Hurricane season is a technology problem too

Every business here has a storm plan for the building. Far fewer have one for the parts of the business that are not in the building — which, increasingly, is most of it.

None of what follows is expensive or difficult. It is only difficult in August.

Your website should survive what your building does not

If your site is hosted locally, or on a machine in the office, it goes down when the power does — at precisely the moment customers are checking whether you are open.

Modern hosting is distributed and unaffected by weather in any one city. If you are not certain which kind you have, that is the question to ask this week rather than in August. "Where is my website physically hosted" is a question your provider should answer in one sentence.

The one page that matters during a storm

Customers want to know one thing: are you open, and if not, when.

Make sure somebody other than you can update that message, from a phone, without a computer — and make sure they have actually done it once. A status line at the top of the homepage, editable in a minute, prevents dozens of calls and a great deal of goodwill damage.

The same message should go to your Google listing, which is where more people will look than your website. Practising it takes ten minutes on a quiet afternoon and is worth more than any of the rest of this list.

Know where your data actually lives

Customer records. Orders. Appointments. Invoices. Photographs of completed work.

If any of it exists only on a computer in the office, treat that as a decision you have made rather than an accident that happened. Off-site copies, verified by restoring one, are the whole of the practice.

It is worth being honest about how common the gap is. Plenty of businesses back up faithfully to an external drive that sits ten feet from the machine it is backing up. That protects against a failed hard disk and against nothing else — not fire, not theft, and not four feet of water.

Phones and payments

Know how to forward the business line to a mobile, and confirm you can still take payment if the internet at the premises is out — most card terminals will fall back to a mobile network, but you want to find that out now.

Both are quick to arrange in advance and impossible to arrange during.

Staff need the plan too

Everyone who might need to post the status update, forward the phone or answer a customer should know how, and should have the logins somewhere reachable that is not the office.

A password manager solves this properly. A note in a drawer at the premises does not, for the obvious reason.

The order things fail in

Worth knowing, because it tells you what to prepare rather than what to worry about.

Mobile networks usually outlast mains power, and mains power usually outlasts fixed internet at the premises. So the phone in your pocket tends to be the last thing working, which is the argument for making sure everything you might need in an emergency can be done from it — posting a status, forwarding the line, checking a booking.

Anything hosted outside the city is unaffected throughout. Anything on a machine in the building is gone the moment power is. That single distinction decides most of what survives.

And the recovery order is rarely the one people expect. Getting back online is usually quick; getting back to knowing what you owe, who was booked, and what was half-finished is what takes a week — and that is a records problem, not a power problem. It is decided in advance by where the records live, not afterwards by how quickly the lights come back.

What to write down while it is calm

One page, kept where your team can reach it from a phone. Not a plan — a list of facts nobody will remember under pressure.

Who hosts the website and how to log in. Where the domain is registered. How to forward the business line, with the actual steps. The login for your Google listing. Where the backups are and how to reach them. Who to call at the payment provider. And who decides, if you cannot be reached.

That last one is the item most often missing and the one that costs the most. A team that knows what to do but not whether they are allowed will wait, and waiting is what turns two hours of disruption into two days.

The week after is the part nobody plans

The storm gets the attention. The expensive part is usually the week that follows, and it is a different problem.

Suppliers want to know what you still owe. Customers want to know whether their booking survived. Insurers want dated evidence of what the premises looked like before and after, and "before" is the half nobody has. Staff want to know whether to come in. Every one of those is a records question, and the answers either exist somewhere reachable or they do not.

Two things make that week shorter, and both are done in advance. Photograph the premises, the equipment and the stock now, while everything is intact, and keep the photographs somewhere that is not the building — that is your "before". And know which of your systems reconstruct themselves from the cloud and which live only on a machine that may not survive, because the second list is what you will spend the week rebuilding by hand.

Reopening is rarely what takes the time. Knowing where you were when you closed is.

Do it in May

Put an afternoon in the calendar before the season starts. Work down this list. Write the answers somewhere your team can find them.

You will have removed most of what turns a storm into a business interruption, and the rest of the season becomes a matter of watching the forecast rather than improvising.