eventually led construction across Los Angeles and Seattle, with a team covering roughly 50 buildings. She was responsible for building out manufacturing, test, reliability, R&D, logistics, office spaces - pretty much any infrastructure the company needed. When she finally left SpaceX, former colleagues who had started or joined hardware companies began calling her. They needed help finding buildings, hiring electricians, understanding permits, and figuring out how to build for production.
“They’re rocket scientists. They’re incredibly talented engineers,” she said. “But they usually don’t know construction. And they shouldn’t.”
Those calls eventually led her to start Ground Control in 2024, which began as consulting services and grew into factory management software. Through expert services and purpose-built software, her team helps hardware companies decide what infrastructure to build now and what to leave flexible as their production needs evolve.
In our conversation, we discussed what goes into building a company’s first factory, why flexibility is so important, what that actually looks like in practice, and how she sees the future of managing constantly changing factory floors.
The hardest part is defining what you want
When I asked Lizzy what most often prevents factories from expanding quickly, assuming they have the demand and the money, her answer came back to the company itself.
“I hate to say, a lot of times it’s defining what you want.”
Defining a factory’s scope requires a series of decisions that depend on one another. For example, the equipment needed is based on the product you’re manufacturing. And the power requirements and floor layout are dictated by what equipment you choose. This ultimately then determines what buildings could work and how much construction is needed to set that machinery up.
It all stems from the first decision around what you’re making, and sometimes that is still evolving. This can also be a question of what production happens in-house and what can be outsourced. “If this part isn’t designed yet, how do I know the machine that I need to build that part?” Lizzy described working with a company that had several possible machines in mind for each equipment category. Each had very different requirements and so planning was a series of assumptions and deciding when to be conservative versus when a middle-of-the-road estimate was reasonable.
It’s a balance between developing a plan that is grounded enough to act on, but still leaves room to evolve as the company learns and scales. Informed future-proofing.
“The best project is no project”
“Elon would say the best part is no part,” Lizzy said. “I say the best project is no project.”
A lot of the work in Lizzy’s role at SpaceX and today at Ground Control is simply understanding the underlying need that’s driving a project. This usually means identifying what work helps the business move forward. At SpaceX, her team managed many projects simultaneously, and her prioritization was based on, as she put it, “which group is the bottleneck right now?”
But then, even when a need is clear, there’s still the question of how much net new infrastructure is actually required. Lizzy gave the example of an engineer who thought they wanted a full ISO clean room. After understanding the cost and schedule implications, the team decided that temporary plastic walls and a rental HVAC unit were enough to support the immediate milestone.
“Figure out what functionality we actually need, and then build that,” Lizzy said. Temporary solutions can buy time as well as information. It’s about designing for flexibility. This usually means keeping space open as much as possible, thinking ahead about electrical distribution, and avoiding permanent infrastructure, or “monuments,” that could inhibit future expansion.
Part of designing for flexibility also means accepting the possibility of rework. At SpaceX, Lizzy’s team would procure equipment with long delivery times while design and permitting continued in parallel. “We were pretty good at redoing stuff,” she joked. But it’s about transparency around risk - what could change and what the company might have to absorb if it did. Moving quickly depends on quantifying risks upfront, accepting them (or not) as a team, and taking action.
At the core of all of this, though, is one simple question that Lizzy kept coming back to in our conversation: “Are you making this harder than it should be?” Often, the fastest way forward is to give yourself less to build, and less to undo.
The Goldilocks problem
When a team chooses its first factory, it has to commit to space before knowing how quickly the business will grow. Too small, and it risks needing to move prematurely; too large, and it risks paying for capacity it can’t use yet.
Assessing options for space from brokers is much more than just comparing rent and square footage. Lizzy rattled off a long list of other factors that are equally important for teams looking to build their first factory: “Does this building have enough power? Is it 480V? Does the outside transformer match the inside switchgear? Are the doors big enough? What’s the concrete slab look like? Are there fire sprinklers? Is the zoning correct? Is the occupancy correct?”
When I asked Lizzy what teams most often need help with, she said power. “It’s one of the biggest potential surprises,” she said. Power requirements depend on what machinery is being used and what cycle times look like. Teams can overestimate how much power they’ll need, or badly underestimate it. As Lizzy put it, “In reality, you’re probably never going to be running all of the machines full bore at the exact same time.”
Lizzy advises teams to first understand what electrical service is available today, whether it can support their expected growth, and what an upgrade would involve. If additional capacity is needed, assuming the utility company can support it, she recommends budgeting roughly a year.
Power also interacts with a factory’s layout. Placing high-power equipment close to the electrical service can save hundreds of thousands of dollars, but also impacts production flow. “Power distribution across a large facility can be an extremely expensive project,” Lizzy said.
A building’s occupancy type is also a factor in how much has to be invested into building out a factory. For example, a warehouse that was used for storage that is being converted to manufacturing probably needs changes to fire protection, restrooms, parking, and exits in order to accommodate the increased amount of people and activity. These are the types of costs that aren’t obvious when comparing property listings, but come out when you submit to a city for building permits.
By identifying these costs upfront, Ground Control helps companies understand how much work is needed before production can start, and which projects can wait. “What do we need to achieve on day one?” she asks. “From there, you can think about how you phase the rest of your projects over time.”
When the factory outgrows the spreadsheet
Another pain point in a hardware company’s real estate journey comes when they finally have enough space and equipment, but now they have to keep track of everything.
Lizzy calls this the point when the business is “not spreadsheetable.” Usually there is an out-of-date CAD file stored somewhere. Someone will screenshot it, paste it in PowerPoint, and draw rectangles for where new machines go, or where existing machines will move to. The corresponding equipment specifications are tracked in Excel. Those files then get passed around to architects, electricians, general contractors, or other stakeholders, leaving that person to reconcile the markups with the equipment list. “You’re cross-eyed looking at this whole thing,” she said.
A CAD drawing doesn’t usually help alleviate this chaos because only a handful of people know how to update it, or have a license. They become “the keeper of the CAD,” while the rest of the team responsible for things like machining, inventory, offices, etc. still needs to collaborate and contribute to the plan.
Ground Control addresses this by bringing the factory layout and the information about what’s inside it into one place. A machine’s specifications, such as its make, model, size, power requirements, cut sheet, history, point of contact are all visible from a single floor plan. “I’ll click a box. It says the machine name with a photo of it. It says the power it needs. And I can see the rest of the building’s power distribution and know where it’s fed from.” This makes it easier to translate a production team’s needs into work that other people can evaluate and execute.
The platform is also designed to be used and updated by different stakeholders. Facilities managers, construction managers, and production engineers can have permission to move things on the floor plan. Anyone in the company can find assets, people, conference rooms, and report facilities issues or request projects. This ensures the plan is useful and relevant beyond the initial buildout.
Ground Control’s consulting work shapes which problems its software should address. Working directly with companies gives the team repeated exposure to where manufacturing companies get stuck. “What are their problems? Where do they struggle? What are the SOS moments along their journey?” The goal is to make routine coordination easier, while keeping a human in the loop for the moments that need it.
Over time, this repository of information can help teams assess bigger investments. If a group asks for another 50,000 square feet, the company can first look at what equipment it already has, where it sits, and how space is being utilized. This helps teams make better decisions around whether to expand or reorganize what they already have, and brings Lizzy’s “best project is no project” philosophy into the software itself.
Leaving room to change
Flexibility takes planning. The hard part is knowing which investments are needed to start production today versus which commitments can wait until a team knows more. These choices also need to be documented and referenceable as the company grows, rather than left to tribal knowledge. After all, the person planning future expansions may not be (and likely isn’t) the one who designed the first production line.
This is where Ground Control’s combination of expertise and software can help teams. The consulting work enables teams to make better-informed initial decisions, while the software gives the people using and modifying the factory a shared place to work from and a traceable source of truth. Keeping the requirements, changes, and history connected to the space means each new decision is better informed. Over time, the company builds an understanding of why its factory works the way it does, and what would need to change to support something different.
“You’re going to continue to layer in different needs as the company learns more about their product, their business model,” Lizzy said. Getting the first factory up and running is one milestone in that process. The company will keep learning, growing, and uncovering unknowns, and both the building and the knowledge behind it need room to evolve.
Author’s note: An LLM was used for light copy editing only (spelling, grammar, and clarity). Content, meaning, tone, and structure remain unchanged.