Your support helps keep our site running! We might earn a small referral fee when you purchase from links in this post, at no extra cost to you, which we REALLY appreciate. All opinions remain our own as always.
Improving Services Based On Customer Demographics
Joseph | Broadband Home Services
Improving Services Based On Customer Demographics
Joseph | Broadband Home Services
about this project:
A broadband company in the United Kingdom was hoping to discover common factors among customers cancelling services. By understanding the frequent reasons WHY customers were ending their contracts, they could put efforts into place to improve their services. This would effectively reduce customer cancellations while providing better service.
Results
Upon analysis, we identified 6 different ‘core’ types of customers. (Often referred to as customer avatars). Two specific avatars were identified as high-risk for ending contracts early. Among these avatars, the pricing of services was a critical factor for cancellation.
This information allowed the company to better tailor their pricing for each specific avatar and greatly improve customer retention.
%
Unique Customer Avatars
Problem Avatars Identified
CHALLENGES
THE CLIENT NEEDED A TEAM THAT COULD WORK WITHIN THEIR LARGE EXCEL DATASET TO DISCOVER WHICH CUSTOMERS WERE AT RISK FOR CANCELLING SERVICES EARLY AND WHY.
The company had a 2TB+ data set compiling their customer history with detailed customer demographics.
They knew this information identify which customers were cancelling services before their contract timeline ended, and hoped to use that information to improve their service options. This would then reduce the number of customers who cancelled their services prematurely.
Unfortunately, Excel is not the easiest software for examining and identifying commonalities without data-analysis trained staff.
The company reached out to OWLLytics to analyze the data and put it into software that could be easily managed by their staff.
CUSTOMER LOSS/ CUSTOMER CHURN ANALYSIS
SOLUTION
OWLLYTICS BUILT A MACHINE-LEARNING MODEL THAT COULD ACCURATELY IDENTIFY MATCHING ATTRIBUTES AMONG THOSE CLIENTS MOST AT RISK FOR CANCELLING SERVICES, PROVIDING A BETTER UNDERSTANDING OF THEIR CUSTOMER NEEDS.
OWLLytics was able to import the data into much stronger software for a much faster analysis.
We used a machine-learning model to identify similar attributes among customers, including those which cancelled services early.
➡ Using this data, we identified six distinct customer avatars, mainly grouped by demographics and services purchased.
➡ We pinpointed specific customer avatars.
➡ Additionally, we identified correlations among the demographics to explore common causes for early cancellation.
➡ Lastly, we provided this information to the company along with suggestions for updated pricing strategies and areas of improvement.
RESULTS
THE BROADBAND COMPANY WAS ABLE TO RESTRUCTURE THEIR PRICING TO BETTER FIT THEIR CUSTOMERS’ NEEDS, ACCORDING TO WHICH CUSTOMER AVATAR THEY BELONGED TO, GREATLY IMPROVING CUSTOMER SATISFACTION.
There were notable differences in the demographic groups of each of the six specific customer avatars, with the most important factor identified as the age of the customer.
We were able to pinpoint the largest level of customer cancellations as being among the elderly demographic.
Upon discovering this, the broadband company was able to review their services and pricing to identify:
➡ Retirement age customers needed a less costly service. Income typically became more restricted upon getting older and they didn’t need as many service features.
➡ Offering a 30-day contract created a steady loss in revenue. It cost the company more money to gain a customer than what the customer returned in revenue (among any avatar).
➡ Family-based customers did not find value in paying for added ‘protection services.’ This group also tended to fall into either limited-service packages or value-added bundles, avoiding mid-range packages.
In general, the broadband company gained a much larger clarity of their customer demographics and their service needs. This allowed them to create more specific and targeted offers to each customer group.
Likewise, it resulted in consistently happier customers who retained their services longer.