How to keep your small business website up, without making it your job.

I have run small-business websites and I have built the monitor that watches them. The people who keep their site up most reliably do four habits, all of them plain. None of them require a contractor.

Renew the TLS certificate on a calendar reminder

Half the "my site is down" messages I receive turn out to be an expired TLS certificate. Most managed hosts renew these for you, but the renewals fail silently when a setting is wrong, and you find out about the failure when a customer finds out about the failure. Put a recurring calendar reminder ninety days out from each renewal cycle. The reminder is five minutes long and asks you to load the site in a private window. If it loads, you are fine for another cycle.

For the deeper walk through what 90 / 30 / 7 / 1-day expiry alert cadence looks like, see the dedicated SSL expiry monitor page. The same article applies whether your cert is Let's Encrypt, DigiCert, or Sectigo — the expiry alert pattern is the same.

Update plugins and themes before the month starts

WordPress, Shopify apps, and any CMS plugin you have installed all ship updates on roughly the same cadence as the vulnerabilities they fix. Pick the first weekend of the month. Update everything. If you cannot remember which plugins you have installed, you have too many; remove the ones you have stopped using. The drift between "installed and forgotten" and "vulnerable in the wild" is shorter than most people think.

The maintenance-automation angle walks through what "automated" actually has to mean for a small shop to be worth paying for, and where dashboards and AI insights fall short of the boring-safeguard test.

Wire up a monitor that pages an actual person

This is where SiteGuardian earns its place. The three habits above reduce the chance of an outage. The fourth habit — a probe that pages somebody when an outage happens anyway — is the part you cannot get from discipline alone. Servers fail for reasons that have nothing to do with you: a hosting provider's storage volume goes read-only, a global CDN config rolls out badly, a nameserver has a bad day.

SiteGuardian's monitoring tier handles the probe, the cadence, and the loud email. It is the habit where it is hardest to imitate the work of a real product with sticky notes.

Keep a written restore path, even if you never use it

Every six months, write down — in a place that is not your head — what you would do if the site stopped answering tomorrow. What host, what login, what backup, what person to call. The document does not have to be long. The exercise of writing it is the useful part. When the outage actually happens, the document reads like a checklist; without it, you are reading your own panicked search history at two in the morning. People who have written the document recover faster. It is the closest thing to free insurance a small shop can have.

If you would rather check whether your own site is restoring cleanly, run the free URL auditnow. It reads the redirect chain, the cache headers, and the TLS chain — the three things that fail silently on a hosted site that has been unchanged for six months.

What is forgettable

Manual uptime checks from your browser. Pet projects in cron. Spreadsheets you fill in. None of those wake you up when the site goes dark at midnight. The boring safeguard is the one that wakes you up — or, better, the one that wakes up the night-shift contractor holding the on-call rotation in their inbox. That is what SiteGuardian does, starting at $19 a month for one site, and it is the only piece of the four that cannot be replaced by willpower.

For shops that already use Oh Dear for the break-the-link crawler side of things, the honest overlap is laid out in SiteGuardian versus Oh Dear. The two tools coexist; most pick the one that fits the daily inbox they are willing to read.