Welcome to part 4 of a 6-part article series on knowledge, the Buyer Journey, marketing/sales synergy, and better sales results. Our last article Which data sources to integrate to gain from Big Knowledge looked at the primary data sources you need to focus on and the value of each, especially the CRM system. This article looks at how machine learning can help produce more precise knowledge from the data sources you have used.
Digital marketing, in general, is very quickly turning what used to be a guessing game into an uninterrupted series of data-driven actions. Effective marketing has become a blend of science, strategy, and creativity. There’s much to know about new ways to drive business with customer data. Machine learning is, for example, a key part of how Luxid client, Visma Software, was able to find 602 new leads and predict 722 customer businesses at risk of churn. (More on that later.)
How does Machine Learning work?
Machine learning is easy to understand. It’s an application of artificial intelligence (AI) that finds patterns in data. It can, for example, predict buyer behavior with a high degree of accuracy. Algorithms that use machine learning automatically learn from experience and improve. Machine learning is a simple thing in theory: you put your data through an algorithm, and it finds patterns. This differs from traditional analytics, where you typically start with an assumption about the data, then select the appropriate analysis method.
A large quantity of data is key to the effective use of machine learning. There is no limit to how much data you can feed into a machine-learning process. With a large enough data set and enough parameters, machine learning will find linkages between data that a human would miss.
You still need a data scientist to combine your data sets and extract the relevant information. Then you need them to help customize the machine learning algorithms, as they are so complicated that you simply can’t build a model by hand. When building from scratch, Python and R are the leading programming language contenders for data science, and both have machine-learning algorithm libraries. We often use RapidMiner for processing data and building machine learning programs.
Uniting data sets for Machine Learning
Many machine learning models are non-linear, which means they can explore the data in greater depth than their linear counterparts, producing more insightful results. With machine learning, raw customer data is fed into an unsupervised learning algorithm that separates contacts by pre-defined actions. In the case of lead scoring, customer behavioral data is fed into a supervised learning algorithm that calculates the likelihood of purchase with a high degree of accuracy.
As with most data analysis, the biggest challenge with machine learning lies in uniting and normalizing data before it’s fed into an algorithm. For high-quality results, you need to start with high-quality data. First, you have to look at variables in the data sets and discard those with many missing values. Then join different data sets together into one large one. This is where time often plays a role. For example, the format of information from a marketing automation platform like Oracle Eloqua is collected over an entire year, while firmographic information is from one point in time. So you need to summarize the activity data in Eloqua to follow the same format as the firmographics data.
Meet, sort, and test
Suppose data sets are significantly different, for example, richer data for one subset of customers. In that case, you need to consider if that’s going to have a negative effect on the machine learning algorithm. If it starts to rely on data that doesn’t exist for other customers or prospects, the results will be skewed.
You must ensure that the data collection process is free of critical flaws. You essentially need to have a good understanding of an organization’s business so that you can be sure that it is correctly reflected in the data. That’s where data governance plays a part in building a solid machine-learning program. The best way to know how data is collected and stored is by meeting with customers and understanding their processes.
Once you have united your data, you should test how well the machine learning model performs to indicate your data quality. For example, split your data set 70/30 and do A/B testing. Use 70 as ‘old data’ and 30 good.
Machine learning in action: Visma Software
The Visma Group of Companies has 900,000 customers in 12 countries and services in five distinct business areas. One business unit, Visma Software, wanted to take its digital marketing to the next level, which involved machine learning. We had two objectives: to analyze and predict the probability of both churn and buying; and to understand the factors, processes, and variables that influence churn & buying probability the most—to prevent churn or Accelerate buying.
Visma Software had plenty of customer data to use. We gathered data from five systems, controlled 70 variables, and analyzed 168,000 data points across 2,722 customer businesses for churn and an additional 6,444 potential purchasers. We combined sales and support data sets enriched with marketing activity data, such as email activity data, website and customer Portal activity data, and firmographic information
In the end, we achieved Prediction model accuracy of 84.6% for churn and 84.1% for sales—very high when it comes to predictive marketing. A big part of this success was the availability of multiple years of activity data. The marketing team identified 602 new leads and 722 customer businesses at risk of churn. The results also provided new insights into the processes and mechanisms at Visma that affect churn and buying. The stimulation and mitigation that will follow will positively impact the bottom line for years to come.