WatchIt's tempting to treat downtime as an occasional nuisance rather than a line item. But once you actually estimate it, the maths is sobering — and it reframes monitoring from a cost into cheap insurance against a much larger bill.
The part everyone counts: lost revenue
The obvious cost is direct: if your store or app makes money online, every minute it's down is revenue that doesn't arrive. A rough estimate is easy — take your monthly online revenue, divide by the minutes you're open for business, and that's your per-minute cost of an outage. For a webshop doing €500,000 a month, even a couple of hours down in peak hours is real money.
The trap is assuming outages happen at convenient times. They cluster around deploys, traffic spikes and campaigns — exactly when revenue per minute is highest. The expected cost is therefore higher than the average suggests.
The part most teams forget
Lost sales are only the visible layer. Below it sit costs that are harder to invoice but just as real: customer trust erodes when people hit a dead site, support gets flooded with "is it down?" tickets, and staff drop what they're doing to firefight. Repeated outages train customers to keep a backup option open — and some never come back.
There's an SEO dimension too. If search engines crawl your site during an outage and repeatedly get errors, it can affect how they treat it. None of these show up on a single invoice, which is exactly why they get underweighted.
Why detection speed is most of the cost
Here's the key insight: total downtime cost is roughly your per-minute cost multiplied by minutes-to-resolution — and a big share of those minutes is detection, not repair. If you find out from an angry customer 40 minutes in, you've already paid for 40 minutes before anyone could start fixing it.
This is the cheapest lever you have. Cutting detection from "whenever someone notices" to one minute removes most of the avoidable cost, because the clock starts the moment something breaks, not the moment a human happens to look. Confirmation before alerting keeps it from crying wolf, and multi-channel alerts make sure the right person actually gets the message.
Cut your detection time with WatchControl
WatchControl runs website and API checks as fast as every minute, confirms before paging to filter out blips, and alerts by email, webhook and SMS so the right person is reached fast. Response-time history and SLA reports help you see the trend, not just the outage.
It starts free with no card, runs from the EU, and turns the most expensive part of an outage — the time before you knew — into a minute or two.