Auto insurance company reduces overstaffing by 20%
A North American auto insurance company with more than 2 million customers needed to improve the customer experience while reducing operational costs. They wanted a faster, automated solution to optimize customer request forecasting and staffing. Using Pecan, the team rapidly generates, implements, monitors, and refines dozens of models that predict customer call volume over short- and long-term intervals, as well as for specific regions and different types of service. Pecan’s predictions are used in their day-to-day operations to better allocate resources and build staff schedules.
Company Size: Over 2 million customers and about $150M in revenue
Solution: Forecast customer requests to optimize resource allocation and staffing
Platform Use Case: Demand forecasting involves trying to determine the likely future need for an item, based on historical data and analytics showing how much of it has…
Data Stack: Microsoft SQL Server
reduction in customer wait time
and 1 year forecasts
What Our Customer Says About Pecan
Manual forecasting affected customer service quality and resource allocation
Before adopting Pecan, one team member created forecasts manually for customer request call volumes. Each set of forecasts required a week of work and, as a whole, only informed planning for only about 60% of the call volume. The team also wanted to reduce overstaffing and other operational costs.
The customer hoped to find a scalable solution that would optimize resource allocation across all its locations to improve service everywhere for its 2M customers.
It was essential to find a partner that offered a flexible yet automated approach to ingest hundreds of millions of rows of data from various data sources, refine models easily based on evolving combinations of variables, and implement top-performing models seamlessly into their workflows.
Predict customer assistance requests in advance
With Pecan, the team developed and implemented over two dozen models into production that generate short- and long-term forecasts. The short-term forecasts predict customer request volume and service type for 3-hour periods within 2 weeks. These forecasts are specific to hundreds of micro-regions in this company’s service area. These highly granular predictions ensure that staff and resources are correctly placed to best meet customer needs.
Additionally, the team retrains and runs another In the context of machine learning, a model is a specific instance or example of an algorithm that has been created based on a particular… quarterly to generate predictions for customer request volume a year in advance.
Pecan empowers the data science team with easy experimentation and data enrichment
The team has continuously experimented with and improved upon the predictive models using Pecan, creating hundreds of models over time — a fast and nearly effortless process with the platform. The team has also enriched their in-house data with external data, such as weather information, which has improved the models’ performance.
Equipped with these forecasts, the company uses business logic to translate predicted call volume into optimized staff and resource allocation.
Improved service boosts customer satisfaction and retention
A customer needing assistance now receives quicker service and more straightforward resolution, thanks to Pecan’s forecasts and the Data science combines statistics, computer science, scientific methods, and business knowledge to analyze, model, and predict using data. The data science toolkit can be used… team’s integration of predictions into their business systems. Of course, they won’t know that Pecan’s forecasts played a role in their service, but they’ll be more likely to maintain their coverage with this company — and to spread the word about their positive experiences.