Marketing mix modeling identifies millions in savings
Use Case
Marketing mix modeling
With over a billion dollars budgeted for marketing, this company has far more options than most for reaching American consumers. But with great budgetary power comes great responsibility. An economic slowdown motivated this marketing organization to find a budget allocation across 20+ channels to improve operating margins and efficiency without reducing revenue.
Pecan’s marketing mix modelingA statistical approach used to quantify the impact of various elements of a company’s marketing strategy on sales and other key performance indicators (KPIs). MMM… (MMM) offered the perfect solution for understanding and optimizing their marketing spending. MMM results now guide better decision-making, increase marketing efficiency, and save money while returning even better outcomes.
Second largest U.S. company in its industry
Company Size: Nearly 20 million customers, 40K+ employees, and a marketing budget of over $1B
Solution: Optimizing marketing spending
Platform Use Case: Marketing Mix Modeling (MMM)
Data Stack: AWS
$1B+
marketing budget managed with Pecan’s MMM
$100M+
overspending identified in national channels, such as linear broadcasting and radio
$200M+
potential savings through reduction or reallocation to more efficient state channels — with improved inbound marketing results
$200M+
in linear broadcast ad spending better attributed and allocated
Challenge
Determining the ideal usage of an extensive omnichannel marketing budget
With a marketing budget exceeding $1B, this company could pursue any and all marketing efforts it desired. But which were worth the investment? That determination was especially difficult for the roughly 30% of the budget allocated for linear TV and radio advertising, where obtaining reliable data is challenging. The company also had a relatively small amount of digital marketing data from ad platforms.
Additionally, the existing impression, click, and attributionThe process of identifying and assigning credit to the various marketing touchpoints that contributed to a conversion or another business outcome, such as a sale… data didn’t provide sufficient analytical details to inform planning, especially since typical attribution methods reveal only what happened in the past. With business performance faltering, the marketing analyticsAnalytics is a business practice that uses descriptive and visualization techniques to gain insight into data; those insights can then be used to guide business… team needed to boost marketing efficiency and make the best possible use of their budget.
Solution
Predicting customer inquiries based on budget allocation
The marketing analytics team was most interested in understanding and increasing inbound inquiries from potential customers. Therefore, Pecan’s experts collaborated with the team on constructing a marketing mix modeling (MMM) solution to predict the number of inquiries based on ad spending.
The modelIn the context of machine learning, a model is a specific instance or example of an algorithm that has been created based on a particular… used about 4 years of data on marketing spending by month, channel, and U.S. state, as well as the number of inquiries in each month. The finished model provides both national- and state-level predictions for the next 12 months, given different levels of spending. Highly accurate predictions for these different time periods inform both short- and long-term budget planning and strategy.
Simulating the impact of potential budget changes with MMM
In addition to predictions, a key feature of MMM is the power to simulate different marketing budget allocations and see likely outcomes. With a click, marketers can examine the potential results of increasing or decreasing state- and national-level spending, as well as the effects of moving spend among channels.
In addition, the Pecan platform’s optimizationBroadly speaking, optimization is a process used to either maximize or minimize an output value by selecting the right input values. In data science, this… capabilities calculate the amount of spend by channel, month, and state, to maximize inquiries while working within customer-defined constraints and objectives. For example, if the objective is to lower the national budget by 5% and reallocate spend to more efficient state budgets, the optimization algorithmIn the data context, an algorithm is a set of instructions for a computer to bring in input data, manipulate it, perform calculations with it,… will automatically determine the optimum ways to allocate the spend. This task would be time-consuming and tedious for a human. However, the efficient, robust Pecan MMM algorithm can calculate ideal allocations in seconds for a variety of scenarios, informing rapid adaptation to changing circumstances.
Giving MMM models nuance with more variables
MMM also reveals the effects of other variables on marketing outcomes. Pecan’s MMM accounts for the lasting effects of marketing (known as adstockThe accumulated effects of past marketing activities on current sales. Adstock is a measure of how much of the current sales are driven by past… or carryoverThe effect that a marketing campaign has on sales after the campaign has ended. This can be positive or negative and can be caused by…). This variable incorporates the ongoing impact of this company’s brand equity, which would continue to generate value even if the company stopped advertising entirely. For this company, the model also incorporates external factors, such as publicly available sales data on relevant products that directly affect inbound inquiries.
Results
Understanding and optimizing marketing spend
With the MMM model in hand, the marketing team has far more insight into how spending in different channels drove customer inquiries. MMM shows each channel’s impact, including channels that don’t provide reliable data on audience responses. That includes the linear broadcast ads that were a key channel for this company. Additionally, the team could use Pecan’s simulation and optimization tools to test different approaches’ impact.
Armed with this information, the team found they could actually boost the number of customer inquiries, while decreasing their marketing spend. This lift in results — while saving money — resulted directly from allocation adjustments informed by MMM. This optimization provides impressive cost-cutting opportunities of over $100M yearly for the company.
While the MMM solution is already proving its worth, it could have an even greater impact in the future. Pecan’s experts are exploring options for refining the model, which may include not just predicting customer inquiries, but also optimizing marketing spend based on both customers’ predicted lifetime value (pLTVCustomer predicted lifetime value (pLTV) is the total amount of revenue a business can predict will be received from a specific customer over the entire…) and cost per acquisition (CPA).
For this company and others, MMM provides extensive predictive insights to inform future-focused, data-driven planning. Even in difficult market conditions, MMM helps marketers make the most of every dollar in their budgets — and make the most business impact.
Contents
It’s time to refine your outreach with customer foresight
Latest Resources
10 Data Breakthroughs That Solve Business Problems
Are you curious about your enterprise and how you can improve your business? Meet our latest Data ScienceData science combines statistics, computer science, scientific methods, and business knowledge to analyze, model, and predict using data. The data science toolkit can be used... breakthroughs.
Predictive Analytics: Getting Actual Value from Machine Learning
In this webinar, we shared the essential components of predictive analyticsPredictive analytics uses data, statistics, and machine learning techniques to build mathematical models that can generate predictions about things likely to happen in the future.... to get value in the shortest time possible.
Stop User Churn in 3 Steps with Predictions
In a constant battle for user time, developers need to keep their users engaged and happy while monitoring reduced engagement and churn rates.