Inside this resource.
6 sectionsA map of the published document. Choose a section to start reading.
View 1 more sections
| Status | Meaning |
|---|---|
| PASS | Intended integration records the sampled purchase once with correct identifiers, value and currency under the selected consent state |
| FAIL | Verified missing, premature, duplicate or incorrect purchase event on the intended path; unavailable evidence is Not checked |
Source: View the complete rubric · Source-defined bands. Your account has not been scored.
Jump to a section
Put this guide to work
Review 15 critical/high/medium tracking checks and create a repair order with evidence.
Your task: Review critical tracking checks. Record the source, date range and reviewer beside the result. Leave anything you have not checked marked Not checked.
Related work: Check the foundations underneath your tracking. · Find the weakest layer of your measurement setup.
Severity: Critical
Prioritize verified event/value or consent defects by actual affected scope. The severity labels and time estimates are operator triage guidance, not measured financial loss or a guaranteed repair SLA.
Check 1: Google Tag Fires on Checkout Confirmation
What you are checking: The intended purchase integration records the correct event once under the selected consent state. Imported/server-side implementations may not fire a browser Google Ads tag.
How to check: Use an authorized test transaction and the appropriate browser/server/import diagnostics. For a browser tag, inspect Tag Assistant; for other integrations, trace the transaction through the native event/import path. Confirm ID, value, currency and duplicate handling, not a green icon alone.
Common failure: Tag fires on the wrong page (cart page instead of confirmation), or does not fire at all because the confirmation page loads a different GTM container or no container.
Time to fix: 15-30 minutes
| Status | Meaning |
|---|---|
| PASS | Intended integration records the sampled purchase once with correct identifiers, value and currency under the selected consent state |
| FAIL | Verified missing, premature, duplicate or incorrect purchase event on the intended path; unavailable evidence is Not checked |
Check 2: Click-Data Storage
Verify sitewide click-data capture under the selected consent state. A separate Conversion Linker is not always required when a Google tag already provides the behavior; see Google's linker guidance. Inspect actual permitted storage/network behavior rather than failing the setup solely for a missing tag name.
| Status | Meaning |
|---|---|
| PASS | Applicable capture/storage behavior verified |
| FAIL | Reproducible defect prevents required capture |
| Not checked | Access or test evidence unavailable |
Check 3: Cross-Domain Tracking Between Cart and Checkout
What you are checking: If your checkout is on a different domain or subdomain (common with Shopify, custom checkouts), session continuity is maintained across the domain boundary.
How to check: Map the actual domains, integration and consent state. For an applicable GA4 web flow, inspect consistent stream/tag configuration, linker behavior and user/session identifiers across navigation; use Realtime/DebugView as supporting evidence. A report glance alone does not diagnose a broken session, and same-cookie-domain subdomains may not require a cross-domain link.
Common failure: A required linker parameter or consistent identifier is lost across an applicable boundary. Verify actual configuration and identifiers rather than assuming every checkout-domain change breaks a session. A GA4 continuity issue can affect Analytics attribution; verify Google Ads conversion behavior separately rather than inferring loss from GA4 sessions.
Time to fix: 1-2 hours
| Status | Meaning |
|---|---|
| PASS | Single session maintained through checkout on different domain |
| FAIL | Verified session discontinuity; same-domain/inapplicable remains N/A and excluded from scoring |
Check 4: Enhanced Conversions Configured
What you are checking: Where applicable and permitted, enhanced conversions follow the supported implementation and selected consent design. Hashing does not itself establish permission.
How to check: Go to your Google Ads conversion action. Verify "Enhanced conversions" is toggled on. Then open Google Tag Assistant, complete a test purchase, and check that the conversion tag shows enhanced conversion parameters (hashed email, phone, or address data) in the tag details.
Time to fix: 1-2 hours
| Status | Meaning |
|---|---|
| PASS | Applicable authorized implementation and native diagnostics verified under the selected consent state |
| FAIL | Verified applicable implementation/data defect; inapplicable is N/A and unavailable diagnostics are Not checked |
Check 5: Consent Choices Match the Selected Design
Verify defaults, updates, accept/reject/withdrawal behavior and relevant ad_storage, analytics_storage, ad_user_data, ad_personalization states against Google's documentation and the accountable requirements review. Basic mode can block tags before consent; advanced mode can send limited denied-consent pings. Blocking tags after decline is not automatically a failure, and modeling availability is not guaranteed by configuration.
| Status | Meaning |
|---|---|
| PASS | Observed behavior matches the chosen valid design and reviewed requirements |
| FAIL | Reproducible choice/state or implementation defect |
| Not checked | Required evidence or accountable review unavailable |
Severity: High
Investigate actual effects on the selected goals and reports; prioritize verified impact and risk rather than assuming every check degrades bidding.
Check 6: No Duplicate Conversion Actions
What you are checking: The selected campaign goal does not count the same economic outcome twice. Multiple primary actions can be legitimate when they represent distinct intended outcomes; inspect custom-goal use too.
How to check: Go to Google Ads > Goals > Conversions. Filter to "Include in Conversions: Yes." Count how many purchase actions appear. Check if multiple actions are tracking the same event (common when GA4 import and Google Ads tag both track purchases).
Common failure: A GA4-imported purchase action AND a Google Ads conversion tag both count as primary. Check whether both actions actually count the same transaction toward the campaign goal. Multiple primary actions can inflate the selected outcome, but the magnitude must be measured.
Time to fix: 30-60 minutes
| Status | Meaning |
|---|---|
| PASS | Selected primary/custom-goal actions represent intended distinct outcomes without duplicate transaction counting |
| FAIL | Multiple purchase actions counting the same event |
Check 7: GA4 Linked to Google Ads
What you are checking: Your GA4 property is linked to the correct Google Ads account.
How to check: Go to GA4 Admin > Product Links > Google Ads Links. Verify the correct Google Ads account ID appears and the link is active. Then check Google Ads > Tools > Data Manager to confirm the connection from the other side.
Common failure: GA4 is linked to an old or wrong Google Ads account. Especially common after UA-to-GA4 migration, agency transitions, or when a brand has multiple Google Ads accounts. Features dependent on that link may be affected; independently configured tags and bidding goals can still work. Verify the actual audience/import/reporting path.
Time to fix: 15 minutes
| Status | Meaning |
|---|---|
| PASS | Correct GA4 property linked to correct Google Ads account |
| FAIL | Verified wrong or missing required association; multiple valid links alone are not a failure |
Check 8: Conversion Values Pass Dynamic Order Totals
What you are checking: The selected action's value and currency match its documented economic definition. A deliberate fixed lead value is different from an accidental fixed purchase-revenue placeholder.
How to check: Sample authorized events using transaction IDs and native tag/server/import evidence, then compare with matching orders. Google Ads aggregate reports do not expose a universal last-ten-orders export. Reconcile currency, taxes, shipping, discounts, refunds and the intended value basis; do not apply an arbitrary 5% tolerance.
Common failure: A static fallback value (like $50 or $1) is hardcoded in the conversion tag. Every order reports the same value regardless of actual order size. Google cannot distinguish a $30 order from a $300 order, so it cannot optimize toward higher-value customers.
Time to fix: 30-60 minutes
| Status | Meaning |
|---|---|
| PASS | Sampled ID/value/currency match the documented order or intentional action-value definition |
| FAIL | Verified unintended fallback, currency or value discrepancy; unavailable order matching is Not checked |
Checks 9–10: Inventory Containers and Direct Tags
List GTM containers, direct Google tags and commerce integrations by owner and purpose. Multiple containers or a direct tag alongside GTM are not inherently wrong. Inspect whether the same purchase actually generates duplicate events or conflicting values; do not remove a legitimate implementation solely because its ID appears in page source.
| Check | Pass | Fail |
|---|---|---|
| 9: Container inventory | Ownership, purpose and actual firing reviewed | Verified duplicate/conflicting behavior |
| 10: Direct-tag coexistence | Integration map and purchase readback coherent | Verified overlapping duplicate events or wrong values |
Severity: Medium
These create data drift over time. Fix them this month.
Check 11: Transaction ID Passed to Conversion Tag
What you are checking: A unique transaction ID is sent with every conversion, so Google can deduplicate if the tag fires more than once for the same order.
How to check: Open GTM Preview. Complete a test purchase. Click on the conversion tag in the Tags Fired panel. Check the tag parameters for "Transaction ID" or "Order ID." It should show a unique value (your order number), not be blank or undefined.
Common failure: Transaction ID field is empty or mapped to the wrong variable. If a customer reloads the confirmation page, Google counts it as a new conversion because there is no transaction ID to deduplicate against.
Time to fix: 30-60 minutes
| Status | Meaning |
|---|---|
| PASS | Unique transaction ID visible in tag parameters |
| FAIL | Empty, undefined, or static value |
Check 12: Micro-Conversions Tracked (Add to Cart, Begin Checkout)
What you are checking: You are tracking add_to_cart and begin_checkout events in GA4, and they are set as Secondary (not Primary) conversion actions in Google Ads.
How to check: Go to GA4 > Events. Look for add_to_cart and begin_checkout. Verify they have data in the last 7 days. Then go to Google Ads > Goals > Conversions. Verify these are listed as Secondary actions (not included in the primary conversion count).
Common failure: An event required for the intended funnel analysis is absent, duplicated or incorrectly selected for the campaign goal. Do not classify every micro-conversion or deliberate goal use as a bidding defect.
Time to fix: 1-2 hours
| Status | Meaning |
|---|---|
| PASS | Applicable funnel events are correct; any Ads import and goal use is deliberate, with purchase-focused goals protected from unwanted proxies |
| FAIL | Verified required event or selected-goal defect; an unnecessary Ads import is not required to pass |
Check 13: Referral Exclusions for Payment Gateways
What you are checking: Payment processor domains (PayPal, Klarna, Afterpay, etc.) are excluded as referral sources in GA4.
How to check: Go to GA4 Admin > Data Streams > your stream > Configure Tag Settings > List Unwanted Referrals. Check that your payment gateway domains are listed.
Common failure: A customer clicks "Pay with PayPal," gets redirected to paypal.com, then returns to your confirmation page. GA4 attributes the conversion to "paypal.com" as the referral source instead of the original Google Ads click. This can affect GA4 attribution. Check the actual journey and reports; it does not establish a corresponding change in independently tagged Google Ads conversions.
Time to fix: 15-30 minutes
| Status | Meaning |
|---|---|
| PASS | Applicable gateway-return referral behavior and selected configuration are verified; no unnecessary blanket domain list required |
| FAIL | Payment gateways showing as referral source in GA4 reports |
Check 14: Attribution Window Matches Your Sales Cycle
What you are checking: Your click-through and view-through attribution windows are set intentionally based on how long your customers actually take to purchase.
How to check: Go to Google Ads > Goals > Conversions > your purchase action > Settings. Check the click-through conversion window and the view-through conversion window. Compare to your actual time-to-purchase data (GA4 > Explore > Time to Conversion, or check your order data for average days between first visit and purchase).
Common failure: A30-day window includes later eligible interactions; that alone does not establish inflated purchases. Or a 1-day click window on a product with a 14-day research phase misses conversions. Either way, the algorithm gets the wrong feedback signal.
Time to fix: 15-30 minutes
| Status | Meaning |
|---|---|
| PASS | Attribution window matches actual purchase cycle within reasonable range |
| FAIL | Default window never reviewed, or window mismatched with sales cycle |
Check 15: Customer Match List Uploaded and Fresh
What you are checking: You have at least one customer list uploaded to Google Ads and it was updated within the last 90 days.
How to check: Go to Google Ads > Tools > Audience Manager > Your Data Segments. Look for customer lists. Check the upload date. Record available diagnostics and the metric definition; no universal 30% match threshold is assumed.
Common failure: An applicable authorized list is incorrectly scoped, outdated for its use or not refreshing as intended. No list or an eight-month upload age alone establishes neither lost bidding effectiveness nor a broken tracking stack.
Time to fix: 30-60 minutes
| Status | Meaning |
|---|---|
| PASS | Applicable permitted list and actual eligibility/diagnostics verified; count/age thresholds are worksheet choices |
| FAIL | Verified applicable list/data-flow issue; no applicable/authorized list is N/A, unknown diagnostics are Not checked |
Score Your Account
Count your results:
| Tier | Checks | Your Passes | Your Fails |
|---|---|---|---|
| Critical (1-5) | 5 | ___/5 | ___ |
| High (6-10) | 5 | ___/5 | ___ |
| Medium (11-15) | 5 | ___/5 | ___ |
| Total | 15 | ___/15 | ___ |
Interpretation
Count checked applicable results only; exclude Not checked and N/A from earned and possible points and show coverage. A pass means the specific observed check passed, not that the complete tracking stack is accurate. Prioritize actual event/value defects and consent behavior, then review integration and reporting limits. No top 30% benchmark or score-to-wasted-spend inference is established.
What This Checklist Does Not Cover
This checklist is the triage layer. It catches the most damaging problems fast.
The full diagnostic covers 78 checks across 5 layers, including:
- Server-side tracking validation and cookie integrity
- Consent Mode behavioral modeling verification
- Data layer integrity and tag firing order
- Cross-platform attribution discrepancy analysis
- GA4 event parameter standardization
- Audience signal quality and refresh schedules
For the complete team-operated workflow, review the Google Ads AI Agentic System. The product page lists the current deliverables, price, license terms and setup responsibilities.
If your score on this checklist was below 12, use the findings to compare the system’s scope with the work your team needs.
Choose who handles the next step.
Use the reference yourself, book a focused paid working session, or have Tegra handle the ongoing channel work.
Search, Shopping, Performance Max and Demand Gen campaigns, plus the research, product feeds, static and video creative, and landing, offer and pre-sale pages behind them.
Explore managed Google Ads → · See paid working sessions →