Monitoring
Zero orders in an hour: how to tell a slow day from a broken storefront
Richard K. · October 10, 2026 · 8 min read
Every store owner has stared at an order dashboard that reads zero and felt that small drop in the stomach. Is it just a slow Tuesday afternoon, or did something break three hours ago and nobody noticed? The honest answer is that you can't tell just by looking. A gut feeling is not a monitoring strategy, and waiting for a customer email asking "did my order go through?" is not a plan either.
Why zero orders is ambiguous without context
An hour without a sale means something different for a boutique doing 15 orders a day than for a store doing 300. It also means something different at 3am on a Tuesday than at 2pm on Black Friday. The raw number, zero, carries no information by itself. What matters is how unusual that zero is given the time, day of week, season, and recent traffic.
This is the core problem with eyeballing a dashboard: humans are bad at judging whether a gap is normal without a reference point. You need a baseline, not a vibe.
Building a baseline instead of guessing
Start by pulling at least 8 to 12 weeks of order history from your store admin. Shopify, WooCommerce, and BigCommerce all let you export orders with timestamps. Break that data down into hour-of-day and day-of-week buckets. You're not looking for a single average order rate, you're looking for a pattern: Monday mornings are quiet, Thursday evenings are busy, the week before a holiday spikes, the week after sags.
From that pattern you can define a reasonable range for each time slot. If your store typically sees 2 to 6 orders between 2pm and 3pm on a weekday, then zero orders in that window is a real deviation worth investigating. Zero orders between 3am and 4am on a Tuesday, for a store with mostly US customers, might just be Tuesday.
This is the same principle behind cart abandonment spikes: reading the signal versus the noise: a number only becomes a signal once you know what normal looks like. Without that baseline, every slow hour looks like a crisis and every real crisis looks like a slow hour.
What actually causes a sudden order volume drop
A genuine order volume drop usually traces back to one of a handful of causes, and they're worth knowing before you start troubleshooting:
Checkout failures. A payment gateway change, a broken discount code, or a shipping rate calculator error can silently empty carts at the final step. See shipping rate calculator errors: how a misconfigured rule empties carts at checkout for a concrete example of how this happens without any obvious error message.
Site or server outage. The storefront loads slowly or not at all, so visitors leave before they ever reach checkout. This is distinct from a checkout bug because the whole site is affected, not just the final step.
Traffic collapse. If your ad campaigns paused, an email send didn't go out, or a referral source dried up, there's simply nobody visiting to place orders. This is a marketing problem, not a technical one, but it looks identical to a broken storefront if you're only watching the order count.
Analytics or webhook blind spots. Sometimes orders are still happening, but your dashboard isn't showing them because a webhook failed or an integration broke. The order that never arrived in your admin: webhook and integration failures explained covers this scenario, where the store is fine but your visibility into it isn't.
Knowing which of these you're dealing with changes what you do next, and the fastest way to tell them apart is to check traffic levels alongside orders. Traffic down and orders down together points to a marketing or external cause. Traffic steady and orders down points to a checkout or technical failure.
Setting up a no orders alert that actually helps
A no orders alert is only useful if it's tuned to your baseline rather than a fixed threshold. A rule like "alert me if there are zero orders in an hour" will either fire constantly for a low-volume store or never fire for a high-volume one until the damage is already significant.
Better alerting compares current order flow against the expected range for that specific hour and day, and flags a deviation, not just a flat zero. For a store with meaningful night-time traffic, that might mean alerting on any 2-hour window with zero orders during business hours, but a 6-hour window overnight. For a highly seasonal store, the baseline needs to shift around key dates rather than staying static year-round.
This is also where order monitoring pairs naturally with uptime and checkout monitoring. If your uptime monitors, status pages, and pingers show the site responding normally but orders have stopped, that narrows the problem to checkout, inventory, or payment processing rather than infrastructure. If uptime checks also show the site down, you have your answer immediately. Cassian™ watches order flow continuously against your store's own history through Order Pulse, and correlates it with uptime checks, so a silent checkout failure and a genuinely slow afternoon don't look the same in your alerts.
A slow day costs you the orders you didn't get. A broken storefront costs you those orders plus every hour it takes you to notice.
What to check in the first ten minutes
When an alert fires, or when you simply notice the order count looks off, work through a short sequence rather than refreshing the dashboard repeatedly. First, check whether the site loads normally from a few devices and networks, not just your own office connection. Second, place a small test order yourself, or walk through checkout to the payment step without submitting. Third, check your traffic source, if visitors have dropped too, the problem is likely upstream of your store (an ad account pause, an email that didn't send, a analytics tracking gap making it look worse than it is).
If traffic looks normal but orders don't, focus on checkout: payment gateway status pages, recent app installs, discount code behavior, and shipping rules are the usual suspects. Document what you find, because the same baseline that flagged the anomaly will help you recognize the pattern faster next time.
Frequently asked questions
- How many orders missing counts as a real anomaly versus normal variation?
- The right threshold depends on your store's own baseline order rate for that specific hour and day of week, not a universal number. A store that normally sees 1 to 3 orders an hour during a quiet Tuesday morning shouldn't panic at zero, but a store that reliably sees 10 or more in that same window should treat a zero-order hour as worth checking. Build your threshold from several weeks of your own historical data broken down by hour and day, then set alerts relative to that range rather than an arbitrary flat count.
- Can a slow day really look identical to a checkout outage?
- Yes, at the level of a single metric (order count), a genuinely slow day and a broken checkout can produce the exact same reading of zero orders. The way to tell them apart is to check a second signal alongside orders, most usefully site traffic and uptime status. If visitors are arriving normally and orders have still stopped, a technical failure is far more likely than ordinary slowness.
- Should I set up a no orders alert even for a small store doing a few sales a day?
- Yes, a no orders alert is arguably more valuable for small stores, because a handful of lost orders represents a larger share of daily revenue than it would for a high-volume store. The key is tuning the alert window to your actual order frequency, a store doing 5 orders a day might set the alert for a 4 to 6 hour gap during business hours rather than a single hour, to avoid false alarms while still catching a genuine outage within the same business day.
Know your normal, trust the deviation
You don't need to watch your order dashboard all day to catch a problem early. You need a clear picture of what normal looks like for your specific store, at your specific hours, and a system that tells you when reality strays from it. That turns a vague, anxious glance at a flat dashboard into a specific, actionable signal: this hour is unusual, go check checkout, or this hour is fine, go enjoy your slow Tuesday.