The short answer: hours are forgiven, weeks are not, and the domain is sacred
Websites go down. Hosts fail, updates break things, certificates lapse — and Google knows it, which is why a short outage, hours to roughly a day, generally leaves no lasting mark: the crawlers hit an error, note it, and come back later to find the site as it was. The trouble is cumulative. As downtime stretches into days, the errors stop looking like an incident and start looking like a condition: crawling slows, and pages that stay unreachable can be removed from the index until they can be fetched again. And none of that is the worst case. The worst case is the domain itself expiring — the name your rankings, your links, and your business email all live on — because that failure does not fix itself when a server reboots.
If you arrived here because your rankings already moved, downtime is one suspect among several; our checklist in why did my Google rankings drop walks the others. This post covers the outage itself: what happens on what timeline, what recovery actually looks like, and the two or three cheap habits that make the whole scenario boring.
The first day or so: Google retries before it judges
When a crawler requests a page and the server answers with an error or does not answer at all, the system does not immediately conclude the page is gone. It retries — on a schedule that stretches out as failures repeat — because the web is full of temporary failures and treating each one as permanent would wreck the index. During a brief outage your pages generally stay indexed, your rankings hold, and the main cost is the real one: customers who found you during those hours could not reach the site. For a local service company, a down site on a busy Saturday loses that Saturday's calls, which is reason enough to care, but it is not a rankings emergency.
Google's own representatives have said versions of this for years: short downtime is normal and recoverable, and the systems are built to tolerate it. The caveat is that tolerance is a slope, not a switch. Every additional day of errors moves you further from 'incident' and closer to 'this site appears to be gone,' and the recovery math changes with it.
When downtime stretches: crawling slows and pages start dropping
Sustained errors change how the crawlers treat the site. A server that keeps failing is wasting crawl effort, so visits space out — which means that even after you fix the problem, the fix is discovered later than you would like. Pages that remain unreachable through repeated checks can be dropped from the index, starting with the ones crawled least often. Rankings do not fade evenly; the site tends to hollow out, with strong pages hanging on longest and the long tail of service and city pages going first — often the exact pages that bring in the specific, ready-to-buy searches.
There is also a quieter failure mode in the same family: the site that is technically up but broken — pages returning errors, a certificate warning scaring off browsers, a migration that left half the URLs pointing nowhere. From a crawler's perspective that is downtime wearing a disguise, and it is more dangerous than a clean outage precisely because nobody's alarm goes off. The owner checks the homepage, sees it load, and never learns the service pages have been erroring for a month.
If part of your revenue comes from commercial accounts, our guide to website development for commercial service companies explains what property managers and procurement teams look for before they request a proposal.
The catastrophic case: your domain expires
Everything else in this post is recoverable on a timeline of days to weeks. An expired domain is different in kind. When the registration lapses, the site stops resolving entirely, your business email on that domain breaks with it, and the name enters a countdown at the registrar after which someone else can register it — and domains with history and traffic are exactly the ones speculators watch for. Getting an expired domain back ranges from a renewal fee inside the grace window to an auction price to simply impossible, and every day of it is a day your rankings, your links, and your customers' bookmarks point at nothing or at a stranger's parked page.
Treat the domain like the deed to the building, because functionally it is. It should be registered to the business, on auto-renew, on a payment method someone monitors, with registrar access held by the owner — not by a departed employee, not solely inside a former web designer's account. More service companies have been hurt by losing control of the name than by any algorithm update, and unlike an algorithm update, it is entirely preventable for the cost of a calendar reminder and a working credit card.
Recovery, and the monitoring that makes this boring
When the site comes back, recovery is mostly a matter of being crawlable and then being crawled. Fix the underlying problem completely — a site that flaps up and down resets the clock each time. Then use Google Search Console: check the coverage and page indexing reports for dropped URLs, and request indexing for your most important pages rather than waiting for the crawlers to wander back on the stretched schedule the outage earned you. If the content and URLs are unchanged, pages generally return as they are re-crawled, over days to a few weeks depending on how long they were gone. What you cannot do is hurry the whole index at once; work the money pages first — homepage, core services, top cities — and let the rest follow.
Prevention is cheaper than all of it, and it is short: an uptime alert that texts or emails you when the site stops responding, so you learn about an outage in minutes instead of from a quiet phone; a Search Console check on a schedule, since its reports will show error spikes you would otherwise miss; domain and hosting renewals on auto-pay with a backup card; and registrar, hosting, and site logins written down somewhere the owner controls. None of this requires expertise. It requires an afternoon, once, and it converts the scenario in this post from a rankings event into a bad morning.
Where ProRouteX fits
None of this is exotic, and that is the point — the companies that suffer from downtime are rarely missing technology; they are missing an owner-level checklist. Every ProRouteX build for local service companies across 45 industries — one company per trade, per market — ships with the basics in place: Search Console set up and verified, the sitemap submitted, URLs structured so a problem is visible instead of silent, and the site delivered in one week so there is time to check it properly. The full structure is on the growth system packages page.
Check two things after you read this: that your domain is on auto-renew under an account you control, and that someone would know within the hour if your site went down tonight. If either answer is no, fixing it is the highest-value marketing work you will do this month.
Frequently asked questions
My site was down for a weekend. Should I worry about my rankings?
Probably not much — outages of hours to about a day are generally tolerated and pages stay indexed. Confirm the site is fully back, check Search Console for error spikes over the following days, and worry about the calls you missed more than the rankings.
How do I know if pages dropped out of Google's index?
Use Google Search Console's page indexing report, which shows excluded and errored URLs, and spot-check important pages with a site search for your domain. Dropped money pages should be re-requested for indexing once the site is stable.
Does downtime hurt my Google Business Profile too?
The profile itself is separate and keeps showing, but its website link leads customers to a dead site, and a prolonged outage undermines the whole presence. The profile surviving is not much comfort if every path from it is broken.
How long does recovery take after a long outage?
Pages return as they are re-crawled, which runs from days for important pages to weeks for the long tail after a long outage. Requesting indexing for your core pages in Search Console and keeping the site stably up are the two things that speed it up.
What is the single most important protection?
Domain control: registration in the business's name, auto-renew on, payment method current, and login access held by the owner. A server outage costs days; losing the domain can cost the entire web presence built on it.
