Helping brands understand the impact of marketing tactics on business outcomes

Half my advertising spend is wasted; the trouble is, I don't know which half.”

John Wanamaker

Marketing Mix Modeling (MMM)

MMM focuses on the big picture, analyzing historical aggregate data to understand the long-term impact of various marketing channels (like TV, radio, print, digital ads) on overall business outcomes (e.g., sales, revenue, brand awareness). It's particularly good at:

  • Strategic Allocation: Helping businesses decide how to allocate larger marketing budgets across different traditional and digital channels over time.

  • Understanding Macro Trends: Identifying the general effectiveness of broad marketing activities, even those that are hard to track individually, like brand-building campaigns.

  • Incorporating External Factors: Accounting for non-marketing influences on business results, such as seasonality, economic conditions, and competitor activities, which provides a more holistic view.

Methodology

MMM typically uses econometric modeling (often regression analysis) to identify statistical relationships between marketing spend, external factors, and business outcomes.

  1. Data Collection: Gathers aggregated historical data on marketing spend (e.g., weekly TV ad spend, monthly digital ad spend), sales figures, and external variables (e.g., GDP, competitor promotions, weather).

  2. Model Building: Statistical models are built to determine the unique contribution of each marketing channel and external factor to the observed business results. For instance, the model might reveal that every $1 invested in TV ads historically led to an average of $X in sales.

  3. Insight Generation: The model quantifies the return on investment (ROI) for each channel, helping to forecast future outcomes and optimize budget allocation for maximum impact.

Problem statements

Marketing Mix Modeling (MMM) and Multi-Touch Attribution (MTA) are two powerful approaches that help businesses understand and optimize the impact of their marketing efforts. They essentially answer the question: "How much did our marketing contribute to our business results, and where should we invest next?"


Multi-Touch Attribution (MTA)

MTA, on the other hand, dives into the granular details of individual customer journeys. It tracks every touchpoint a customer interacts with on their path to conversion (e.g., clicking a paid search ad, visiting a website, opening an email, seeing a social media ad). MTA excels at:

  • Tactical Optimization: Providing insights for optimizing specific campaigns, ad creatives, and audience targeting in real-time or near real-time.

  • Understanding Customer Journeys: Revealing the sequence and impact of different marketing interactions, helping marketers understand which touchpoints are most influential at various stages of the customer's decision-making process.

  • Granular Performance Measurement: Assigning credit to each touchpoint that contributes to a conversion, offering a more precise view of channel performance down to the individual impression or click.

Methodology

MTA relies on user-level data and various attribution models to assign credit.

  1. Data Collection: Gathers data from various digital platforms (e.g., ad servers, website analytics, CRM systems) to track individual user interactions across different marketing channels. This often involves using cookies or unique identifiers.

  2. Attribution Model Application: Different models are applied to distribute credit among the touchpoints. Common models include:

    • Last-Click: Gives all credit to the final touchpoint before conversion.

    • First-Click: Gives all credit to the first touchpoint.

    • Linear: Distributes credit equally among all touchpoints.

    • Time Decay: Gives more credit to touchpoints closer to the conversion.

    • Algorithmic/Data-Driven: Uses sophisticated statistical methods (like Markov chains or shapley values) to assign credit based on the actual contribution of each touchpoint, considering the sequence and interaction effects.

  3. Insight Generation: Provides a detailed understanding of the most effective touchpoints and paths to conversion, enabling marketers to optimize budgets and campaigns at a highly granular level.

Case Studies

Helping Viking Cruises navigate stormy waters during the global pandemic

  • What should we know about the services you provide? Better descriptions result in more sales.

  • What should we know about the services you provide? Better descriptions result in more sales.

Dominating the living room with Sonos home theater products

  • What should we know about the services you provide? Better descriptions result in more sales.

  • What should we know about the services you provide? Better descriptions result in more sales.

The Lime Truck: Enabling robust lead scoring by solving for marketing-contributed opportunity value

  • What should we know about the services you provide? Better descriptions result in more sales.

  • What should we know about the services you provide? Better descriptions result in more sales.