Step 1: Attribution
Overview: Attribution is an outdated form of digital marketing measurement. It helps identify which marketing activities are driving conversions. It's channel-based and focuses only on a small part of the customer journey.
Components:
Last-Click Attribution credits the final touchpoint before conversion. Simple but often overlooks earlier influences.
Multi-Touch Attribution distributes credit across multiple digital touchpoints, offering a slightly more holistic but still skewed view of the customer journey.
Benefits:
Identifies key digital touchpoints that influence conversions.
It helps allocate digital sales activation budgets more effectively.
Our recommendation:
Use last-click only to test the channel performance of sales activation campaigns.
DON'T USE multi-touch attribution, as it provides skewed data based on a small portion of the customer journey.
Step 2: Human Intelligence Incrementality
Overview: Human analysis of integrated marketing, sales and customer data to identify patterns and correlations that can improve marketing effectiveness. By human expertise, it aims to focus on the metrics that matter across SEE, THINK, DO, and CARE and measure the incremental impact by correlation over the short to medium term (weekly to 12 months).
We've tested the theory of drawing correlations first in an agile environment to build "data fitness" within teams. This step is critical for data adoption before investing in Steps 3 and 4.
Components:
Human eye observed correlations: Marketing analysts AND marketers examine integrated data for insights.
Agile testing: Implement iterative testing to validate the impact of these insights.
Benefits:
Leverages human domain expertise across marketing departments and experience to uncover subtle traction points.
Enhances the value of metrics that matter across the different stages (SEE, THINK, DO, CARE) of the customer journey.
Measurebyte best practices:
Develop a weekly habit of regular data review (we call these IdeationRooms) and develop correlation analysis 'muscle'.
Use agile methodologies to test and refine insights quickly.
Step 3: Artificial Intelligence Incrementality
Overview: Utilising complex AI and machine learning models to analyse data and understand the incremental impact of marketing efforts over the medium term (4 to 12 months).
Components:
Deep machine learning: AI models analyse large datasets to identify incremental sales and customer retention impacts.
Agile testing: Continual testing to confirm the validity of AI-driven insights.
Benefits:
Provides deeper, data-driven insights that can be missed by human analysis.
Improves precision in measuring the effectiveness of marketing activities.
Measurebyte best practices:
Integrate Measurebyte (or your own) models into data systems for seamless analysis.
Regularly validate AI insights through agile testing to ensure the accuracy and effectiveness of marketing.
Step 4: Marketing Mix Modeling (MMM)
Overview: Combining AI-driven insights with econometric analysis to optimise the overall marketing strategy over the long term (12 months+).
Components:
Integration of AI Incrementality and Econometrics: Uses AI insights and external factors like weather, competition, seasonality, and economic conditions.
Holistic analysis: Evaluates the combined effect of various factors on marketing performance.
Benefits:
Provides a comprehensive view of all factors affecting marketing effectiveness.
It helps in making strategic decisions that consider a wide range of variables.
Measurebyte best practices:
Use MMM to test different marketing scenarios and strategies for the year ahead.
Continuously refine the model to incorporate new data and changing market conditions.