Why do my Meta Ads campaigns turn off by themselves?

Quick answer: trace the status change before turning anything back on

A Meta Ads campaign usually appears to turn itself off because a person, automated rule, schedule, account control, billing event, restriction, or connected API tool changed its effective delivery state. The visible switch tells you the current state; change history and account-level evidence tell you what caused it.

Do not immediately reactivate every affected object. Record the campaign, ad set, and ad IDs, the first time delivery stopped, the last impression timestamp, the displayed status at each hierarchy level, and any warning message. Then identify the actor and trigger before deciding whether restarting is safe.

• User symptom: an object that was delivering is now off, paused, inactive, or no longer producing impressions.
• Control level: account, campaign, ad set, ad, automated rule, integration, or user access.
• Time grain: the exact status-change event and the delivery window around it.
• Likely mechanism: manual action, automation, schedule, limit, billing, policy, or API control.
• Decision produced: restore delivery, correct a rule, remove a limit, secure access, or leave the object off.

Keep an unexpected shutdown separate from adjacent delivery problems

This diagnosis starts when an object was on and then changed state or stopped because a control acted on it. It is different from an active campaign that never spends, an ad stuck in Preparing or review, a rejected ad, or delivery that continues briefly after a deliberate pause.

The distinction matters because restarting an object cannot fix auction underdelivery, incomplete review, or a policy decision. Conversely, changing bids or creative is unnecessary when an automated rule is repeatedly pausing a healthy campaign.

• Active but not spending: the switch is on, but eligibility or auction constraints prevent delivery.
• Preparing or in review: the ad has not completed a pre-delivery platform stage.
• Rejected or restricted: policy or account enforcement blocks eligibility.
• Spend after pausing: delivery, reporting, pause, and billing clocks do not stop at the same instant.
• Unexpected shutdown: a status, schedule, limit, or automation event stopped an object that had been running.

Freeze one incident with IDs, timestamps, and hierarchy states

Choose one affected campaign and record immutable IDs rather than relying on names that may have changed. Capture the account time zone, the approximate stop time, last delivered impression, current status, effective status, budget, schedule, bid strategy, and warning banners at campaign, ad set, and ad level.

Check whether only one child stopped or the entire hierarchy became inactive. If the campaign is on but one ad set is off, start at that ad set. If many unrelated campaigns stopped together, move outward to account spending controls, billing, permissions, restrictions, or shared automation.

• One campaign ID plus every affected child ID.
• Last impression and spend timestamps in the ad account time zone.
• Configured status and effective delivery status at every level.
• The exact warning, error, or account-quality message.
• A list of other objects that stopped during the same interval.

Use change history to identify who or what flipped the switch

Open change history for a window that begins before the last normal delivery and ends after the shutdown. Filter by status, budget, schedule, rule, and account changes. Record the timestamp, object ID, old value, new value, and actor for each relevant event.

An actor may be a named user, business partner, automated rule, bulk edit, API integration, or system action. Do not assume a colleague remembers every change: scheduled actions and third-party tools can run outside business hours, while bulk edits may affect objects whose names look unrelated.

• Match the status-change timestamp to the delivery stop.
• Check campaign, ad set, and ad history instead of only the parent campaign.
• Review bulk edits and duplicated objects created near the incident.
• Preserve the actor identity before permissions or tools are changed.
• Export or screenshot evidence if the event may need escalation.

Audit automated rules, schedules, and third-party tools

Automated rules are a common cause of campaigns that repeatedly turn off. Inspect active and recently disabled rules for conditions based on spend, CPA, ROAS, frequency, result count, time, or naming patterns. Verify the evaluation window, attribution setting, schedule, time zone, action, and scope. A rule can be logically correct but operate on incomplete same-day data or more objects than intended.

Also review portfolio tools, agency scripts, bid-management platforms, connectors, and custom API jobs with permission to update campaigns. Compare their execution logs with the Meta change event. If reactivation is followed by another pause at a predictable interval, recurring automation is more likely than a random platform fault.

• Rules with pause, stop, budget, or schedule actions.
• Broad filters that unintentionally include new or renamed campaigns.
• Short evaluation windows that react to immature attribution data.
• Account and tool time zones that shift the intended execution hour.
• Connected apps, scripts, and partner users with campaign-management access.

Check end dates, spending limits, billing, and account enforcement

A lifetime-budget campaign or ad set can stop at its configured end date even when the creative and targeting remain valid. Account spending limits, prepaid balances, payment failures, and credit constraints can halt broader delivery. Compare the exact stop time with schedules, remaining budget, billing events, and account-level limits.

Then inspect Account Quality and asset status. A policy restriction, identity problem, security review, or disabled payment method can make an object effectively inactive without resembling a manual pause. Keep these causes separate: extending a schedule will not resolve billing, and replacing a card will not resolve a policy restriction.

• Campaign and ad set start and end dates.
• Lifetime budget consumed versus remaining scheduled time.
• Account spending limit and prepaid or credit balance.
• Failed payment, disabled payment method, or overdue balance.
• Account, Page, business, or user restrictions affecting delivery.

Treat an unknown actor as a security incident

If change history names an unfamiliar person, partner, app, or business, do not simply turn the campaigns back on. Review administrators, employees, partners, system users, connected apps, access tokens, and recent login or security alerts. Remove unrecognized access through the appropriate business controls and require stronger authentication for authorized users.

Preserve IDs, timestamps, screenshots, billing records, and the list of changed objects before cleanup. Unexpected pauses may be protective or may accompany unauthorized budget, destination, or creative edits. Verify those fields as well before restoring delivery.

• Confirm every person, partner, system user, and connected application.
• Review destination URLs, budgets, creatives, and payment settings for other changes.
• Rotate credentials or tokens through approved security procedures.
• Document the incident before removing evidence-bearing access.
• Escalate unknown administrative or financial activity immediately.

Reactivate with a controlled test, not an account-wide toggle

Once the cause is known, make the smallest responsible correction. Disable or narrow one faulty rule, extend one intended schedule, resolve one verified billing issue, or restore one authorized permission. Reactivate a single controlled object first and monitor whether its status remains stable through the next automation cycle.

Confirm that review completes, eligibility returns, impressions resume, spend stays within guardrails, and the intended destination and optimization event remain unchanged. If the object turns off again, compare the new change-history event with the first incident instead of repeatedly toggling it on.

• Fix the proven control before reactivation.
• Test one campaign or ad set with a known budget ceiling.
• Wait through the next relevant rule or schedule interval.
• Verify status, impressions, spend, destination, and event selection.
• Document the permanent owner for each automation and account limit.

How an AdSpecIt-style audit helps diagnose campaigns that turn off

An AdSpecIt-style audit can inventory campaign, ad set, and ad states alongside schedules, budgets, limits, delivery history, change evidence, optimization settings, and performance. That account-wide view helps separate a single object-level pause from a shared account control or automation pattern affecting many campaigns.

The useful output is a cause-and-action map: identify the first affected layer, match the stop time to a change or control, assess whether the actor was authorized, and recommend the smallest safe recovery. This prevents teams from treating every shutdown as an algorithm mystery and reduces the risk of restarting campaigns that were stopped for a valid billing, policy, security, or profitability reason.