The Morning the Dashboard Went Red: A 24-Hour Plan for a One-Person Business
It's 6:47 on a Tuesday morning. You open your laptop with coffee in hand and your site returns a 502. Not a slow load. A wall. Your host's status page says "investigating." Your inbox has two emails from customers asking if you're still in business, and your payment processor is throwing errors that make no sense at 7 a.m.
For a company with an IT department, this is an incident. For a one-person business, it's an existential afternoon.
The good news: most of what turns a two-hour outage into a two-week revenue hole is decided before the outage happens. What follows is the plan to write this week, and the order to run it in when the lights go out.
Before anything else: the three things you actually need
Not a business continuity framework. Three things:
- A backup you have personally restored from. Not a backup that exists — one you've tested. Open the archive, confirm the files are there, confirm the database dump imports. Do it once a quarter. A backup nobody has opened is a hope, not a backup.
- A way to reach your customers that doesn't live on your website. An email list on a separate provider, plus phone numbers for your top twenty clients. If both your site and your email run through the same platform, you don't have redundancy, you have one vendor with two logos.
- A static page you can publish somewhere else in under ten minutes. One HTML file, hosted free, with your contact info and one sentence explaining what's happening. Write it now. It takes twenty minutes and it will feel ridiculous until the day it doesn't.
That's the whole prep list. Everything below is execution.
Hours 0–1: confirm the scope and stop guessing
Before you touch anything, figure out what's actually broken. Three questions:
- Is it the host, the DNS, or my own code? Check a down-detector, then check your DNS records from a second network. If your registrar is the problem, no amount of host support will help.
- Is my email affected? Send yourself a test from an outside address.
- Can customers pay? Run a live test transaction on your own card for a small amount.
Write the answers down with timestamps. You will forget by hour four, and your host's support chat will ask.
Then open one ticket per failing service — not one per symptom — and post to the provider's status page thread so other people's reports reinforce yours.
Hours 1–2: put up the holding page
Publish the static page you wrote in advance. Point a subdomain at it if you can, or put it behind your existing domain at a different path. It needs exactly four elements:
- A headline: "We're aware of an issue and we're on it."
- What still works: "Email and phone are unaffected."
- What to do right now: one link or one phone number.
- A timestamp, updated every few hours.
That timestamp is the whole trick. An undated error page reads as abandoned. A page that says "updated 9:15 a.m. ET" reads as a business with a pulse.
Hours 2–4: tell people before they ask
Reactive communication feels safer. It isn't. The customers who find out by hitting an error are the ones who start wondering whether to take their business elsewhere.
Three messages, in this order:
To your list. Subject line: "Quick note — our site is down, here's what to do." Two paragraphs. What happened, what still works, when you'll update them. No apology spiral, no technical detail they can't use.
To your top twenty clients, individually. Four sentences. Same content, human voice. If a payment is due this week, tell them to hold off and you'll resend the invoice.
To anyone with an open order. This is where outages cost real money. If a delivery, appointment, or file transfer depends on your site, name it explicitly and give them the workaround.
Draft all three templates on a normal day and save them in a single document. Under stress, you'll write a rambling message, second-guess it for an hour, and send it too late.
Hours 4–8: payments, specifically
A frozen payment processor is worse than a dead website, because it touches money people have already committed.
If your processor is down or has frozen your account: don't wait on the support queue alone. Have a second processor already approved and connected to a backup checkout link. Stripe and PayPal are not interchangeable for every business, but almost everyone can keep a second rail alive for emergencies.
For invoices already outstanding, switch to a manual path — a payment link from the second provider, or a bank transfer with clear instructions. Send it to affected customers with a one-line explanation.
Then, when it's back up, reconcile manually. Every outage produces a handful of payments that landed twice or not at all. Give yourself a block on day two to untangle it.
Hours 8–24: the recovery checklist
Once the site is back, resist the urge to close the laptop and pretend it didn't happen. In the next twelve hours:
- Test the three customer paths end to end: browse, buy, receive email confirmation.
- Check for orders or form submissions that came in during the outage and never reached you.
- Update the holding page one final time with "resolved" and a link back.
- Send a short follow-up to your list: back up, nothing lost, here's the phone number if anything looks off.
- Export the day's server and error logs to a folder you'll actually find again.
The part nobody writes until it's too late
Within a week, write the postmortem — for an audience of one. What broke, how long it was down, what you tried that didn't work, and the single change that would have shortened it. Then make that change.
And put a recurring 30-minute block on your calendar every quarter labeled "outage drill." Open the backup, publish the holding page, send yourself the customer email. It feels silly. It takes half an hour. It's the difference between a bad morning and a bad month.
