Customer Success Analytics
Unlocking insights from customer complaints using Numbers Station
A leading B2C retail company was facing a significant challenge with a recent increase in customer churn rates. The company had a large volume of customer complaints but lacked the resources to analyze the data beyond manual annotation. The company deployed the Numbers Station Data Intelligence Suite to categorize customer complaints. Using Numbers Station's Data Transformation Assistant, data analysts with no AI expertise developed custom complaint categorization models. They were able to identify that a large number of complaints were due to delivery issues in New York State. The analyst team shared these insights with the delivery operations team, who was able to address the issue and ultimately improve customer satisfaction and reduce churn.
Problem
A leading B2C retail company was facing a significant challenge with recent increase in customer churn rates. Most of their customer analytics had been based on structured data that was amenable to analysis (transaction and customer databases). However, existing analyses based on structured data could not explain the sudden spike in churn rates. The company also had a large volume of customer complaints which was highly unstructured data but lacked the resources to analyze the data beyond manual inspection. The company was seeking a more scalable solution to understand the root cause of customer dissatisfaction and reduce churn rates.
Goal
Categorizing customer complaints to understand the root cause of customer churn.
Challenge
The company's main challenge was the lack of resources to analyze the unstructured complaint data. The team in charge of analyzing that data was mostly composed of data analysts with no ML or AI expertise. They tried to manually analyze the complaints but quickly realized that they would need a more scalable solution to derive any meaningful insights.
Solution
The company turned to Numbers Station' Data Transformation Assistant to make sense of their complaint data. Using Numbers Station AI transforms, analysts with no technical expertise were able to build categorization models on top of their ServiceNow complaint tickets. Ultimately, they were able to identify patterns and trends in customer complaints, and found serious delivery issues in New York state.
Results
With this information, the company was able to work with the delivery service to fix the issue and improve their delivery process, resulting in a reduction in customer churn rates and an improvement in customer satisfaction.
More customer service use cases
Use Numbers Station to connect ticketing systems with revenue data and derive ticket prioritization strategies to improve overall customer satisfaction.
Transform open text fields in product feedback surveys into actionable insights. Understand what features your customers like the most or the least to drive your product roadmap.
Derive insights from support call transcripts, understand how agents are doing using sentiment analysis. Communicate feedback to provide high-quality customer service.