日本語で読む
Let’s be honest about what’s actually happening. You need to be able to fire people.
Your company is winning globally right now, but you and I know the internal reality is different. You are running a Ferrari engine inside a tractor chassis. You have a workforce that waits for orders instead of taking initiative. You have a structure where credit flows up, but blame flows down.
You are trying to run a digital company using the JTC system designed for a 1940s factory. It helped rebuild Japan after the war, but it doesn’t work in global, digital markets.
You may have had consultants in bad suits pitch you “digital transformation” or “job descriptions.” They set up committees, wrote new mission statements, and absolutely nothing changed. That’s because they are trying to install new software on broken hardware. It is theater.
Then you have the brute force option, like Rakuten. You mandate English, import a thousand foreigners, and try to overwrite your Japanese identity with Silicon Valley logic. It may work financially, but it destroys the soul of the company. You end up with a culture of mercenaries who don’t care about your history.
You aren’t going to do either of those.
The Dejima Strategy
You cannot simply “change the culture” of the main company. The law is like a wall. You cannot fire the legacy staff, and you cannot force them to act like entrepreneurs.
So stop banging your head against the wall. You are going to walk around it.
You are going to build a Dejima.
Back in 1636 the Shogunate needed Western technology but didn’t want Western chaos. So they didn’t open the borders. They built a man-made island in Nagasaki. They contained the infection but extracted the value.
That is your blueprint. You are going to keep your current company—the Mainland—exactly as it is. Let the generalists run it. Let them have their safety and their seniority. They are the Maintainers. Their job is to keep the lights on.
Then, you build a completely separate structure for the specialists: the Builders.
Here is exactly how you execute this.
1. The Physical Quarantine
The Dejima cannot be on another floor of your HQ. If the Builders share a cafeteria with the Maintainers, the culture will die. The gravity of the Mainland is too strong.
You need to rent an office in a different district, far from your current HQ. It needs its own badge access, its own IT stack, and its own coffee machine. It needs to feel like a foreign country.
2. The Firewall
This is where most companies fail. You cannot allow internal transfers.
You cannot send a 50-year-old Section Chief to the Dejima just because he’s bored. The Dejima is not a rehabilitation center.
If a Maintainer wants to join the Dejima, they must submit a formal resignation from the parent company. They lose their tenure. They lose their retirement track. They re-apply as a fresh hire.
This filters out the tourists. Only the people who are truly hungry for risk will cross that bridge.
3. Sovereign Command
The Dejima requires a specific chain of command to survive the gravity of the Mainland. It does not report to the CEO or HR. It reports directly to you.
You must hire a Wartime CEO from the startup or VC world to run the daily operations. This leader needs to understand burn rates and velocity better than they understand internal approvals. Find someone who has failed at least once.
Your role remains deeply hands-on, but the nature of your control changes. You define what gets built by setting the strategic targets, the capital allocation, and the deadlines. You do not manage how it gets built. You provide the mission and the air cover; they provide the execution.
You manage this relationship through a strict interface. The Mainland provides capital and IP; the Dejima provides finished products. There are no alignment meetings or consensus committees. Accountability is binary. You judge the Dejima solely on the metrics. If they hit the numbers, you protect them. If they miss, you cut the funding.
4. The Venture Studio Model
This unit operates as a factory designed to manufacture startups. In a standard department, half of a team lead’s time is lost to internal bureaucracy. Here, the Dejima leadership handles the Platform (e.g. legal, finance, and HR) using streamlined, external standards. This allows the teams to focus entirely on the product.
You define the strategic targets, such as a new product for the US market, and they commission specific teams to solve those problems. These teams function as autonomous units with their own P&L. They operate with total independence to hit the target you set.
You manage risk through milestone-based funding rather than annual budgets. Projects start with a small Proof of Concept budget to validate the hypothesis. If the data looks good, they unlock the full production budget. If it does not, you cut the fuel line immediately. This allows you to fail cheaply on the bad ideas so you can double down on the winners.
5. The Incentive Structure
To compete with global tech firms and venture-backed startups, you must offer ownership. A high salary is not enough to attract founders, and a subsidiary cannot offer stock options in the main company. The solution is a Shadow Equity model hard-coded into specific projects.
You allocate a percentage of the project’s value (let’s say 20%) to the founding team. This is a contractual claim on the upside. If the project hits its metrics, the team does not just earn a bonus; they earn a Liquidity Event. Once the product reaches a defined scale, the Mainland buys out the team’s shadow equity for a lump sum based on a pre-defined valuation formula (e.g., revenue multiple) locked in the initial contract. The Mainland CFO gets no vote on the price.
This structure solves the retention problem without creating zombie employees. To realize the value of their equity, the team must ship the product and hit the growth targets. If they leave before the launch, they forfeit their stake. But once the value is proven and the buyout occurs, they are free to take their winnings and leave, or double down on a new project. This aligns their greed perfectly with your growth.
6. The Kill Switch
You must pay Global Market Rates to compensate for the lack of safety. To attract talent that would otherwise go to Google or a top startup, the base compensation needs to be significantly higher than the Mainland standard. However, this premium comes with zero tenure.
All staff operate on short-term, renewable contracts tied directly to project milestones. For a new hire, this might be a six-month probation; for a project lead, a one-year term. This avoids the legal trap of indefinite employment and ensures you are never stuck with a salary that isn’t generating value.
Employment is strictly contingent on results. If a project is cancelled or metrics are missed, the contract is simply not renewed. There are no performance improvement plans and no reassignments to other departments. This reintroduces the one thing missing from your current culture: Healthy Fear.
Conclusion
I won’t lie to you. This strategy comes with a heavy social price. It explicitly creates two castes: the high-risk, high-reward Builders and the low-risk, stagnant Maintainers.
In the West, this kind of wealth gap leads to social unrest. But Japan is the only nation that can navigate this, because you have a demographic cheat code.
You are losing 600,000 workers a year to aging. You don’t need to fire the Maintainers to make room for the future. Nature is downsizing your workforce for you. The “Maintainer” class is not a permanent underclass; it is a sunset generation.
This allows you to do something no American CEO can do: You can keep the safety net for the old guard while building a ruthless engine for the new guard.
This reveals the critical difference between the Dejima of 1636 and the Dejima of 2026.
The original Dejima was built to be a cage. It was designed to keep the future small and contained forever.
Your Dejima is a seed. It is designed to grow.
As your Maintainers retire, the Mainland will shrink. AI will eat the jobs of most paper-pushers. But as your Builders succeed, the Dejima will expand. Slowly, year by year, the logic of the island will overtake the logic of the continent.
The Dejima is not just an innovation lab; it is the lifeboat being built while the Titanic slowly sinks on purpose.