Weekly Attribution Review Template

Why Weekly Attribution Reviews Matter

Attribution drift happens silently. A pixel stops firing. A UTM parameter gets dropped in a redirect chain. A platform changes its conversion window. None of these trigger alerts—they just slowly shift revenue from one channel to another.

Weekly reviews catch these shifts while they're still small. A 5% misclassification in week one becomes 20% by month end if unaddressed. For a $100k/week revenue brand, that's $20k in misdirected budget by the time quarterly reporting happens.

The goal is not perfect attribution (impossible). The goal is consistent attribution—the same rules applied the same way every week, so trends are real and not artifacts of measurement drift.

Core Metrics to Audit Weekly

Start with five numbers. Pull them every Monday morning from your primary analytics platform (GA4, Shopify, or custom warehouse). Compare each to the prior week and the same week last year.

Revenue by channel (organic, paid search, social, email, direct, referral, other) - threshold: no single channel should swing more than 15% week-over-week without a known campaign change
Conversion rate by channel - threshold: if any channel drops below its 90-day rolling average by more than 20%, flag for pixel audit
Cost per acquisition (CPA) by paid channel - threshold: if CPA increases more than 10% without corresponding volume increase, check for attribution lag or pixel loss
Assisted conversions / multi-touch attribution - threshold: if assisted conversion share swings more than 5 percentage points, validate your attribution model settings
Unattributed revenue - threshold: should stay below 5% of total. Anything above 8% signals a tracking gap

Data Integrity Checklist

Before interpreting the numbers, verify the pipes are clean. Use this checklist every week.

Pixel firing: Pull last 100 orders from Shopify. Manually check 10 random orders. Verify the conversion pixel fired on the thank-you page (check browser console, network tab, or pixel inspector tool). If fewer than 9/10 fired, escalate to dev
UTM parameter consistency: Spot-check 5 active campaigns. Verify utm_source, utm_medium, utm_campaign are populated and match your naming convention. Check for typos, case mismatches, or missing parameters
Platform sync lag: Compare yesterday's revenue in Shopify to yesterday's revenue in GA4. Threshold: should match within 2%. If gap is larger, wait 24 hours and recheck (GA4 can lag up to 48 hours)
Conversion window alignment: Confirm all platforms use the same attribution window (e.g., 30-day click-to-conversion). If they differ, document the difference and adjust comparisons
Segment definitions: Verify your channel groupings haven't changed. If GA4 rules were updated, rerun last week's data with new rules to ensure apples-to-apples comparison

Failure Modes and Escalation

Some issues require immediate action. Others can wait for the next sprint. Use this decision tree.

Immediate (fix this week): Pixel not firing on thank-you page, unattributed revenue above 10%, revenue discrepancy between Shopify and GA4 exceeding 5%, UTM parameters missing from active campaigns
High priority (fix within 2 weeks): Conversion rate drop exceeding 25% in a single channel, CPA increase exceeding 20% without volume explanation, assisted conversion share swinging more than 10 percentage points, new traffic source appearing in GA4 but not in Shopify
Standard priority (next sprint): Channel revenue swinging 10-15% without known cause, minor UTM inconsistencies in paused campaigns, attribution model refinements, documentation updates

Common Attribution Mistakes to Catch

These errors appear in most DTC operations. A weekly review catches them before they distort budget allocation.

Direct traffic inflation: Untagged traffic defaults to 'direct.' Check if email, SMS, or retargeting campaigns are missing UTM tags. If 'direct' exceeds 20% of revenue, investigate
Last-click bias: If your platform only uses last-click attribution, organic and email appear worthless. Document this limitation and use assisted conversions as a secondary metric
Conversion window mismatch: If you're comparing 7-day attribution (Facebook) to 30-day attribution (GA4), you'll see Facebook underperform. Standardize or adjust comparisons
Mobile app traffic: If you have an app, verify in-app conversions are being tracked and attributed correctly. App traffic often gets lumped into 'direct' if not configured properly
Affiliate and partner channels: If partners drive traffic via their own links, ensure their traffic is tagged and segmented. Untagged partner traffic often gets misclassified as organic

Template: Weekly Attribution Audit Log

Use this format every week. Keep a running log in a shared sheet. This creates a paper trail for debugging and forecasting.

Date: [Monday of review week]
Revenue by channel: [Organic: $X, Paid Search: $X, Social: $X, Email: $X, Direct: $X, Other: $X]
Week-over-week change: [% change for each channel]
Data integrity notes: [Pixel status, UTM check, platform sync, conversion window, segment definitions - pass/fail for each]
Anomalies detected: [List any metric outside threshold]
Root cause (if known): [What changed - campaign launch, platform update, technical issue, etc.]
Action items: [What needs to be fixed, who owns it, deadline]
Confidence level: [High / Medium / Low - how much do you trust this week's data]

Setting Up Automation

Manual audits work, but automation catches more. Set up alerts for the five core metrics so anomalies surface without waiting for Monday.

Use GA4 alerts (free) to flag when conversion rate drops more than 20% or revenue swings more than 15%. Set up Shopify webhooks to log all conversions to a database, then query it daily for unattributed orders. If your analytics platform supports it, schedule weekly reports that compare current week to prior week and prior year - the comparison view surfaces drift faster than raw numbers.

Questions

FAQ

How do I know if my attribution model is wrong?

Compare assisted conversions to last-click conversions. If assisted conversions are more than 30% of total conversions, your last-click model is understating certain channels (usually organic and email). This isn't 'wrong' - it's a limitation you need to document. Run a secondary analysis using assisted conversions to see if budget allocation decisions would change. If they would, your model is hiding important data.

What's the difference between unattributed revenue and direct traffic?

Unattributed revenue is orders that don't match any conversion event in your analytics platform - usually because the pixel didn't fire or the customer cleared cookies. Direct traffic is traffic that arrived without a referrer (typed URL, bookmarked, or referred from a non-tracked source). Direct traffic is attributed; unattributed revenue is not. High unattributed revenue (above 8%) signals a tracking problem. High direct traffic (above 20%) signals either strong brand awareness or missing UTM tags.

How often should I recalibrate my attribution model?

Quarterly minimum. Review your model settings (conversion window, channel groupings, assisted conversion rules) every 13 weeks. If you made major changes to your marketing mix (launched a new channel, killed a campaign, changed platforms), recalibrate immediately. Document each change so you can explain year-over-year comparisons.

What should I do if two platforms show different revenue numbers?

First, confirm they're measuring the same thing. Shopify reports all orders; GA4 reports only orders with a tracked conversion event. If Shopify shows $100k and GA4 shows $95k, the $5k gap is likely unattributed orders. Wait 48 hours for GA4 to fully process data, then recheck. If the gap persists and exceeds 5%, audit your pixel firing and UTM implementation. If the gap is consistent (always 5%), document it and use it as your baseline going forward.

Want this on your account?

Thirty minutes. Bring the number that keeps you up.

Book a demo