Your account broke on a Saturday. You found out on Tuesday.
That gap is where PPC budgets actually die. Most account disasters are detection failures, so we built the agency around detection: automated checks every 30 minutes, humans every week, incident reports with timestamps instead of adjectives.
Recent detections
Anonymized, from the live alert log
- Conversion tag heartbeat lost after a client site releasepaused within 22 min
- Budget cap removed by a bulk edit on a franchise accountrestored within 35 min
- Top ad group disapproved: destination mismatchresubmitted same morning
- CPC anomaly: brand term +240% vs rolling baselinecompetitor bid war flagged
- Merchant Center feed rejected 61 items on a price mismatchfixed before the morning peak
Across current engagements: 26 accounts watched, 41-minute median from anomaly to alert, €212k of runaway spend intercepted in twelve months.
Five ways accounts die quietly
| The failure | What the radar does about it |
|---|---|
| A conversion tag dies in a site release and smart bidding flies blind. | Tag heartbeat check, every 30 minutes. Silence for two cycles fires an alert. |
| A bulk edit quietly removes a budget cap and spend triples overnight. | Pacing check against plan, every 30 minutes, with an absolute daily ceiling. |
| Your best-performing ad gets disapproved on a Friday evening. | Disapproval scan on every cycle, diffed item by item, feeds included. |
| The landing page 404s after a CMS change while ads keep serving. | Uptime and response checks on every active final URL. |
| CPCs drift up for weeks and nobody can say when it started. | Anomaly detection against a rolling baseline, so drift gets a start date. |
Who's watching
Wout Peeters
Co-founder & Principal · Antwerp
Wout spent 11 years in PPC, the last four as head of search at a Benelux retail group, where an undetected weekend tag failure cost €9,400 and a fair amount of his composure. He built the first version of the monitoring stack that week and has been arguing that detection beats brilliance ever since.
Lena Verhoeven
Co-founder, Client Education & Reporting · Ghent
Lena spent 8 years agency-side watching clients receive reports they never read because nothing in them was written for humans. At Radarcampaign she writes the guides, the plain-language incident reports, and the onboarding that makes clients dangerous enough to check our work.
- Certified on Google Ads
- Certified on Microsoft Advertising
- Monitoring checks every 30 minutes
The €9,400 weekend
- Before the radar
- Tag died Friday 19:10, found Monday. €9,400 spent against a dead conversion signal.
- Same failure, a year later
- Detected 07:32, Saturday. Campaigns paused 08:10.
- Damage the second time
- 38 minutes of spend: about €130.
- What changed
- Nothing clever. A heartbeat check every 30 minutes and a human on call.
What gets checked, and how often
Every 30 minutes
- Spend pacing vs plan
- Conversion-tag heartbeat
- New disapprovals, feeds included
- Landing-page availability
- CPC anomalies vs baseline
Daily
- Budget-cap integrity after bulk edits
- Search-term waste screen
- Merchant Center item status
- Competitor auction movement
Weekly, by a human
- Full account review
- Change log with reasons
- Experiment reads
- Plain-language client note
Standalone monitoring is €390/mo for up to three accounts; full management from €990/mo. Details on the pricing page.
Who's watching your account right now?
If the honest answer is "nobody," the strategy call is free and the first thing we'll do is put that right.
Get a free strategy callOr go straight to monitoring, managed PPC or account rescue.