In the world of digital marketing, understanding which touchpoints lead to conversions is crucial. However, tracking and optimizing conversions across multiple platforms like Facebook Ads, Google Ads, and Google Analytics 4 (GA4) is far from straightforward. Let’s dive into the technical aspects and challenges of cross-platform attribution.
The Complexity of Modern Customer Journeys
Today’s customer journeys are complex, often involving multiple devices and platforms. Users might see a Facebook ad on their phone, later search on Google from their laptop, and finally convert on a tablet. This multi-touch, cross-device behaviour creates significant challenges for accurate attribution.
Technical Challenge: Cross-Device Tracking
Different platforms use various methods to track users across devices:
- Deterministic matching: Uses unique identifiers like user logins.
- Facebook’s advantage: Logged-in users across devices.
- Google’s Edge: Gmail, YouTube, and other service logins.
- Probabilistic matching: Uses machine learning to infer cross-device usage based on IP addresses, browsing patterns, and other signals.
Neither method is perfect, leading to discrepancies in cross-platform reporting.
Platform-Specific Attribution Models
Each platform uses its attribution model, which can lead to overlapping or conflicting conversion claims.
Facebook Attribution
Facebook uses a “last-touch” model within its attribution windows. However, it also offers data-driven attribution models considering the full user journey within its ecosystem.
Technical note: Facebook’s pixel fires on page load, which can sometimes lead to over-attribution if users don’t complete an action.
Google Ads Attribution
Google Ads provides several attribution models:
- Last Click
- First Click
- Linear
- Time Decay
- Position-Based
- Data-Driven Attribution (DDA)
Technical insight: Google’s DDA uses Shapley values from cooperative game theory to distribute credit among touchpoints.
Google Analytics 4 (GA4) Attribution
GA4 uses a data-driven approach by default, which differs from Facebook and Google Ads.
Key difference: GA4’s data-driven model considers all touchpoints, including non-paid channels, which can lead to different attribution results than platform-specific ad reporting.
Side Note: More about GA4 you can read here, in my previous blog post.
The “All Conversions” Conundrum in Google Ads
Google Ads’ “All Conversions” metric includes:
- Standard conversions
- View-through conversions
- Cross-device conversions
This comprehensive approach can provide a more complete picture but may also lead to higher conversion numbers than other platforms.
in the other words: all_conversions = standard_conversions + view_through_conversions + cross_device_conversions
Technical Challenge: View-Through Attribution
View-through conversions are particularly tricky. They’re counted when a user sees an ad but doesn’t click and later converts. The technical implementation varies:
- Google uses a 1-day window for display ads.
- Facebook allows customizable view-through windows.
This disparity can lead to significant differences in reported conversions between platforms.
Data Discrepancies and Their Causes
Several factors contribute to data discrepancies across platforms:
- Different attribution windows: Facebook: Customizable, up to 28-day click, 1-day view Google Ads: 30-day click, 1-day view (customizable) GA4: 90-day lookback window
- Sampling: GA4 may use sampling for large datasets, while ad platforms typically don’t.
- Time zone differences: Conversions might be attributed to different days due to platform-specific time zone settings.
- Cookie limitations: Intelligent Tracking Prevention (ITP) and ad-blockers can limit cookie-based tracking.
The Impact of Privacy Regulations and Tracking Prevention
Recent privacy regulations (GDPR, CCPA) and tracking prevention measures (like Apple’s ITP) have further complicated attribution.
Technical impact: These changes have reduced the effectiveness of third-party cookies, pushing platforms towards first-party data and probabilistic attribution methods.
Why Precise Optimization is Challenging
Given these complexities, precise optimization across platforms becomes a significant challenge:
- Data silos: Each platform has its own data, making it difficult to get a unified view.
- Attribution conflicts: Different models can assign credit to the same conversion differently.
- Delayed data: Some conversions (especially in B2B) can happen long after the initial touchpoint, outside of standard attribution windows.
- Incremental lift measurement: It’s hard to determine the true incremental impact of each channel or campaign.
Advanced Solutions and Future Directions
To address these challenges, marketers are turning to:
- Marketing Mix Modeling (MMM): Uses statistical analysis to measure the impact of multiple marketing inputs on sales or conversions.
- Multi-touch Attribution (MTA): Attempts to assign credit to multiple touchpoints in the customer journey.
- Unified measurement approaches: Combining MMM and MTA for a more comprehensive view.
- AI and Machine Learning: Leveraging advanced algorithms to process vast amounts of data and uncover attribution insights.
In conclusion, while platforms like Facebook Ads, Google Ads, and GA4 provide valuable insights, the complexity of modern digital customer journeys means that no single platform can provide a complete picture. Marketers must use a combination of tools, models, and analysis techniques to get closer to understanding true attribution and optimizing their marketing mix effectively.
Remember, the quest for perfect attribution is ongoing. As technology evolves, so will our ability to track and attribute conversions more accurately. Stay curious, keep testing, and always be ready to adapt your attribution strategies.