Customer Information Data Management Systems | Pitney Bowes
AAA reduced the time it took to create models from 5 months to just 6 weeks.
The Auto Club Group’s Advanced Analytics team now efficiently analyzes data and gains rapid insight with the help of Spectrum Self Service and Spectrum Miner™.
Federation of auto clubs with 50 million members
The ACG is AAA’s second largest club
6 million customers in the southern region
Help team respond quicker to requests
Produce more models in a shorter time
Analyze volumes of customer data to improve business
Increased predictive model accuracy
Saved 1.5 hours per project & 126 hours per month
Boosted productivity to 13 projects per employee per month
“The ad hoc requests for data analysis prevented us from developing the deeper statistical analyses and models needed. At the same time, the processes were taking too long. We required tools that would give the business units direct data access. This freed up my team to model faster.” Kristin Rahn, Director,
AAA Advanced Analytics,
The Auto Club Group
The Auto Club Group’s Southern Region—serving Florida, Georgia, and Western Tennessee—offers roadside assistance, travel planning, insurance and other financial services to more than 6 million customers. With nine business units (including member acquisition, renewal, travel, road service calls, rewards, market data, insurance, other product sales and public relations), the non-profit member service organization analyzes volumes of customer data to inform business strategy for improved operations.
Each business unit within the organization relied on the Advanced Analytics team for data access and analysis, as well as for statistical model development. The company required easy-to-use technology to help the modeling team to respond more efficiently to inquiries and also needed a solution that would allow many of these requests to become self-service.
Company-wide, The Auto Club Group Southern Region was challenged to double sales targets in order to meet revenue goals. The Advance Analytics team which supports the internal business units with predictive models and ad-hoc analytics, acknowledged however that using the traditional statistical tools the turn-around time could be up to six months to complete just one modeling project.
The advanced analytics team needed a way to shorten the modeling cycle (including data preparation and transformation, accounting for missing values and outliers, ensuring data availability, exploratory data analysis and model development) while at the same time increasing their analytic output. The pressure on the analytics team was two-fold:
1. Produce more models in a much shorter period of time
2. Free up analytic time by enabling the business units to answer some analytic questions for themselves
The company selected Spectrum Miner™ to simplify the modeling process for the analytics team and speed up the handling and output of data. Spectrum Miner™ enables the Advanced Analytics team to create one big aggregate table including all of the relevant data inputs, and to use Spectrum Miner™ on this table to create models more efficiently and more accurately.
The company also selected Portrait Self Service Analytics to provide business units with the ability to service their own data requests. Using Portrait Self Service Analytics, the team prepares nine analytic tables per month which are readily accessible to scores of people within each business unit to review for their own business insights. Accounting, for example, is able to evaluate cost per road service call by area and membership to determine profitability for road service providers.
“Implementing Pitney Bowes Software’s solutions was a double success. We resolved 75 percent of incoming business unit requests for data analysis and reduced the time it took to create models from five months or more to just six weeks.” Kristin Rahn, Director,
AAA Advanced Analytics,
The Auto Club Group
The Auto Club Group’s Advanced Analytics team now efficiently analyzes data and gains rapid insight, increasing predictive model accuracy and reducing modeling time from five months or more to just six weeks.
The team saved approximately 1.5 hours per project resulting in about 126 hours saved each month with employees averaging 13 projects per person per month.
150 employees now have access to Portrait Self Service Analytics, enabling them to extract and analyze data specific to their business unit.