tl;dr
Google channels will look a bit stronger after this update
Referral and Direct are expected to go down
The changes fix several under-the-hood issues that were causing inaccurate or inconsistent attribution — your data is now more reliable, not just different
Enhanced Journey Consistency and Stability
If you ever noticed small inconsistencies between different analysis runs, this is what fixed them. The entire data processing pipeline has been overhauled.
Unified sessionization
All calculations now happen in a single, unified step. Customer journeys are built once and remain consistent — no more subtle differences between runs.
More stable journey stitching
Shared user identifiers are now handled more carefully, preventing customer journeys from being incorrectly merged. Reporting is more stable over time as a result.
Improved Attribution Accuracy
Six specific fixes to credit channels more accurately and make attribution data more complete.
1. More accurate order counting
A bug was causing attributed orders to be undercounted — particularly for accounts with high order volumes. This has been fixed. All relevant orders are now included in your attribution analysis.
2. Smarter channel prioritisation
When a single session contains multiple touchpoints, credit is now assigned to the highest-ranked channel. The hierarchy is:
Paid Channels (e.g. Paid Search, Paid Social)
Owned Channels (e.g. Email, SMS)
Referral Channels
Other Channels (e.g. Direct, Organic Social)
Example: A user clicks an email campaign (Owned), then types your URL directly (Direct) in the same session. Credit now goes to the email campaign — not the direct visit. Your active marketing efforts get the credit they deserve.
3. Precise influencer tracking
A bug was causing duplicate touchpoints for influencer and partner sales. The fix counts unique discount code redemptions only — you may see a slight decrease in total influencer touches, but the number is now accurate.
Improved Google Ads coverage
4. Improved Google Ads coverage
Klar now reads gclid data directly from Shopify — even when there is no Pixel tagging on the visit. This improves Google Ads attribution coverage for sessions that wouldn't previously have been captured.
5. PPS touchpoint capping
PPS responses are now capped at one inferred touchpoint per order journey. Previously, answering multiple PPS questions (e.g. "Instagram" to one, "Influencer" to another) would create multiple redundant nodes in the journey. This no longer happens.
6. Post-conversion cooldown window
A short cooldown period now follows every conversion. Touchpoints triggered in this window — such as automated transactional emails or "Order Confirmed" page refreshes — are excluded from the next order journey. This prevents attribution noise from post-purchase activity being credited to a future order.
What to expect in your data
Across the board, these changes make attribution data more accurate rather than just different. The most noticeable directional shift:
Google channels will appear stronger — primarily due to improved gclid coverage and smarter channel prioritisation
Referral and Direct will appear lower — as they previously captured demand they didn't actually create