Ads Stopped Running Alert

Tell you when an account has gone quiet, before a declined card costs you a day of traffic.

An account can stop serving without anything looking wrong. A card is declined, a billing threshold is hit, someone pauses the last enabled campaign, a conversion tracking change trips a policy check. Google does not phone you. The first sign is usually a quiet day in your own reporting, noticed the morning after.

This automation checks whether the account recorded any activity at all in the last few hours, and tells you when it did not.

It is a reporting automation. It never changes your account.

How it decides

Each run reads account level performance split by hour, in the account's own timezone, and looks at the last Hours to check completed hours. If the chosen metric was zero in every one of those hours, the account counts as offline.

The current hour is deliberately excluded. Google Ads reporting lags by a few minutes, so an hour that has only just started legitimately shows zero and would raise a false alarm every run.

Because the test is "zero in every hour", a single impression anywhere in the window means the account is serving and no alert is raised. This is a check for an account that has stopped, not a check for an account that is underperforming.

Settings

Hours to check is how long the account has to be silent before it counts as offline. Anything from 1 to 168 hours. The default of 6 is long enough to survive a quiet night on a low volume account without waiting so long that a real outage runs all day.

Metric to check is which signal has to be absent, chosen from two:

• Impressions is the right answer for almost every account. It is the first thing to disappear when serving stops, and it is there even in an hour that earned no clicks.
• Clicks is stricter, and only suitable for an account with enough traffic that a quiet hour is itself unusual. It catches something impressions cannot: ads that still serve but have stopped attracting anyone, which is what a broken tracking template or a mangled headline looks like. On a low volume account it will alert on ordinary quiet nights, so start with impressions and move to clicks only if you know what an hour without one means for this account.

Hours are counted in the Google Ads account's own timezone, the same one the schedule on the General step uses, so an hour here means an hour where the account actually lives.

When it emails you

You get an alert on the run where the account first looks offline, not on every run while it stays that way. Flowboost compares against the last alert it sent in the past seven days: if that run was already offline, the account has not just gone down and there is nothing new to say.

When the account comes back and later goes down again, that is a new alert, because a run that found the account serving ends the incident. The next outage is judged fresh rather than compared against the one before it.

Choosing a schedule

Match the schedule to how much a silent hour costs you. Hourly is the shortest useful cadence and means you hear about an outage within an hour of the window closing. Daily is enough on a small account, but a daily run with a 6 hour window can miss an outage that started and ended between runs.

A window shorter than the gap between runs leaves blind spots. If you run every 6 hours, check at least 6 hours.

Reading a run

Open Status updates in the automation's row. The report shows whether the account was judged offline, the metric and window used, and the value recorded in each hour checked, so you can see whether the account went quiet gradually or all at once.

What to do when it fires

Check in this order: billing (a declined card or a reached spending limit is the most common cause), then campaign status (someone may have paused the last enabled campaign), then policy (an account level suspension stops everything at once). All three are visible in the Google Ads interface within a minute.