How a Major Personal Hygiene Brand Optimized Its Media Mix and Increased Total Media Revenue by 5.5%

The Challenge: Measuring Marketing Effectiveness Across a Complex Product Portfolio 

MASS Analytics collaborated with a major personal hygiene brand to measure the effectiveness of its marketing activities and improve its Marketing Return on Investment (MROI). The engagement presented a distinct structural challenge: the client was following a product substitution strategy, which meant all related metrics had to be grouped according to that strategic logic rather than standard product categories. 

Modeling was conducted at a product group level and across a number of different regions, adding further complexity to the data architecture. The client needed answers to several layered business questions: the total impact of marketing on sales; the most effective media channel on both an aggregate and a detailed level; the effect of cannibalization within its portfolio; the impact of competitors’ strategy; and the impact of COVID-19. Above all, the brand wanted to identify the optimal allocation of its budget across media channels. 

The Proposed Approach: Agile Workflow, Pooled Modeling, and Granular Analysis 

The team adopted an agile workflow that began with data preparation. Using MassFeeds—specifically the embedded template of Nielsen—data preparation was automated. Three years’ worth of monthly data was prepared and uploaded to MassTer, MASS Analytics’ marketing mix modeling platform. 

The first modeling challenge was data scarcity. To increase the number of fitted points and ensure more reliable estimates, the team applied a pooled modeling technique. This approach allowed the model to draw strength across regions and product groups rather than relying on any single thin data slice. 

To answer the client’s multilayered questions, the modeling results were analyzed on both a granular and an aggregate level. Digital media, for example, was examined holistically and then broken down by individual touchpoints such as Facebook and Instagram. Television was similarly analyzed both as a total channel and by spot length. This dual-layer analysis ensured that strategic recommendations could be drilled down into specific tactical executions. 

The Optimization Phase: From Response Curves to the Optimal Execution Range 

The final step of the engagement was optimization. Its purpose was to evaluate the client’s new strategy and uncover the optimal media mix, given that in recent years the budget had been weighted mainly toward digital media. 

To advise the client on how to allocate budget optimally, the team needed to measure two things for each channel: 

The output of this analysis was the Optimal Execution Range: a recommended interval, derived from each channel’s saturation characteristics, from which the client could isolate its optimal budget. 

Step 1: Building the Response Curve 

The starting point was the creation of the response curve, which represents how much revenue is generated at each budget level. In its raw form, this curve does not give a clear indication of what budget is optimal. For example, the analysis showed that the current budget allocated to TV was approximately 280k and sitting near the flat part of the curve—suggesting it was close to saturation—but the response curve alone did not provide a precise measure of how saturated the channel truly was. 

Step 2: Creating the Profit Curve 

The next step was to create the profit curve by subtracting spend from the revenue generated at each budget level. After reaching its peak—the Max Profit point—the curve begins to drop. It eventually reaches zero and falls below it (negative profit) at the point where spend exceeds revenue. The Max Profit point marks the start of the Optimal Execution Range. 

Step 3: Defining the Saturation Point with the Cumulative Profit Curve 

The third step was to create a cumulative profit curve, which is the rolling average of the profit curve. The saturation point is located where this cumulative curve crosses the max profit curve. This intersection defines the endpoint of the Optimal Execution Range. 

The percentage saturation of a media channel is then calculated by dividing the current budget by the value at the end of the Optimal Execution Range. In the TV example, the current spend of 280k yielded a **percentage saturation of 96%**—confirming that the channel was operating extremely close to its saturation threshold and that marginal returns were minimal. 

Results and Insights: A 5.49% Increase in Total Media Revenue 

The process described above was performed for both TV and digital media. The Optimal Execution Range for each medium was delivered to the client as a first read on where each channel saturates. 

At a later stage, these curves were used simultaneously to provide a comprehensive perspective on how to reallocate budget efficiently and increase ROI. By shifting spend away from saturated channels and toward channels with stronger marginal returns, the team was able to increase the total revenue generated from media by 5.49%. 

The optimization recommended specific budget shifts across the media mix: 

• TV: Budget was reduced by 20%, with a corresponding revenue impact of only -1%, confirming the channel was heavily saturated. 

• Social Media: Budget increased by 30%, generating a 47% increase in revenue. 

• YouTube: Budget increased by 20%, generating a 27% increase in revenue. 

• Google: Budget increased by 25%, generating a 10% increase in revenue. 

• Programmatic Display: Budget increased by 10%, generating a 37% increase in revenue. 

The result was a more efficient media mix that redirected investment from low-marginal-return TV spend into higher-performing digital touchpoints, unlocking net incremental revenue across the total portfolio without increasing the overall media budget. 

Why Optimization Requires More Than Modeling Alone 

This case illustrates that marketing mix modeling delivers its full value only when it is extended into optimization. Building a robust model answers the question of what happened and why. Optimization answers the question of what to do next. 

By combining pooled modeling to overcome data scarcity, granular analysis to capture touchpoint-level dynamics, and a rigorous three-step optimization methodology, the personal hygiene brand moved from descriptive measurement to prescriptive action. The identification of saturation points and the construction of an Optimal Execution Range turned abstract coefficients into concrete budget boundaries that the marketing team could act on immediately. 

For brands managing complex portfolios across multiple regions, the lesson is clear: media mix optimization is not merely about spending more on what appears to work. It is about understanding precisely where each channel begins to lose efficiency and reallocating within those boundaries to maximize total portfolio return.