Why Attribution Matters
True understanding of impact
See how touchpoints work together from first interaction to conversion.
Budget & ROI
Allocate spend to what actually influences outcomes; optimise with confidence.
Resolve competing claims
Platforms often claim the same conversion — attribution helps reconcile the story.
What Is Attribution?
The practice of assigning conversion credit to channels, campaigns, creatives, and touchpoints across the journey.
Key Attribution Models
Last Click
100% credit to the final touchpoint; simple but ignores prior influence.
First Click
Credits initial discovery; great for awareness read, but misses later nudges.
Linear
Equal credit to all touches; balanced but blunt.
Time Decay
More credit near conversion; suits shorter cycles.
Position-Based
40% first + 40% last; 20% split mid-funnel.
Data-Driven
Algorithmic weights from observed impact; most accurate with enough data.
Last Click — The Default
Easy to understand and common in native platforms, but tends to over-invest bottom-funnel tactics.
First Click — Top-Funnel Insight
Highlights discovery channels and brand building; should be paired with other models for long paths.
Linear & Time Decay
- Linear: good for omni-channel reads.
- Time Decay: respects recency; realistic for short cycles.
Position-Based (U-Shaped)
Elevates discovery and decision moments; may under-read mid-funnel nurture.
Data-Driven Attribution (DDA)
Available in Google Ads, GA4, Meta, etc. Requires meaningful conversion volume and solid tagging. Expect learning time and differences vs single-touch models.
Google Ads vs GA4 Attribution
Google Ads: prioritises last Google Ads click; often shows higher counts.
GA4: multi-channel models with DDA default; includes non-Google touches.
Tip: set a source of truth, document windows, and reconcile deltas.
Cross-Platform Challenges
- Platform competition: multiple walled gardens claim the same conversion.
- View-through complexity: inconsistent weighting and windows.
- Privacy limits: iOS/cookie changes create blind spots.
Build unified dashboards and focus on incrementality to counter biases.
View-Through vs Click-Through Attribution
Clicks are stronger intent signals, but upper-funnel video/display/DOOH need view-through reads with sensible lookback windows.
Multi-Touch Attribution (MTA)
Maps the full path and weights touches algorithmically; powerful but data- and infra-intensive (most common in enterprise contexts).
Use Case: NZ Health Campaign
First touch
YouTube awareness (~25%).
Mid-funnel
Native content (~25%).
Last touch
Google Search (~50%).
DDA showed 30% CVR drop without YouTube; native engagement lifted conversion likelihood by 45%.
Attribution & Budget Planning
- Compare models and study top conversion paths.
- Shift budget to undervalued initiators and effective mid-funnel assists.
- Validate with incrementality tests; review quarterly.
GA4 Attribution Setup Essentials
Define key events
Purchases, forms, calls — validated and consistent.
Enable Google Signals
Cross-device visibility for truer paths.
Use path reports
Visualise journey and effective combos.
Primary model
Set DDA; compare with first/last for perspective.
Attribution in Meta Ads
Use compare attribution settings (e.g., 1-day click, 7-day click, 7-day view). Match windows to journey length and pair Meta reads with GA4/Looker for a blended view.
Mistakes to Avoid
- Assuming platform-reported ROAS is reality.
- Not comparing attribution models.
- Treating all conversions equally (AOV/LTV matter).
- Over-crediting brand search; ignoring offline touches.
How KBR Supports Attribution Strategy
Audits & model comparisons
Expose gaps between platform reports and real business performance.
Unified dashboards
Blend data across channels for a single view of journeys and paths.
Team education
Workshops that turn attribution into daily decision-making.
Ongoing optimisation
Reallocate budget with confidence as models and markets evolve.
Want Attribution That Drives Smarter Decisions?
We’ll help define the right model, reconcile platforms, and tie reporting to real outcomes.