The Marketing Analytics Food Chain – What You Should Know and Why

The Marketing Analytics Food Chain – What You Should Know and Why

Marketing Analytics Maturity Phases

In -a recent post, Apollo founder Bryan Mason discussed the data needed to support advanced marketing analytics methods. In this series of posts, I will be discussing how to use this potentially enormous data stream across the four phases of the analytics food chain, some of the most popular machine learning tools used in performing advanced analytics and some of the specific use cases for advanced analytics that are used to understand marketing effectiveness for both companies and advertising agencies.  For each use case, I will describe the marketing questions you can expect to answer and the appropriate marketing analytics methods that can be applied.

Marketing analytics solutions can be divided into four categories: descriptive, diagnostic, predictive and prescriptive. The chart below provides a nice summary of each category and the incremental value added each level provides.

Descriptive analytics generally answers the question “What happened?”. This is what we most often see as output from clients for their marketing campaigns and activities in the form of charts, reports and/or dashboards. The best reporting schema provide updated information in real time so that stakeholders can interpret the results and make decisions on what to do next. A good dashboard may even have some drill-down capabilities that will yield additional insights. However, descriptive analytics doesn’t address the issue of why a campaign or marketing activity performed as it did.

Diagnostic analytics is meant to address the question of “Why did this happen?”.  Often, diagnostic analysis is referred to as root cause analysis. Statisticians and data scientists mine the historical data to find correlations between the action desired (called the dependent or target variable) and all of the factors that influence that action (independent variables). Models are built to find patterns in the data that determine the variation in the target variable that is “explained” by the variation in the independent variables thought to influence the target. In advertising, for example, these models can be used to find out which channels or tactics drove sales during a particular campaign. Examples of other questions that can be answered through diagnostic analytics include “how do we avoid repeating execution errors in a campaign?” or “what do we need to focus on to repeat a successful campaign?”.

Predictive analytics takes things a step further by building models off of the historical data to answer the question “What will happen in the future?”. These models use something called machine learning techniques, bringing together a number of data mining methodologies, forecasting methods, predictive models and analytical techniques to analyze current data, assess risk and opportunities, and capture relationships and make predictions about the future. In marketing, one use case is that of marketing mix modeling, where each marketing activity’s impact on sales can be measured and then used to build a forecast simulator that predicts changes in future sales based on differing “mixes” of marketing spend across channels. Predictive analytics is becoming more attractive to companies and ad agencies now that technology has been developed to analyze big data and by embedding predictive analytics techniques in more advanced machine learning tools that non-data scientists can make use of. While this seems to be more efficient, in many cases it can be dangerous in that it requires a level of understanding in explaining how the statistics lead to the conclusions that non-statisticians just do not have

Prescriptive analytics builds upon the other three steps in the analytics chain as well as applying business rules to infer actions to influence future desired outcomes. Businesses can assess a range of prescribed actions and their possible outcomes, factoring in the risks associated with each action. This is a relatively new branch of analytics that is not yet commonly used but is gaining in interest among larger organizations, particularly in the Consumer Package Goods (CPG) space.

Where are you in your marketing analytics journey? Are you getting real, actionable insights out of your data? Apollo consultants are here to help you use marketing analytics to maximize your company’s success.


Ed Holmquist

Apollo Consulting Group, Providence, RI

No Comments

Post A Comment

We do not sell or share your personal information.