With aggressive growth plans, a $1 billion private equity-backed software company needed to understand its best short-term opportunities to increase revenue from current customers. Years of acquisitions, however, created an environment of fragmented customer and product data. Building a growth strategy backed by real data was next to impossible—until we came equipped with a proprietary machine learning platform and capabilities designed specifically for this type of analysis.
Our team worked with the software company and its C-suite to:
The software company now has insight that will help it build value in a way that others cannot. Better yet, we delivered that insight in just weeks.
cross-selling opportunity in the first two years – including $788 million in year one
upgrade potential from existing customers
churn risk identified, allowing the company to mitigate
The software company has plans to grow revenue to $1.5 billion, and its new CEO needed to quickly introduce growth strategies to deliver on that goal. The CEO wanted data-backed analysis to answer key questions:
But before the company could answer these questions, it had to address underlying data issues. The organization has grown rapidly through mergers and acquisitions, and its five business units did not share data. This meant the company did not have consolidated customer and product data that it could use to understand the potential for cross-selling and predict customer behavior – keys to future growth.
That’s when we came in.
The software company’s CEO worked with West Monroe in a previous role and has seen, first hand, the impact of our unique approach – powered by Intellio™ Predict, a proprietary advanced analytics platform that uses machine learning to evaluate revenue streams and predict growth potential with precision.
We rapidly assembled a multidisciplinary team with software industry experience as well as data management and data science expertise.
Our team first made sure the client’s data could support analysis. We imported 28 million rows of customer and product data from 16 Salesforce, Siebel, and Oracle data tables. One of our key tasks was “de-duping” account data using ML-DQ and creating a universal identification approach that would allow account-level analytics.
Then, the team used our proprietary platform to analyze data and develop precise, quantified answers to the company’s questions. We used a credit score-like approach to value customers – analyzing 30 million combinations of customer and product data points to produce 7.74 million scores for 70,000 customers. This included 96 cross-selling scores per customer. Each customer also received a score indicating likelihood of upgrading in a product family, as well as a score reflecting risk of churn. Our machine learning platform identifies important features in the likelihood particular behaviors and then finds optimal splits in the data to separate high versus low probability of upgrade and churn.
Because our platform includes established predictive models and tools, we were able to complete this analysis in a matter of weeks.
The software company now has insight that will allow it to build value in ways that others cannot. It can see cross-selling white space exists and develop strategies to capture it. Our analysis showed that the best cross-selling opportunities exist in three of the five business units, and 20% of customer/product family combinations are responsible for 81% of potential cross-selling value. The company can now give its salespeople specific lists of who to call and what to sell. And, it can establish clear accountability for developing growth targets – something it could not do before.
The company understands where to act to maximize upgrade potential and reduce churn. For example, 91% of churn risk resides in 20% of customer/product families – allowing it to focus proactive retention efforts more precisely.