The Marketing Mix Problem
Optimizing the Marketing Mix enables companies to determine the optimal spending level and the budget allocations along any dimension of interest (customer segments or sub-segments, campaigns, channels, media, products, time, frequency, etc.). This optimizes the impact of marketing resources and enhances a company’s chances of connecting and engaging with the ideal buyer.
Typical Issues in Marketing Mix Analysis
- Company Sophistication: Marketing mix analysis is a complex problem and in most companies, such decisions are driven by historical performance, gut feel, or what the competition is doing. An increasing number of companies now are turning to data and analytics based decision making. However, in order to derive full benefit from this capability, companies need to attain a certain level of maturity in people, processes and technology platforms. According to Forrester Research, there are 4 levels of maturity for a company wishing to optimize its marketing mix: Guesstimators, Advice Seekers, Spreadsheet Junkies, and Predictive Analyzers. The primary difference between these 4 levels is the availability of data and the sophistication with which companies use data to optimize marketing mix.
- Multiple Objectives: Even if a company is in the Predictive Analyzer stage where it builds sophisticated models to analyze marketing mix, the marketing mix optimization problem itself is fairly complex. Multiple objectives might exist in a typical marketing mix problem – typically revenue, customer satisfaction, user engagement, trials, foot traffic, etc. Usually it is not possible to optimize all objectives at the same time, and trade-off decisions need to be made
- Optimization vs Simulation: Many vendors in the market offer a simulator for marketing mix and do not actually optimize an outcome (even though it might be labeled as “optimization”). While a simulator has its uses, at some point, true optimization is needed
Management Foresight Solution
Management Foresight works with companies of all maturity levels. Our approach is simple – if a company is not sophisticated in its people, processes and tools to do full-fledged marketing mix optimization, we focus on smaller solvable problems like a single brand, market, or customer segment to demonstrate success.
Management Foresight uses a powerful proprietary approach for marketing mix optimization that uses the Gamma AnalyticsTM AI Suite:
- Companies can evaluate how to invest across their product portfolio across multiple markets
- We perform true multi-objective optimization that yields the pareto-optimal results for multiple objectives (as opposed to using linear combinations of objectives like some vendors do). So you can solve for various goals like revenue, customer satisfaction, user engagement, trials, etc
- Although we perform true optimization, we believe that once you have a few optimal scenarios, the best decisions around marketing mix are taken by humans with deep contextual market knowledge. It is possible that the optimal answer recommended by analytical approaches may not be the one selected for implementation in a market. Through our “Governance” techniques, we facilitate an objective decision making process that takes all aspects into account, including mathematically optimal answers, as well as marketer experience and judgment
- We provide thorough “what-if” and other sensitivity analyses that generate multiple scenarios to simulate the impact of market situations, competitive threats, and changes in investment/resources. This approach is more deliberate and sophisticated compared with simulators commonly available in the market
Governance
We at Management Foresight believe that you can have the best tools at your disposal but they may be useless unless you make business decisions based on the information. To derive the best benefits of marketing mix modeling and optimization, we design and set up a governance process, that enforces regular disciplined evaluation and decision-making based on all available metrics, data and information.