Silent overspend is the expensive kind. The campaigns you know are running hot are the ones you are watching. The ones that drain quietly, while you are focused on everything else, are the ones that show up as a number you do not recognize when you finally open the account on Monday morning.
Amazon Ads has no native spend anomaly detection. You can set a daily budget. You can set bid caps. But if a broad match keyword starts pulling traffic in a way you did not predict, or a campaign gets misconfigured after an update, or a listing goes sideways and your ads keep running to nowhere, the platform does not tell you. You find out when you look. And when you look is not always soon enough.
Spend Alerts is the layer that watches when you are not. It is included on every Off Hours plan, requires no configuration, and starts the moment you connect an account. This post explains exactly how it works: what it watches, how it detects problems, what a real alert looks like, and where it fits alongside the four rule types that make up the rest of the platform.
A story about a Thursday
Northlane Goods sells home goods on Amazon. They run Sponsored Products campaigns across a catalog of about 40 products. Most campaigns are phrase and exact match. One auto campaign runs as a discovery layer, and a broad match campaign runs on their top-selling storage line to capture longer-tail queries.
On a Thursday in March, one keyword in that broad match campaign started pulling traffic from a search term that was technically relevant but very high-volume. The term had a low conversion rate for Northlane's product. The campaign had a daily budget with headroom. Clicks came in all day Thursday, all day Friday, through the weekend.
Nobody noticed until the weekly report landed on Monday. Four days. Spend on that one campaign ran at nearly four times its 14-day average. By the time the report was reviewed and the keyword was addressed, the damage was done. That is not a recoverable situation. It is an avoidable one.
This is the problem Spend Alerts solves. Rules cover what you predicted. The expensive surprises are the ones you did not. A broad match gone viral, a campaign running against a suspended listing, a misconfiguration after an account update: none of these are scenarios most sellers think to write a rule for in advance. They are exactly the scenarios that cost the most.
Three things Spend Alerts watches for
Most ad monitoring tools watch for one thing: spend too high. Spend Alerts watches for three, because two of them are just as costly and much easier to miss.
Spikes
A spike fires when a campaign's daily spend exceeds its 14-day rolling average by 20 percent or more. The threshold is calibrated to your account's own baseline, not a generic industry number. A campaign that normally runs $300 per day trips an alert at around $360. One that normally runs $3,000 trips it at $3,600. The math is relative to your normal, so accounts at every spend level get protection that fits.
Drops
Drops are the underappreciated half of the problem. When a campaign goes quiet, because a listing is suppressed, a campaign was accidentally paused, an inventory issue pulled a product from consideration, spend falls and revenue follows. That is a real business impact, and it is one that looks like silence rather than noise. Silence is easy to miss. Spend Alerts flags it.
A drop alert fires when spend falls meaningfully below baseline. The trigger is symmetric to the spike threshold: a sustained decline that breaks the 20 percent band in the downward direction raises a flag, because something changed and you should know about it.
Drift
Drift is the one nobody talks about, but it is the one that does the most invisible damage over time. No single day trips a threshold. Spend is up 7 percent this week compared to last week. It was up 9 percent the week before. It was up 6 percent the week before that. Each week looks approximately fine in isolation. Over six weeks, you are running at nearly 60 percent above where you started, and your unit economics have shifted underneath you without any single moment you could point to.
Spend Alerts runs a weekly pass specifically to catch this pattern. The daily check catches sharp changes. The weekly pass catches the compounding slow ones. Both matter.
Zero setup. Watching from day one.
A safety net you have to configure before it can catch you is not really a safety net. If it requires setup, it requires you to remember to do the setup. And the situations that require a safety net are precisely the ones you did not plan for.
Spend Alerts has no configuration step. There are no thresholds to set, no rule templates to fill out, no alert conditions to define. Connect an Amazon Ads account to Off Hours and monitoring starts. The system builds a 14-day rolling baseline from your account's own historical data, so by the time two weeks have passed, every campaign has a personalized normal to measure against.
This is the meaningful difference from Off Hours performance rules. Performance rules are powerful precisely because they are configurable: you define the trigger (ACoS above 40 percent for three consecutive days), you define the action (reduce budget by 15 percent, send an alert), and the rule executes exactly as written. But they require you to anticipate the scenario first. Spend Alerts requires nothing. It watches for things you did not see coming, which is the category of problem it is actually solving.
What an alert tells you
An alert that says "your account spent more than usual" is not useful. By the time it reaches you, you already know something is wrong. What you need to know is where, specifically, the problem is.
Spend Alerts names the culprit campaigns. Each alert includes which campaigns drove the change, and by how much in both percentage and dollar terms. If three campaigns are within normal range and one is running 80 percent hot, the alert names that campaign and shows the variance. You open the alert, you know exactly where to look, and you can act within minutes rather than spending time diagnosing first.
Honest about Amazon's data lag
Amazon's reporting data does not land instantly. Depending on the account, the marketplace, and the time of day, there can be a 3 to 12 hour lag between when clicks and spend occur and when they appear in the reporting API. Any monitoring system that ignores this will generate false alarms, because it is comparing partial data against a full baseline.
Spend Alerts checks on settled data. When a day's numbers have not fully landed, the alert system waits. Rather than fire a false spike alert because yesterday's 6pm to midnight spend has not propagated yet, it surfaces a "data still loading" status and holds the evaluation until the numbers are complete. This matters because a monitoring system that cries wolf becomes the kind of thing sellers start ignoring, which defeats the purpose entirely.
False alarms are not a minor inconvenience. They train you to dismiss alerts. A system that fires accurately and only on settled data is one you can actually trust, which is the only kind worth having.
How you receive alerts
Alerts are delivered by email. If you have connected Slack to your Off Hours account, they also arrive there. The alert lands in your inbox without any action required on your part. No dashboard to check, no report to run. The information comes to you.
For agencies managing multiple client accounts, this matters significantly. You cannot monitor ten accounts by refreshing ten dashboards. The alerts normalize across all connected accounts, so an anomaly on any one of them surfaces to you on the same channel, through the same format, with the same campaign-level detail. Spend monitoring at scale without a proportional increase in the time it takes to do it.
Spend Alerts vs performance rules: the distinction
These two things look similar from a distance and serve genuinely different purposes.
Spend Alerts watches and tells you. It requires no configuration, watches every account automatically, and flags deviations from baseline whether you expected them or not. It is reactive in the good sense: it responds to what is actually happening in the account, not what you predicted would happen.
Performance rules act. You define a trigger condition, such as ACoS above a threshold or spend below a floor, and you define an action, such as reduce budget by 20 percent or pause the campaign. When the condition is met, the rule executes. Performance rules are how you automate responses to predictable scenarios. They require configuration because you are encoding your own judgment about what constitutes a problem and what the right response is.
The two work together. Spend Alerts catches the surprises you did not plan for. Performance rules handle the situations you did. Neither replaces the other. The common patterns of Amazon ad wasted spend benefit from both layers: rules to address predictable problems, alerts to catch unpredictable ones.
The floor under all four rule types
Off Hours has four rule types: dayparting, budget rules, event rules, and performance rules. Each one automates a specific layer of account management. Dayparting controls when campaigns run. Budget rules adjust spend levels on a schedule. Event rules handle tent-pole moments. Performance rules react to metric shifts. Together they cover the scenarios you thought about when you set them up.
Spend Alerts is what sits under all of them. Rules are only as good as the predictions that created them. A dayparting rule that pauses campaigns from midnight to 6am does nothing if a misconfigured campaign is exempted from the rule and runs all night. A budget rule that boosts spend for Prime Day does nothing to stop a broad match keyword from running wild three weeks before the event. Spend Alerts watches regardless of what rules are active, because the problems it catches are precisely the ones the rules did not cover.
Think of it as the difference between a planned budget and an actual statement. The rules represent the plan. Spend Alerts is the check on reality. Keeping ACoS under control long-term requires both: the discipline of well-built rules and the visibility to know when something outside those rules is happening.
Included free on every plan
Spend Alerts is not an add-on. It is not a premium feature or an upgrade tier. Every Off Hours account gets it, from the first day of the free trial through every month of paid use. Off Hours is $149 per month flat per account, 14-day free trial, no credit card required. Spend Alerts is on from day one.
The logic behind including it at no extra cost is straightforward. The four rule types automate what you plan for. The thing that protects the plan from unplanned surprises should not cost extra. It should just be there. See how Spend Alerts works in detail or start a trial and connect your first account to see it running on your own data.
Frequently asked questions
What are Amazon Ads spend alerts?
Amazon Ads spend alerts are automated notifications that fire when an ad account's spending behavior deviates meaningfully from its recent baseline. A good spend alert catches three distinct problems: sudden spikes, drops where a campaign goes quiet due to a pause or listing issue, and slow drift where spend creeps up week over week in a way that never trips a daily threshold but compounds into a significant problem over time. Amazon does not offer native anomaly detection, so this monitoring layer has to come from a third-party tool or manual review.
Do spend alerts require any setup or configuration?
Off Hours Spend Alerts require zero configuration. Connect an Amazon Ads account and monitoring starts immediately. The system builds a 14-day rolling baseline from your account's own data, so the thresholds are calibrated to your normal, not a generic industry average. There are no rules to write, no thresholds to set, and no templates to configure.
What counts as a spend spike in Off Hours?
Off Hours flags a spend spike when a campaign's daily spend exceeds its 14-day rolling average by 20 percent or more. The baseline is account-specific, so a campaign that normally spends $500 per day would trigger an alert at around $600. The alert names the specific campaigns driving the change and shows the dollar variance, not just a percentage, so you can evaluate severity at a glance.
How are spend alerts different from performance rules?
Spend Alerts watches and tells you. Performance rules act. Spend Alerts require no configuration and fire as soon as something deviates from baseline, whether you expected it or not. Performance rules require you to define a trigger condition and an action. The two work together: Spend Alerts catches surprises you did not plan for, while performance rules automate responses to the scenarios you did predict.
Off Hours includes Spend Alerts on every plan. Connect your Amazon Ads account and the watching starts from day one. No configuration required. Start a 14-day free trial.