What the 2011 tsunami taught me about business continuity

Business continuity for systems means the business can keep taking orders, communicating and getting paid when a building, a device, a vendor or a key person is suddenly unavailable. In Japan, where earthquakes, tsunamis and typhoons are part of normal planning, that depends less on any particular tool than on knowing what the business really runs on, holding its data and accounts where they can be reached from anywhere, and not relying on one person’s memory. The time to build that is before anything goes wrong.

In March 2011, I was in Miyako, Iwate when the Tōhoku earthquake and tsunami changed the shape of the coast.

That is not a clean business anecdote or a founder myth. It was physical, local, messy, and irreversible. The ground moved. The sea came in. Assumptions that had felt stable in the morning were useless by afternoon.

When I talk about digital infrastructure in Japan now, people sometimes hear a technical service: websites, data, workflow automation, AI tools, CRM. Those are part of the work. Underneath them is the question Miyako left me with: what happens when the system you trusted no longer works?

Video: He Turned Chaos Into a Career with Matt Ketchum from Career Thoughts Over Coffee.

What does a disaster reveal about systems?

Miyako is a coastal city in Iwate Prefecture. During the 2011 Tōhoku earthquake and tsunami, parts of the Sanriku coast were hit by catastrophic waves, and Miyako was among the places devastated.

I do not treat that as abstract history. I was there.

The lesson I took from it has repeated itself in very different settings since: systems fail first in the gap between the map and the ground. A seawall is a model of expected risk. A warning system is a chain of interpretation, trust, timing and movement. A town plan is a bet about how people, roads, buildings and habits will behave under stress. When the stress arrives, the abstraction is judged by the details.

Business systems work the same way. A company says it has a CRM, but the real customer knowledge lives in one salesperson’s LINE history. A town says its vacant houses are tracked, but the useful information is in paper files, old maps and neighbors’ memories. A founder says the website explains the business, but the real value is in years of unrecorded experience. From outside, the system exists. On the ground, it is fragile.

Why do weak business systems look fine until they fail?

In stable conditions, weak systems can look fine for years. A paper workflow looks fine until the person who understands it retires. A vendor relationship looks fine until the price changes or the vendor closes. A WordPress site looks fine until nobody remembers how it was built. A spreadsheet looks fine until it becomes the only record behind a major decision.

Companies rarely collapse in one wave. They lose knowledge one retirement at a time, let a website decay one missed update at a time, and let data become untrustworthy one copied spreadsheet at a time. Then the owner wants to sell, expand, automate, hire or hand over the business, or a disaster closes the office for a week, and the system is not ready. The failure was not sudden. It had just not been visible.

Many of these systems are locally sensible and culturally reinforced, which is why generic advice to “move to the cloud” rarely lands. Move what, for whom, in which language? Which paper process is a legal requirement and which is habit? Which employee is quietly acting as the system that connects everything else?

How should a small business in Japan plan for continuity?

Japan has a well-developed idea of business continuity planning (事業継続計画, BCP), and the Small and Medium Enterprise Agency runs a certification for SME continuity plans (事業継続力強化計画). The formal plan matters less than a few practical questions about your systems:

Can people work if the office is closed? Email, files, accounting and customer records should be reachable from a laptop or phone anywhere, through services such as Google Workspace or Microsoft 365, freee or Money Forward, rather than only from a server or PC in the office.

Are the important records backed up somewhere else? Cloud services protect against a lost building, but not against deleted data or a locked account. Keep a separate copy of critical records and check that it can be restored.

Can you reach staff and customers? Keep an up-to-date contact list outside your main systems, and decide how you will check staff safety (安否確認) and tell customers what is happening, whether by email, LINE official account or your website.

Does access depend on one person or one device? Admin passwords, two-factor codes and recovery emails often sit with one employee. If that person is unreachable, can anyone else get in? I cover this in treating domains, accounts and access as business assets.

What still runs on paper or fax? Orders arriving by fax at an office machine stop when the office does. Receiving faxes as PDF, as described in the guide to moving off fax, removes one single point of failure.

Is the knowledge written down? For each critical process, write down how it works, who does it and what to do if they are unavailable. A short checklist that someone else can follow beats a perfect manual nobody reads.

Why does local knowledge matter as much as the plan?

One of the hardest lessons of a disaster is that official information and local knowledge are not substitutes for each other. Official systems coordinate response, set standards and publish warnings. Local knowledge explains the shortcuts, the weak points, the trust networks and the places where the map is wrong.

Japanese business works the same way. A foreign company can read market reports and still misunderstand how decisions are made. A software vendor can translate its interface and still miss the workflow. A recommendation can fail because it ignored hanko culture, accounting habits, bilingual staffing or the one person who has “always handled that.” Good continuity planning asks what the paper is doing socially as well as administratively, which processes exist because of law, and which analog practice should be kept rather than removed.

Where this shows up in my work

The industries I keep returning to are the ones where the official records are thin and the best information sits in fragments: rural property, outdoor and regional tourism, independent music, and small businesses running on analog workflows. I co-founded Akiya & Inaka and helped build it by collecting and structuring scattered property records, and I talked through why akiya are a data and trust problem before they are a real estate opportunity in the Unpacking Japan conversation.

For businesses, the work is the same instinct applied to their systems: map the real workflow rather than the official one, find where knowledge lives, separate constraints from habits, turn tacit knowledge into documents and checklists, choose tools only after the operating reality is clear, and build ownership in so nothing depends on one person remembering everything. AI helps when that foundation exists. Without it, AI mostly speeds up confusion.

Preparedness is respect for reality

Preparedness is not paranoia. For a Japanese business it might mean clear documentation, bilingual process notes, a CRM that reflects real customer relationships, a vendor map, a tested backup and a workflow that survives when one person is away. None of this is glamorous. Real infrastructure usually becomes visible only when it is missing.

If you suspect your business is held together by undocumented workarounds, paper, inherited systems or knowledge that only exists in people’s heads, a Diagnostics review maps what you run on and where it would break. For overseas-owned teams, see digital infrastructure for foreign-owned SMEs in Japan.

---

Further reading: the hidden cost of good enough systems · why good Japanese businesses are hard to find, trust and buy from · how to digitize paper workflows