March 27, 2019

Introduction to Predictive Marketing Analytics

Matti Airas

Lead Business Consultant, Marketing Science Team, Customer Experience Management, Tieto

Predictive Marketing is the process, tools and rules for applying predictive analytics to making marketing and sales decisions. The objective of Predictive Marketing is to anticipate which marketing and sales actions will most likely lead to the desired customer behaviour and to carry out those actions.

Predictive Marketing is not a replacement for more traditional marketing approaches. Marketing and sales decisions will continue to be based mostly on product releases, inspirational ideas, what competitors and peers are doing, and what has worked well in the past. What Predictive Marketing will do is supplement traditional marketing with a new, more analytical method of sorting and prioritising marketing and sales actions.

One of the key aspects of Predictive Marketing is Predictive Marketing Analytics (PMA), which involves using historical customer data to predict future outcomes and trends. For example, PMA can tell you which accounts to target for churn prevention and which leads are most likely to become customers and are, thus, most worth pursuing. PMA is typically and predominantly done using computer algorithms (e.g., Machine Learning).

Graphic 1: Predictive Marketing Analytics Process
Graphic 1: Predictive Marketing Analytics Process

During the last five years, the majority of B2C companies have started utilising Predictive Marketing Analytics. But PMA is really nothing new. The individual methods and formulas have been around for almost fifty years. So what is behind its recent growth in popularity? The explanation is that three critical enablers have simultaneously matured to the point at which PMA has become easily accessible to almost any company.

  1. Marketing Clouds and CRM-centric sales generate a massive amount of behavioural data. These include sales interactions, digital customer engagement, social media, loyalty, support, etc.
  2. Data extraction has become easier and cheaper. Just a few years ago, it took weeks or even months to extract, transform, and load data for analytics. Modern martech applications make data extraction simple and easy to automate using data integration tools.
  3. Computing costs have plummeted. Machine learning requires a massive amount of computing power. This used to be very expensive. With cloud computing more or less following Moore’s law, the cost of storing and processing data has fallen drastically.  

Predictive Marketing Analytics isn’t an absolute science. Like traditional marketing, it is some combination of art, intuition, and science. But it does provide companies with the ability to more reliably forecast customer behaviour.

We will continue with this topic in April when Matti Airas introduces how predictive marketing adds value to different business processes.

Interested in hearing more? Contact us for a free consultation and let's see how Predictive Marketing Analytics can work for you.

The blog was originally posted on idBBN.com.

Stay up-to-date

Get all the latest blogs sent you now!