Proper measurement is essential for gaining a clear understanding of your marketing costs and the value that each platform delivers. Without accurate data, it is impossible to know how much to spend and where to allocate resources effectively. Platforms like Google, Facebook, Tiktok, and LinkedIn each have unique strengths, but you need to know the actual value you’re getting from each to budget and prioritise your efforts moving forward.
Attribution Models and the Customer Journey
Attribution models, often viewed through tools like Google Analytics (GA), are different ways of explaining how various marketing interactions contribute to a conversion. They offer differing levels of accuracy depending on the tool used and help explain how one or more touchpoints lead to an outcome, such as a sale or a lead.
Traditionally, businesses rely on the marketing funnel to understand the customer journey. The funnel guides potential customers through stages like awareness, consideration and decision-making. At the top, people learn about a product, and as they move down, they evaluate options before making a purchase decision. This model assumes a linear path from start to finish, with marketing tactics designed to smoothly guide customers through each stage.
Copyright © 2024, Silkee Ltd. All rights reserved.
However, Google’s “Messy Middle” concept challenges this linear approach. Today’s consumer journey is far more complex, with customers bouncing between exploration and evaluation stages in a repetitive, non-linear fashion. They might interact with search engines, reviews, and social media multiple times before making a decision, resulting in an unpredictable and chaotic path. This highlights the limitations of the traditional funnel and shows how marketing interactions are less about leading customers through a straightforward journey and more about engaging them across various touchpoints.
The Rise of Data Silos and the Importance of Attribution
In reality, customers jump between different platforms and touchpoints as per Google’s “Messy Middle,” making attribution models vital for tracking these complex interactions. However, data is becoming increasingly siloed. Each platform is guarding user behaviour more tightly—Facebook only tracks interactions within its own ecosystem, while Google does the same. This makes it harder to get a complete picture of the customer journey across multiple channels.
An attribution model is a framework used to assign credit for a conversion to different touchpoints in a customer’s journey, such as ads, social media, emails, or website visits. Several core attribution models exist:
Core Attribution Models:
First-Touch Attribution:
Definition: Assigns 100% of the credit to the first interaction.
Scenario: A user first discovers a brand via a Facebook ad, receives an email, and later clicks a Google ad to purchase.
Example: 100% credit goes to the Facebook ad, as it introduced the customer to the brand.
Last-Touch Attribution:
Definition: Assigns all credit to the last interaction before purchase.
Scenario: A user interacts with a blog post, sees a retargeting ad on Instagram, and purchases after clicking a Google ad.
Example: 100% credit goes to the Google ad as it was the last interaction.
Linear Attribution:
Definition: Distributes credit equally across all touchpoints.
Scenario: A customer watches a YouTube video, clicks a Facebook ad, receives an email, and completes a purchase via a Google search ad.
Example: Credit is split equally across all four touchpoints, with each receiving 25%.
Time-Decay Attribution:
Definition: Gives more credit to touchpoints closer to the conversion.
Scenario: A user clicks on a display ad, reads a blog, clicks a Facebook ad, and finally buys via a Google search ad.
Example: The Google ad gets the most credit (50%), followed by the Facebook ad (30%), blog post (15%), and display ad (5%).
Position-Based Attribution (U-Shaped):
Definition: Assigns more credit to the first and last touchpoints.
Scenario: A user engages with a Facebook ad, interacts with an email, and finally purchases via a Google search ad.
Example: 40% credit goes to both the Facebook ad and Google search ad, and the email gets 20%.
Data-Driven Attribution:
Definition: Uses machine learning to assign credit based on the actual influence of each touchpoint.
Scenario: A customer reads a blog, clicks a Facebook ad, watches a YouTube video, interacts with an email, and purchases via Google.
Example: Based on historical data, credit might be distributed as 10% to the blog, 25% to Facebook, 15% to YouTube, 20% to email, and 30% to Google.
There is no single "best" attribution model that works for every situation, as the ideal model depends on the specific goals, customer journey, and complexity of the marketing strategy.
The Importance of Conversion Value in Attribution
When assessing attribution models, it’s crucial to consider conversion value. Just as in basketball, where different shots (3-pointers vs. 1-point free throws) carry different values, conversions in marketing also vary. If we only track the number of conversions without accounting for value (e.g., a small purchase vs. a large one), we may over-credit lower-value interactions.
Imagine two basketball players—one primarily makes difficult 3-point shots, while another focuses on easier free throws. If we only count the number of shots made, we might incorrectly give more credit to the player taking easier shots. In marketing, failing to assign proper conversion values can lead to overvaluing frequent but lower-value touchpoints. By incorporating conversion value, businesses can better understand which interactions drive high-value outcomes and adjust their strategies accordingly.
Factoring in Lifetime Value (LTV) Beyond Immediate Conversions
Attribution models often focus on immediate conversion value, but this overlooks the lifetime value (LTV) of a customer. LTV estimates the total value a customer brings over their entire relationship with the brand, not just from a single transaction. Focusing only on immediate conversions can lead businesses to prioritise short-term gains, while LTV encourages long-term thinking.
For instance, one channel might generate more one-time purchases, while another brings in fewer but repeat customers. By factoring LTV into attribution, businesses can better recognise the value of loyal, long-term customers, ensuring that marketing efforts are focused not just on quick wins, but on building sustainable relationships.
Adapting to Siloed Data with Geo Tests, Brand Lift, and Conversion Lift Tests
As cross-channel tracking becomes harder due to privacy regulations and data silos, marketers need new methods to understand the real value of their campaigns. Geo tests, brand lift tests, and conversion lift tests have become vital tools in this landscape.
Geo Tests: Split different geographic regions into test and control groups, exposing one to a campaign and comparing the results with the other. This helps reveal incremental impact, even without user-level data.
Brand Lift Tests: Measure the impact of campaigns on brand awareness, favourability, and intent by comparing groups that saw the campaign with those that didn’t. This method captures long-term brand perception shifts that aren’t tied to immediate conversions.
Conversion Lift Tests: Show the incremental conversions generated by a campaign by comparing conversion rates between a test group and a control group. These tests isolate the direct impact of the campaign on sales, even when cross-channel tracking is limited.
Why These Tests Matter
In today’s privacy-focused, siloed-data landscape, these testing methods offer a privacy-compliant way to measure effectiveness and guide resource allocation. They allow businesses to understand which marketing activities are driving incremental growth and long-term brand value, ensuring data-driven decision-making even in a fragmented ecosystem.
Ready to take your digital advertising to the next level? Contact us today for a free audit and see how our team can help you maximise your performance across all platforms.
コメント