Unlocking CPG Market Share: A Marketing Mix Modeling Case Study on Budget Optimization

A leading CPG giant was investing over $500 million annually in marketing, yet its market share was stagnating, and profitability was eroding. The core issue was a complete lack of visibility into which channels were actually driving sales and which were draining resources. Traditional attribution models were failing to capture the impact of offline media and complex trade promotions, making a strong case for a comprehensive marketing mix modeling approach to untangle the intricate web of consumer touchpoints and external market forces. The leadership team could no longer justify its nine-figure budget based on historical inertia and gut feelings. They needed a data-driven, quantitative method to prove and improve their marketing effectiveness. This case study details how Quantzig’s advanced analytics engagement not only identified over 20% in budget inefficiency but also provided a dynamic framework for reallocating spend. This strategic shift led to a measurable 4.2% uplift in incremental sales and a significant improvement in marketing ROI within just two quarters, transforming the marketing function from a perceived cost center into a proven engine of growth.

Key Highlights

• Stagnant Growth Despite Massive Marketing Spend

  Our client, a global CPG company with a portfolio of household-name brands, was grappling with flat sales despite a marketing budget exceeding half a billion dollars. Their primary objective was to shift from a volume-at-all-costs mindset to one of profitable growth. The core challenge was that decades of accumulated marketing wisdom were proving ineffective in a fragmented media landscape. The company lacked a unified analytical framework to compare the effectiveness of a national TV campaign against a targeted social media push or a retailer-specific trade promotion. This led to budget allocation based on precedent rather than performance, threatening the company's competitive position and alarming stakeholders who demanded greater accountability for the massive marketing expenditure.

• Inability to Measure True Marketing ROI

  The client's marketing budget was a 'black box.' They were unable to accurately distinguish the impact of their diverse marketing activities—from TV and radio ads to digital campaigns, trade promotions, and in-store displays—on overall sales. A fundamental problem was the inability to separate base sales (what they would sell anyway) from incremental sales directly attributable to marketing efforts. This ambiguity made it impossible to calculate a credible marketing ROI analysis for any given channel. Consequently, marketing budget allocation was a contentious process driven by negotiation and intuition, not by data-backed evidence of what truly moved the needle on sales and market share. The lack of a clear performance metric was a major strategic vulnerability.

• Advanced Econometric Modeling for Granular Insights

  Quantzig's solution was to design and implement a bespoke marketing mix modeling (MMM) framework. This was not an off-the-shelf tool but a deeply customized analytical engagement. We began by integrating over three years of weekly data from more than 50 disparate sources, including syndicated scanner data, internal shipment records, media agency reports, and economic indicators. Our data scientists then developed a series of advanced econometric models. These models statistically isolated the impact of each marketing lever while controlling for external factors. Crucially, the models incorporated business realities like advertising 'adstock' (the lingering effect of ads) and the law of diminishing returns on spend.

• Achieved 18% Improvement in Marketing Effectiveness

  The analytical engagement delivered a profound impact, headlined by an 18% improvement in overall marketing effectiveness. This meant the client could achieve the same sales volume with an 18% lower budget, or reinvest those savings into high-performing channels to accelerate growth. The insights from the marketing mix modeling directly led to a strategic reallocation of over $30 million, shifting funds from low-ROI traditional media to high-impact digital and in-store activation. This data-driven pivot resulted in a 4.2% increase in incremental sales and improved sales forecast accuracy by over 15%, enhancing supply chain and inventory planning. The project provided the C-suite with the credible, data-driven justification for marketing spend they had long sought.

Problem Statement

A premier Consumer Packaged Goods (CPG) corporation found itself at a critical juncture. Despite commanding significant market presence and investing hundreds of millions in marketing annually, the company faced a persistent challenge: an inability to scientifically measure the return on its marketing investments. The leadership team was flying blind, allocating massive budgets based on historical patterns and qualitative feedback rather than empirical evidence. This lack of clarity created significant business friction, pitting the marketing team, who advocated for larger budgets, against the finance team, who demanded accountability and proof of ROI. The core of the problem lay in a fragmented and chaotic data landscape. Data from various sources—Nielsen/IRI scanner data, internal financial records, digital analytics platforms, and media agency reports—existed in isolated silos, each with its own metrics and reporting cadence. There was no single source of truth to provide a holistic view of performance. This data disarray made it impossible to answer fundamental strategic questions: Was the multi-million dollar investment in television advertising more effective than targeted digital campaigns? How much did in-store promotions cannibalize future sales? Without a robust marketing mix analysis, the company was caught in a cycle of inefficient spending, missed opportunities, and an ever-growing pressure to justify its marketing strategy to an increasingly skeptical board.

• Fragmented Data Ecosystem : The client struggled with a tangled web of data sources. Critical retailer data arrived in inconsistent formats, media spend reports from different agencies lacked a common taxonomy, and internal sales data was not granular enough. Harmonizing these disparate time-series datasets into a single, analysis-ready format was a monumental task that their internal teams were not equipped to handle, preventing any meaningful CPG marketing analytics.
• Isolating Marketing Impact : A key challenge was separating the true impact of marketing activities from a host of external factors. Sales could rise or fall due to seasonality, competitor actions, changes in consumer confidence, or even the weather. Without the ability to statistically control for these variables, any analysis of marketing effectiveness was fundamentally flawed and led to incorrect conclusions about channel performance and ROI.
• Measuring Cross-Channel Effects : The client's marketing efforts did not exist in a vacuum. A TV ad might drive a consumer to search online; a social media campaign could increase the effectiveness of an in-store display. The existing analytical capabilities could not measure these crucial synergistic (or cannibalistic) effects between channels. This meant they were potentially under-investing in channels that acted as key influencers in the customer journey, simply because their direct impact was difficult to measure.
• Static and Inflexible Budgeting : The company was locked into a rigid annual budgeting process. Budgets were set once a year and rarely adjusted, regardless of in-flight campaign performance or shifts in the market. This static approach prevented any form of marketing spend optimization during the fiscal year. Opportunities to double down on a successful campaign or pull funding from a failing one were consistently missed, leading to a significant waste of resources and a slower response to competitive threats.

The breaking point arrived during the annual budget review. The Chief Marketing Officer presented a compelling narrative for a 10% budget increase to fend off a new competitor. The Chief Financial Officer responded not with a counter-argument, but with a simple, direct question: 'Show me the incremental dollar of profit we get for every new dollar you want to spend, broken down by channel.' The silence in the room was deafening. The marketing team could produce reams of data on brand awareness, engagement rates, and media impressions, but they could not provide the one financial metric the board truly cared about. The budget request was denied, and the existing marketing spend was frozen pending a full audit. This public and painful moment crystallized the reality that the status quo was no longer survivable. The company's inability to speak the language of financial accountability had become a critical business risk. It was clear that they needed more than just better reports; they needed a fundamental transformation in how they measured and managed marketing performance, which led them to seek an expert solution in marketing mix modeling.

Objectives

• Quantify Sales Drivers : The primary objective was to move beyond assumptions and statistically determine the precise contribution of each sales driver. This involved creating a model that could quantify the percentage of sales driven by base demand, TV advertising, digital marketing, trade promotions, pricing changes, and other key variables. Achieving this would provide an unprecedented, fact-based understanding of the business.
• Optimize Budget Allocation : A crucial goal was to build a forward-looking simulation tool. This tool would empower marketing leaders to run various 'what-if' scenarios, such as 'what happens to our sales if we shift $10M from TV to digital?' The objective was to enable data-driven decisions on budget allocation to maximize overall marketing ROI, rather than relying on historical precedent or gut feel.
• Establish a Credible ROI Metric : The engagement sought to establish a consistent and reliable methodology for calculating marketing ROI across all channels. This would create a 'common currency' for performance, allowing for apples-to-apples comparisons between vastly different activities like a Super Bowl ad and a search engine marketing campaign. This single metric would become the cornerstone of marketing accountability.
• Enable Agile Marketing Planning : A strategic objective was to break free from the rigid annual planning cycle. The goal was to provide the tools and insights necessary for a more dynamic and agile approach to marketing management. This would allow the team to make data-informed adjustments to the marketing plan on a quarterly or even monthly basis, responding effectively to real-time performance data and market changes.

Solution Implemented

Quantzig addressed the client's challenges by deploying a comprehensive, multi-phased marketing mix modeling engagement designed to deliver not just a one-time analysis, but a sustainable analytics capability. Our approach was centered on translating complex data into clear, actionable business strategies. The first phase involved a thorough data audit and harmonization process, creating a unified analytical dataset from over 50 fragmented sources. In the second phase, our team developed sophisticated econometric models to dissect historical performance and quantify the impact of every marketing dollar. The final phase focused on operationalizing these insights through an interactive budget simulation tool and a strategic insights report. This end-to-end solution provided a clear view of marketing effectiveness and a practical roadmap for future investment decisions, empowering the client to navigate the complexities of modern marketing with data-driven confidence.

• Data Aggregation Engine : Consolidated over 50 disparate data sources into a single analytical dataset.
• Econometric Model Build : Developed multiple regression models to quantify the impact of all marketing drivers.
• ROI and Contribution Analysis : Decomposed sales into base volume and incremental volume driven by each activity.
• Budget Optimization Simulator : Created an Excel-based tool for scenario planning and optimal budget allocation.
• Strategic Insights Report : Delivered a final presentation with actionable recommendations for the next fiscal year.

Technologies Used

• Data Platform: Python & SQL : We utilized Python, with its powerful Pandas and NumPy libraries, to perform the heavy lifting of data wrangling, cleaning, and transformation across dozens of large, unstructured datasets. This allowed us to automate the process of creating a harmonized, weekly time-series dataset. A PostgreSQL database was implemented as the central repository for this clean data, enabling efficient querying and providing a scalable, single source of truth for the entire marketing mix analysis, ensuring data integrity throughout the project lifecycle.
• Modeling: R & Scikit-learn : For the core statistical modeling, we leveraged the R programming language, renowned for its extensive libraries dedicated to advanced econometrics and time-series analysis. This was crucial for accurately implementing complex modeling components like adstock and saturation effects. Concurrently, Python's Scikit-learn library was used for rapid prototyping, feature engineering, and cross-validation of models. This dual-platform approach combined R's statistical depth with Python's versatility, ensuring the development of robust and highly accurate predictive models that captured the nuances of the client's business.
• Visualization: Power BI : To translate complex model outputs into actionable business intelligence, we developed a suite of interactive dashboards using Power BI. These dashboards went far beyond static charts, allowing users to dynamically explore the results of the marketing mix modeling. Executives could view high-level ROI summaries, while brand managers could drill down into weekly channel performance and competitive trends. The Power BI dashboard also served as the front-end for the budget simulator, making sophisticated scenario planning accessible to non-technical users.
• Cloud Environment: AWS : The entire analytical workload was executed on Amazon Web Services (AWS) to ensure scalability and performance. We used Amazon S3 for durable and cost-effective storage of the large raw and processed datasets. The intensive model training and data processing tasks were run on Amazon EC2 instances, allowing us to provision the necessary computational power on-demand. This cloud-based approach provided the project with the agility to scale resources as needed and accelerated the project timeline by avoiding dependencies on the client's constrained on-premise IT infrastructure.

Request a demo

Results and Impact

Quantzig's marketing mix modeling solution delivered transformative clarity, replacing years of debate and conjecture with data-driven facts. The client could, for the first time, see the precise ROI for every channel and the point of diminishing returns for each dollar spent. This newfound confidence in their data led to the immediate and decisive reallocation of over $30 million from underperforming, high-cost traditional media into high-growth digital and targeted in-store initiatives. The engagement successfully answered the board's pressing call for financial accountability and, more importantly, fundamentally repositioned the marketing department from a cost center into a verifiable and predictable driver of revenue growth. The final report and the accompanying budget simulation tool became the new 'source of truth' for all strategic marketing decisions, embedding an analytics-first mindset into the organization's DNA and resolving the client's core problem of justifying its marketing spend.

Qualitative Impact

• From Guesswork to Data-Driven Daily Execution : The operational rhythm of the marketing team was fundamentally altered. Daily stand-up meetings, which once revolved around subjective discussions of creative concepts, were now anchored by a review of the Power BI dashboards showing near-real-time performance against the model's predictions. The conversation shifted from 'I think we should do this' to 'The model shows this channel is over-saturating; let's test a reallocation.' Budget requests were no longer lengthy prose documents but concise proposals backed by outputs from the budget simulator, showing the expected lift in incremental sales. This change embedded a culture of continuous optimization and made data-driven decision-making a daily habit, not a quarterly review exercise. The insights from the marketing mix modeling for consumer packaged goods became the team's operational playbook.
• Enabling Strategic 'What-If' Planning at the Speed of Business : Previously, critical strategic questions like 'What if a competitor doubles their TV spend?' or 'What is the financial impact of shifting $5M from Q2 to Q3?' were unanswerable, leading to strategic inertia. The budget optimization simulator, delivered as part of the solution, completely changed this dynamic. The leadership team could now independently run dozens of such scenarios in minutes, visualizing the impact on sales, profit, and market share. This capability transformed strategic planning from a static annual event into a dynamic, ongoing process. It enabled them to make bold, data-backed pivots, such as pre-emptively increasing investment in performance marketing to counter a competitive launch and confidently defending the budget for long-term brand-building activities by demonstrating their modeled impact on base sales.
• Building a Culture of Accountability and Cross-Functional Trust : Perhaps the most significant impact was the cultural shift. The marketing mix modeling project created a common language and a single, trusted set of metrics that bridged the historical divide between Marketing, Sales, and Finance. For the first time, Finance could see a clear, quantified link between marketing spend and financial results, leading to increased trust and more collaborative budget discussions. The Sales team began using the model's outputs to understand how national marketing campaigns influenced regional sales and foot traffic, fostering better alignment. This breakdown of organizational silos and the establishment of a shared, data-driven reality fostered a powerful culture of mutual accountability for driving profitable growth.
• Paving the Way for a Sophisticated, Future-Ready Analytics Function : The overwhelming success and tangible financial impact of the marketing mix modeling engagement created massive organizational buy-in for advanced analytics. The project served as a powerful proof-of-concept, demonstrating the immense value locked within their data. The client is now positioned to climb the analytics maturity curve. They are actively planning the next phase of their journey: integrating the top-down MMM framework with bottom-up multi-touch attribution (MTA) models to create a unified measurement system. Furthermore, they are exploring how to incorporate more granular datasets, such as weather patterns, local event calendars, and social media sentiment, to further refine the model and unlock even deeper CPG marketing analytics insights, securing their competitive advantage for the future.

How Quantzig Can Help

Quantzig's success in this engagement is a direct reflection of our profound expertise in marketing analytics, cultivated over 18 years of dedicated experience. Our specific mastery in marketing mix modeling for the CPG industry was the critical factor that enabled us to navigate the client's complex challenges. Unlike generic analytics providers, we possess a deep, ingrained understanding of the nuances of CPG data—from the complexities of syndicated scanner data to the intricacies of trade promotion calendars. This domain knowledge allows us to build econometric models that are not just statistically sound but also commercially relevant. We recognize that a successful MMM solution is not merely a complex algorithm; it's a tool for business transformation. Our approach combines statistical rigor with a relentless focus on translating model outputs into actionable insights and user-friendly simulation tools. This unique blend of technical depth and business acumen is what allows us to consistently address complex problem statements of this nature. We don't just deliver a model; we deliver a clear, data-driven path to improved marketing effectiveness and profitable growth. Our extensive experience has shown us that the ultimate goal of marketing ROI analysis is to empower decision-makers with the confidence to act, a principle that guided every phase of this successful engagement and is the hallmark of Quantzig's approach to optimizing marketing spend for CPG brands.

Quantzig's Expertise in Marketing Mix Modeling

• Deep CPG Domain Knowledge : Our analysts understand the unique dynamics of CPG marketing analytics, including the impact of trade promotions, retailer data complexities, and shopper behavior, ensuring our models accurately reflect real-world market conditions.
• Advanced Econometric Modeling : We move beyond standard regression, employing advanced techniques to precisely measure adstock, diminishing returns, and cross-channel synergies, providing a more accurate and defensible picture of marketing effectiveness.
• Actionable Strategy and Simulation : Our deliverables are not static reports. We provide interactive tools and strategic roadmaps that empower marketing teams to make smarter, data-backed decisions continuously, turning insights into immediate action.