Common Attribution Mistakes on Shopify

Q: What lookback window should we use?

Align to your purchase cycle. Calculate median days-to-purchase by cohort (e.g., paid social customers, organic customers). Set lookback window to 1.5x the median. For most DTC, 14-30 days is standard. Audit quarterly as seasonality shifts purchase cycle length.

Q: How do we know if attribution is wrong?

Run an incrementality test. Pause your largest paid channel in one region for 2 weeks; compare revenue in the paused region to a control region. If attributed ROAS is 3:1 but incrementality test shows 1.5:1, attribution is overstating channel contribution. Use test results to calibrate your model.

Last-Click Attribution as Default

Last-click attribution credits the final touchpoint before purchase with 100% conversion value. Shopify's native analytics default to this model. The failure mode: a customer sees a Facebook ad on Day 1, clicks an organic search result on Day 7, and the organic channel receives full credit despite the ad driving initial awareness.

Diagnosis: Run a cohort analysis. Compare customer LTV by first-touch channel vs. last-touch channel. If organic appears 3x more valuable than paid in last-click but your paid spend is driving 60% of new customer acquisition, last-click is misrepresenting channel contribution.

Threshold: If last-click attribution shows >80% of conversions from a single channel and your media mix is diversified, audit the model. Last-click should rarely show such concentration unless one channel genuinely dominates the funnel.

Last-click credits final touchpoint with 100% value
Ignores awareness and consideration stages
Overvalues bottom-funnel channels (organic, direct)
Undervalues top-funnel channels (paid social, display)

Ignoring Cross-Device Journeys

A customer browses on mobile, adds items to cart, then completes purchase on desktop. Shopify's standard tracking treats these as separate sessions. The failure mode: mobile is credited with the add-to-cart event, desktop with the purchase, and the relationship between the two is invisible.

Cross-device tracking requires either first-party data (logged-in users) or third-party pixel matching. Most Shopify stores have <30% logged-in checkout rates, meaning 70% of journeys are fragmented across devices.

Diagnosis: Check your Google Analytics 4 cross-device reports. If device-level conversion rates show desktop at 4% and mobile at 0.8%, but mobile sessions outnumber desktop 3:1, cross-device abandonment is likely the issue, not mobile channel quality.

Mobile and desktop tracked as separate sessions
Cart abandonment appears channel-specific when it's device-specific
Requires first-party data or pixel matching to resolve
GA4 cross-device reporting is the diagnostic tool

Misaligned Lookback Windows

Lookback window is the time period during which a touchpoint can receive attribution credit. Facebook defaults to 7-day click, 1-day view. Google Ads defaults to 30-day click. Shopify's native analytics use 30-day session-based windows.

Failure mode: A customer clicks a Facebook ad on Day 1, browses for 20 days, then returns via direct search on Day 25 and purchases. Facebook's 7-day window excludes this customer entirely, while Google's 30-day window may credit the ad. The same conversion is attributed differently depending on which platform's window is used.

Decision rule: Align lookback windows to your actual purchase cycle. For fashion (7-14 day cycle), use 14-day windows. For considered purchases (30+ days), use 30-day or longer. Audit by cohort: segment customers by days-to-purchase and set windows 1.5x the median cycle time.

Facebook: 7-day click, 1-day view (default)
Google: 30-day click, 30-day view (default)
Shopify: 30-day session-based
Mismatch creates inconsistent credit assignment

Conflating Correlation with Causation

A store increases Instagram spend by 50% and sees revenue rise 40%. Attribution reports show Instagram driving the growth. Failure mode: a seasonal lift, competitor exit, or press mention caused the revenue increase, and Instagram merely correlated.

True causation requires incrementality testing. Run a holdout test: pause Instagram ads in one geographic region for 2 weeks while maintaining spend elsewhere. Compare the paused region's revenue to the control region. If the paused region's revenue drops 15%, Instagram's true incremental contribution is ~15%, not the 40% attributed.

Threshold: If a single channel shows >50% YoY growth in attributed revenue without corresponding changes in spend efficiency (CPC, ROAS), run an incrementality test before reallocating budget. Correlation without testing is a common path to budget misallocation.

Seasonal lifts appear channel-driven
Competitor exits create false attribution spikes
Incrementality testing isolates true causation
Holdout tests are the diagnostic standard

Not Accounting for Assisted Conversions

Assisted conversions are touchpoints that contributed to a purchase but were not the last click. A customer clicks a Google ad (assisted), then returns via email (last-click) and purchases. Standard Shopify reporting credits email with 100% of the conversion.

Failure mode: Email appears highly efficient (low cost per conversion) while Google Ads appears inefficient (high cost per conversion). In reality, Google drove the initial customer acquisition and email drove the return visit. Reallocating budget away from Google based on this misreading will shrink the top of the funnel.

Measurement: Use multi-touch attribution models. Shopify does not natively support this, but Google Analytics 4's data-driven attribution or Shopify apps (e.g., Littledata, Ruler Analytics) can assign fractional credit across touchpoints. Start with a 40-20-40 model: 40% to first touch, 20% to middle touches, 40% to last touch. Adjust based on your funnel shape.

Last-click ignores assisted touchpoints
Email appears more efficient than it is
Paid acquisition channels appear less efficient
Multi-touch models distribute credit across journey

Ignoring Dark Traffic and Direct Conversions

Dark traffic is traffic from sources that don't pass referrer information (email clients, messaging apps, QR codes, offline mentions). Direct traffic includes users who type the URL or use bookmarks. These are often grouped together and treated as 'earned' or 'organic'.

Failure mode: A customer receives a text message with a link to the store (dark traffic), clicks it, and purchases. Attribution reports show 'direct' as the source. The SMS campaign is invisible, and budget is reallocated away from SMS based on false attribution.

Diagnosis: UTM parameters are the control. Require all marketing links to include utm_source, utm_medium, utm_campaign. Any traffic without UTMs is either dark traffic or a tracking gap. Set a threshold: if >15% of traffic is untagged, audit your link-building process. Use UTM audits (Supermetrics, Ruler Analytics) to catch missing tags before they distort attribution.

Dark traffic: email clients, messaging apps, QR codes
Direct traffic: bookmarks, typed URLs
Both appear as 'organic' or 'direct' without UTMs
UTM parameters are the control mechanism

Attribution Checklist for Shopify Operators

Use this checklist to audit attribution setup and identify failure modes before they distort budget decisions.

Confirm all marketing links use consistent UTM structure (source, medium, campaign, content)
Document your purchase cycle length by cohort (days from first touch to conversion)
Set lookback windows to 1.5x median purchase cycle time across all platforms
Run an incrementality test on your largest paid channel (2-week holdout, 1-2 geographic regions)
Compare last-click attribution to multi-touch attribution; if channel rankings differ by >30%, investigate
Audit dark traffic: ensure SMS, email, QR codes, and offline mentions are tagged
Check cross-device tracking: confirm GA4 cross-device reports show device-level conversion rates
Document your attribution model (last-click, first-click, linear, data-driven) and review quarterly

Questions

FAQ

Should we use last-click attribution on Shopify?

Last-click is a starting point, not a decision framework. Use it to identify which channels drive final conversions, but pair it with first-click and multi-touch models to understand the full journey. If last-click shows >80% concentration in one channel, audit for misattribution before reallocating budget.

How do we track customers across devices on Shopify?

Native Shopify analytics do not track cross-device journeys. Use GA4 (free, cross-device reporting included) or third-party apps (Littledata, Ruler Analytics). Require customer login at checkout to enable first-party cross-device matching. If login rates are <30%, expect 70% of journeys to remain fragmented.

What lookback window should we use?

Align to your purchase cycle. Calculate median days-to-purchase by cohort (e.g., paid social customers, organic customers). Set lookback window to 1.5x the median. For most DTC, 14-30 days is standard. Audit quarterly as seasonality shifts purchase cycle length.

How do we know if attribution is wrong?

Run an incrementality test. Pause your largest paid channel in one region for 2 weeks; compare revenue in the paused region to a control region. If attributed ROAS is 3:1 but incrementality test shows 1.5:1, attribution is overstating channel contribution. Use test results to calibrate your model.

Want this on your account?

Thirty minutes. Bring the number that keeps you up.

Book a demo