The FOAK Location Selection Framework: Lessons from Choosing a Gigafactory Site

In the end, it had to be a political decision.

After half a year of sifting through more than twenty locations and countless meetings with local officials, our future 8 GWh factory came down to two options.

• Tatarstan: a highly industrialised region east of Moscow, with a strong industrial park, permits ready, and—most importantly—right next to our main customer, Kamaz.

• Kaliningrad: a Russian exclave in Europe, wedged between Poland, Lithuania and the Baltic Sea, on the unfinished site of a nuclear power plant.

Every rational criterion pointed to Tatarstan: customer proximity, supply chain, logistics, readiness, taxes, and even seasonal conditions. Kaliningrad won, and on the face of it it was for one reason: money. The regional government offered to fund over 20% of the project.

But the deeper truth was political. The abandoned nuclear power plant site belonged to Rosatom, our sole shareholder. It had become a political liability, too expensive to complete and too costly to abandon. Repurposing it for a lithium-ion factory was a convenient way to write a new story.

That decision came at a price. The location was far from our customer, the site was incomplete, workforce relocation was inevitable, and logistics were riskier. I signed the agreement in September 2021. Six months later, Russia invaded Ukraine, and every hidden risk of Kaliningrad was realised at once.

Me, (on the left), signing an agreement with Kaliningrad region governor, 2021

The FOAK Location Selection Framework

Your demo or FOAK (First-of-a-Kind) site is one of the most consequential choices you will ever make. Costs, operations, timelines, and sometimes survival depend on it. Politics, investors’ agendas, or personal bias can tip the balance away from logic. My advice: use a structured framework and watch for irrational influences.

Here’s the FOAK location selection framework I wish more teams followed:

1. Stay close to your primary customer

• Lower delivery costs: your FOAK prices will already be high.

• Faster iterations: constant feedback and site visits save months of emails and Teams calls.

• Visible commitment: being near your anchor customer builds trust in the riskiest stage.

  
  

2. Balance against supplier proximity

• If inputs are easy to ship, prioritise customer proximity.

• If inputs are heavy, hazardous, or hard to move - consider suppliers first.

• Factor in CBAM, tariffs, and carbon footprint.

• Stress test for geopolitical risks, lockdowns, and trade disruptions.

  
  

3. Run a complete logistics risk assessment

• Quote current costs, then check historical spikes and their causes.

• Map transport bottlenecks, political choke points, and customs delays.

• Think of tail risks: blocked canals, fragile bridges, and sanctions.

  
  

4. Test the physics and infrastructure of the site

• Weather windows for construction works (snow, heat, floods, storms).

• Site topography and environmental restrictions.

• Power, water, wastewater, roads - are they already built or just promises?

  
  

5. Check labour and living conditions

• Is the workforce available locally?

• If not, what’s the cost of relocating and housing them?

• Don’t forget families: schools, healthcare, recreation.

  
  

6. Understand local politics

• Meet officials early to learn pain points, local factions and rivalries.

• Map potential blockers (farmers, NGOs, NIMBY groups).

• Look for incentives: land, permits, training centres, financing.

  
  

7. Move quietly and fast

• Don’t signal commitment too early — land prices spike, ownership shifts, and projects stall.

• Stay in stealth until the lease or purchase is signed. Only then go public.

Conclusion

Industrial parks can sometimes shortcut many of these steps—premade infrastructure, pre-cleared permits, eager local officials. But even then, walk through the checklist.

It is always cheaper to catch a mistake on paper than halfway through pouring concrete.

And remember: not every decision will be rational. Some will be political. Your job is to know the difference, and to push logic as far as it will go before the politics take over.

👉 I’m writing a book: “The FOAK Framework”, on how to scale up climate hardware. This story, and frameworks like the one above, are part of it (they come in an expanded version in the book). If you’d like early access to draft chapters, drop me a note!