STARK: Customer retention through Churn Prediction

STARK is a nationwide hardware store and lumber yard that caters to both private and professional customers. STARK has physical stores all around the country and a large online shop with a wide range of products. 

In close collaboration with STARK, Delegate has developed a churn tool that, using AI, can assess whether a customer is about to leave. With the new solution, STARK has activated all the data they already had and gained a strong weapon in the fight to retain and win customers.

All research shows that it is easier to retain customers than to get new ones. Therefore, it is crucial for STARK to be able to predict when a customer is about to leave, in order to do everything to retain the customer. 

"Our strong focus on retention is about spotting customers who are acting less for various reasons, in order to proactively accommodate these customers. Satisfied customers are the goal, and there is of course also a lot of economics involved in retaining more customers." 

–  Jørn Fogh, Sales Excellence Director at STARK 

The first step in retaining a customer is to discover that they are about to leave. This prediction is also called Churn prediction. 

"Churn covers a declining or terminated engagement among customers or members. It can be customers who switch to a competitor. A customer who suddenly buys significantly less; a customer who stops subscribing to your services or a member who leaves. Churn Prediction is about predicting these events." 

- Erik David Johnson, Principal AI Specialist at Delegate 

Churn prediction requires large amounts of data and intelligent models to analyze these. STARK already had historical data on customer behavior, and combined with knowledge and expertise from Delegate, an AI model has been fed with this data and trained to assess the risk of churn among customers. 

But an AI model is not better than we make it. The preparation of data for training the AI model behind STARK's churn prediction used over 900 unique parameters. These are, of course, parameters that have been selected in collaboration between STARK and Delegate based on close collaboration and a very early business clarification with STARK's management, business experts, and data responsible. 

How does STARK utilize the new knowledge in practice? 

Insight must be operationalized to be valuable. With the new AI-based churn solution, STARK gets an updated risk assessment on all customers every week. This allows them to target their retention initiatives towards customers with the highest churn risk - or the opportunity to capture customers who show an increasing risk over time. 

STARK is using the churn solution directly in their Dynamics 365 solution. If a customer is at high risk in the churn analysis, tasks will be created for the responsible sales employee. In this way, STARK can directly contact their customers to retain them with offers or simply give them attention with a relevant dialogue. 

STARK can also use churn risk data as input for their marketing initiatives. In practice, they can look at churn risk and its development over time, and use it to segment their communication to customers through emails and ads. All follow-up processes are set up in the system, and there is no employee who needs to keep track of the analyses in a static model. The AI model automatically initiates a marketing process when a customer reaches a critical percentage. 

"The ability to create an automated process targeted at customers who do not come to our stores as often will contribute positively to customers' experiences of being seen and, hopefully, also our sales." 

–  Jørn Fogh, Sales Excellence Director at STARK 

With the help of Delegate, STARK has activated all the data they already had and gained a strong weapon in the fight to retain and win customers. They quickly translate the large amount of data into behavioral patterns. In this way, they can predict when a customer is, for example, about to move their business to one of STARK's competitors. 

The value of the individual customer is quite significant. All research shows that it is both easier and more profitable to hold on to your customers than to have to go out and find new ones. STARK has therefore set all sails to succeed in minimizing churn. Together with Delegate, STARK has embarked on a data journey. Here, it is intelligent models and artificial intelligence that have put customer retention in system and automated the processes around churn. 

This is a completely new and effective method that saves STARK's employees time, which also results in customers feeling seen. STARK uses their new churn solution directly in Dynamics 365, and every week STARK automatically receives an updated risk assessment of all customers. 

"The collaboration with Delegate on the development of our churn prediction solution has been extremely constructive. Throughout the process, Delegate worked hard to communicate, involve and explain the otherwise fairly complex material being worked on. We met specialists who really understand numbers - and who also worked hard to explain the material to us as customers. Throughout the project, we had a clear alignment and status on the project, and the delivery was also professionally delivered." 

- Flemming Malm, Senior Manager Project Management at STARK"

"The ability to create an automated process targeted at customers who do not come to our stores as often will contribute positively to customers' experiences of being seen and, hopefully, also our sales."

Jørn Fogh, Sales Excellence Director at STARK

Would you like to be contacted?
Fill out the form

We would be more than happy to help your organization. Fill out the form and we will contact you as soon as possible.

Morten Fritsch, Director

mfr@delegate.dk
+45 53 53 69 59

Latest Cases

ISS: An Optimal Workplace Using IoT Data

Delegate har i tæt samarbejde med Microsoft og ISS skabt en IoT Data platform som ved hjælp af IoT-teknologi giver indblik i faktorer som påvirker komforten på arbejdspladsen.

Start your next project with us

en_GBEnglish