Earliest signal of broken tracking, landing-page regression, or auction-side shift.
At a glance
Real-time alert that fires when daily conversions fall sharply against a same-day-of-week baseline. The earliest signal of broken pixel-tracking, a landing-page regression, or an auction-side shift. Read this card BEFORE making any “ROAS dropped” decisions, often the conversion drop is measurement, not performance.
| The formula | (today.metrics.conversions vs same_dow_avg.metrics.conversions) - 1. Compares today’s conversion count against the average of the same day-of-week over the prior 7 days (so Mondays compare to Mondays). Alert fires if the drop is greater than 25%. |
| GAQL resource + metric | FROM customer with segments.date for daily aggregation. Field: metrics.conversions. Conversion-rate variant: (metrics.conversions / metrics.clicks) * 100 for the same comparison. |
| Account currency (single by design) | Not currency-relevant (conversion count is unitless). The downstream impact is currency-dependent, but this card flags the count drop. |
| Conversion attribution model (configurable) | Whatever you have set in Google Ads. DDA is default for new accounts, Last click for older. Critical: if you switched the model recently, this alert may fire spuriously because the new model credits conversions differently. Always check Tools > Attribution before assuming a real drop. |
| View-through inclusion (excluded by default) | Uses primary conversions. A drop here does NOT flag if all-conversions held but primary tanked, that pattern usually means a primary conversion action was disabled or the pixel for it broke. |
| Bot / IVT filter | Google’s Invalid Click Filter applies. A sudden bot-spike in clicks (without conversions) does NOT inflate this alert because the comparison is on conversions, not the conversion rate. The conversion-rate variant of this alert IS sensitive to bot spikes, see CTR Decline for that pattern. |
| Micros conversion | Not applicable to this card. |
| Real-time vs ingestion lag | This is a real-time alert. It runs hourly against fresh data, but conversions take 1-4 hours to ingest. Daily totals can swing materially in the first few hours. The alert is suppressed until 14:00 local time to allow morning ingestion to settle, otherwise it would fire every morning. |
| MCC aggregation | Per child account. MCC managers see one alert per affected account, the engine does not roll up. |
| Time window | RT (real-time). Comparison baseline is prior 7-day same-day-of-week. |
| Alert trigger | drop > 25% vs prior 7D same-DOW. |
| Roles | owner, marketing |
Calculation
Calculated automatically from your Google Ads data. See the At a glance summary above for what the metric tracks and the worked example below for a typical reading.Worked example
A UK fashion brand. Tuesday 14 Apr 26, 18:00 local time. The alert fires.| Day | Day of week | Conversions | Vs same-DOW avg |
|---|---|---|---|
| Tue 7 Apr 26 | Tue | 38 | baseline |
| Tue 14 Apr 26 (today) | Tue | 14 | -63% vs prior Tue |
| Tue (prior 7-day rolling avg) | Tue | 36 | (the alert baseline) |
| Metric | Today | Same-DOW avg | Delta |
|---|---|---|---|
| Spend | £905 | £880 | +3% |
| Clicks | 1,124 | 1,080 | +4% |
| Conversions | 14 | 36 | -61% |
| Conv. rate | 1.25% | 3.33% | -62% |
| ROAS (estimated) | 1.5x | 4.2x | -64% |
- Spend and clicks are normal, conversions cratered. This is the textbook signature of a tracking break. If targeting/auction had shifted, clicks or spend would have moved too.
- The ROAS card will look catastrophic (1.5x vs 4.2x baseline). The instinct is to pause campaigns. Do NOT pause campaigns yet. Run the diagnostic checklist:
- Step 1: Pixel check. Open the brand’s checkout success page in incognito, complete a £1 test order, watch Tag Assistant or the GTM debugger for the conversion fire. Most likely cause: a developer pushed a code change to the checkout page yesterday and broke the conversion tag.
- Step 2: GA4 cross-check. Look at GA4 Realtime > Conversions for the last 4 hours. If GA4 also shows the drop, the conversion broke. If GA4 shows normal conversions, only the Google Ads pixel broke.
- Step 3: Commerce platform check. Look at Shopify or BigCommerce orders for the same window. If orders held normal volume, only Google’s measurement broke; revenue is fine. If orders also dropped, traffic is converting in real life too, this becomes a real performance issue.
- Step 4: Attribution model check. Tools > Attribution. If someone switched DDA -> Last click overnight, conversions get re-attributed away from Google Ads to other channels. Real conversions, just credited elsewhere.
- Don’t pause until you’ve cleared steps 1-3. A spurious pause based on broken measurement costs more than a few hours of accurate measurement.
- If steps 1-3 all show normal conversion behaviour, the issue is between Google’s reporting and your account. Wait for the next ingestion cycle (typically 4 hours). Some Google Ads conversion-pipeline lags do produce one-day-only “ghost drops” that resolve overnight.
- If steps 1-3 confirm a real drop, then check landing-page status (down? slow? was a Magento/Shopify deploy at the same time?), then check competitor activity (CPC spike?), then promo timing (did a 30%-off promo end yesterday?). In that order.
- Pixel check first, always.
- Don’t restructure campaigns on a single-day alert.
- 7-day rolling average is the rebound test, if today is a one-day blip, wait.
- If the alert fires three days in a row, escalate (the gap is structural).
- If it fires on a Monday after a sale ended Sunday, expected, not actionable.
Sibling cards merchants should reference together
| Card | Why pair it with Conversion Drop |
|---|---|
| Google Ads Clicks vs Conversions | The diagnostic chart. If clicks held while conversions dropped, the cause is measurement (tag), not auction. Open this card immediately when the alert fires. |
| Google Ads ROAS | The headline that will look catastrophic when this fires. Don’t act on the ROAS drop until you’ve diagnosed this alert. |
| Google Ads Spend Anomaly | If spend held stable but conversions dropped, tag breakage. If spend also dropped, auction-side shift. Two different fixes. |
| Google Ads Alert: Conversion Tracking Broken | The deeper diagnostic, watches click-to-conversion ratio over time and surfaces a likely tag-break. |
| Google Ads CTR Decline | If CTR dropped at the same time, the cause is more likely auction (creative fatigue, competitor entry) than tracking. |
| Google Ads Conversion Rate Trend | The longer-window context. A one-day drop is noise; a 7-day decline is structural. |
| GA4 Revenue by Channel | Critical cross-check. If GA4 also shows the drop, the conversion really broke. If GA4 is fine, only Google Ads pixel broke. |
| Shopify Total Revenue | The truth side. If commerce-platform orders held while Google Ads conversions cratered, no real performance issue, just a measurement gap to fix. |
Reconciling against the vendor’s own dashboard
Where to look in Google Ads UI: Google Ads > Overview shows daily conversions on the main chart. Hover the latest day to see today’s count vs trailing average. Google Ads > Reports > Predefined > Time > Day gives the day-by-day tabular view this card runs over. Filter to last 14 days, compare today vs same-DOW prior week. Google Ads > Tools > Conversions > Status shows pixel-fire status per conversion action; this is where a tag-break shows up FIRST as “No recent conversions” warnings. Other Google Ads views that look like the same number but aren’t:- All conversions trend: includes view-through. A primary-conversion drop with All-conversions stable suggests one specific conversion action broke (often the primary purchase pixel), not all of them.
- Conversions by conversion action: per-action breakdown. If one specific action dropped to zero while others held, that’s the broken pixel. Diagnostic gold.
- Search Lost IS: a coverage metric, not relevant to a conversion drop unless lost-IS-budget shows a budget-cap kicked in.
- Conversions modelled vs observed: Google’s modelled conversions can fluctuate independently. A drop in observed but stable modelled may indicate a Consent Mode issue, not a tag break.
| Reason | Direction of divergence |
|---|---|
| Real-time event-ingestion lag (1-4 hour). The card runs hourly; Google Ads UI updates at variable cadence. | This card may flag a drop slightly earlier than Google Ads’ own anomaly indicator. |
| Suppression window. The alert is suppressed before 14:00 local time to allow morning ingestion to settle. | Google Ads UI shows the partial day raw and may look like a drop when in fact it’s just incomplete. |
| Attribution-model change. If model switched mid-window, Google Ads UI re-renders all rows; this card’s baseline (prior 7-day) does not retroactively update. | Card may flag a drop where Google Ads UI shows none, until the baseline rolls forward. |
| Currency. Not relevant for conversion count. | None. |
- Tag broke but real orders held (most common; cross-check commerce platform).
- Real demand dropped (promo ended, season turned, competitor launched).
- Attribution shifted (someone switched DDA <-> Last click; some channels lost credit, others gained).
- Conversion action was disabled in Google Ads (someone toggled it off; orders are still happening, just not tracked).
- Consent Mode v2 denial rate increased (more EU/UK users opted out; observed conversions drop, modelled conversions fill some gap).
| Card | Expected relationship | What causes legitimate divergence |
|---|---|---|
shopify.total_revenue (UTM-filtered to Google CPC) | If Google Ads conversions dropped but Shopify orders from Google CPC held, the issue is measurement, not performance. | Pixel breakage, Consent Mode denial increase. |
bigcommerce.total_revenue | Same logic as Shopify. | Same. |
google_analytics.ga_revenue_by_channel (Paid Search) | GA4 should show a similar shape if the underlying drop is real. If GA4 is normal, only the Google Ads pixel broke. | Different attribution windows; GA4’s last-non-direct may credit conversions Google Ads attributes to different campaigns. |