I recently joined the board of QuantFarm a SaaS platform created to harness the power of data with technology and bring data-led innovation to service and products companies. While deep data analytics using machine learning, AI, and Blockchain technologies all sounds very cool –my simple mind required a use case to get on board.
The incredible QuantFarm team (PHD elites) asked me the two biggest problems I encountered when running my previous software companies. Without missing a beat, I blurted out; churn and customer acquisition cost. Here is how churn me loose works.
First let’s define churn. Simply put customer churn is when a customer ceases relationship with a company. Churn is a death blow to companies and prevention is better than a cure. Customer churn reduces profitability through revenue loss. Churn also results in greater marketing and re-acquisition costs. For most companies, the customer acquisition cost (cost of acquiring a new customer) is higher than the cost of retaining an existing customer, sometimes by a factor of 15 times more expensive. Also, the probability of selling to an existing customer is a lot higher than to a new prospect.
A good place to start “churn me loose” is a prediction stage to better understand which customers are going to churn and what actions will have the greatest retention impact on each customer. Churn Analytics helps in reducing customer churn with targeted proactive retention. The first part of the solution a segmentation objective; to identify homogenous groups within customers for determining similar patters of transactions and churn behavior from past data. Segmentation algorithms will work in tandem to segment the customers based on their behaviors (usage, profitability, life time value, complaints etc.) At a bare minimum 20 functions and 20 customers (400 comparisons) will have to be analyzed.
After that we start churn scoring. Churn scoring predicts the possibility of churn of individual customers and the impact of the factors on the churn customers. Churn prediction algorithms create two lists; churn and no churn customers.
Once the lists have been created a prediction churn time is assigned. This survival analysis helps to predict time to churn for a segment of customers, compares time to churn between two or more segments. We then gather data to determine the churn time of customers based on the relationship of co-variables such as, time to churn, price increases, customers dissatisfaction, and/or inferior product quality.
Churn me lose as a service helps;
- · Minimize acquisition costs and increase marketing efficiency
- · Keep customers engaged and loyal over time
- · Decrease the likelihood that competitors will lure existing customers
- · Activate and strengthen the existing customer base
- · See the value of individual customer loss and create targeted strategies
- · Proactively detect customer value loss and take measures
Prevention is better than cure: Customer churn reduces profitability through revenue loss. http://quantfarm.com/