High-Tech Commentary: From Paper to Digits at a Small Surplus Lines Carrier
How did one little company in Hatboro, Penn., persuade 66 of the top wholesale insurance agencies in the U.S. to place about one-third of their business there? The short answer is that this company, Penn-America, provides a unique franchise value to general agents underwriting excess and surplus lines products for small businesses in small cities and towns through some 30,000 retail brokers. The long answer is the story of what it takes to create that franchise value for busy, successful entrepreneurs: relative market exclusivity, access to senior management and technology that is focused on their needs, not the carrier’s.
When I joined Penn-America in 1987, I was the IT Department. The team I soon started to build focused on the back-office functions of a paper-based insurance company. The halls of our building were lined, literally, with ceiling-high shelves of folders. Today, with 17 people providing IT services to our staff and general agents—and hardly a file folder to be found—we’ve come a long way from the Agents Advisory Council meeting in 1994 when a few attendees tentatively asked us, “Can you accept data from us … electronically?”
We immediately conducted a survey of our agents that included just three questions: Do you use a computer for account currents? Can you save output to a diskette instead of paper? Would you like to earn $500 for your agency? Shortly thereafter, our first “electronic data transfer” began and soon we were proud of the fact that 21 percent of our agents were mailing diskettes to us once a month.
From the perspective of 2004, when we exchange data with our agents in virtually every medium, from paper through images to XML, it’s amazing to remember that as recently as 1994, “Do you use a computer?” was a serious question. In 1995, we were exploring how to use the new medium called e-mail, and in 1996 we published our first Web site. The years between 1994 and 1997 were like the Wild West for small insurance carriers’ IT departments. There weren’t many of us out there. The environment changed constantly. There was danger around every corner and the electric charge of opportunity was in the air. Despite the tumult of the times, we took on the challenge of assuring that our still-growing systems were Y2K compliant and even installed our first high-speed Internet connection.
I speak with authority because, at many points along the way, we panned for gold in the wrong data streams. But we’ve learned hard and crucial lessons about the process required to deliver on the promise of technology:
Listen to the agents. They know what they need. Recognize that they need different things and that there are many “right” ways to accomplish a task. Provide a range of tools that make it easy to accomplish the task.
Use multi-disciplinary teams to understand the problems and to invent the solutions. Keep listening. Change course when it’s clear you’ve taken a wrong turn.
Keep sharpening your tools. Between our first forays into using technology to support our general agency network and today, we learned and then applied these lessons in four distinct phases. From the Wild West (1994-1997) we moved into building a basic infrastructure for a small community of pioneers (1997-1998), recruited new settlers to help us expand the community (1999-2000) and methodically built what has become a thriving metropolis—with plans to expand and upgrade our highways (2001-2004).
In 1997, we launched the first of several comprehensive “listening” programs that shaped the development of our agent-centered technology infrastructure. We hand-picked 50 employees at every level of the company and formed them into teams. Each team was assigned a different “customer” group, such as employees, investors and general agents. Each team had a goal: listen to the assigned audience and find ways to make our relationship better and more efficient—and launch a solution within six months.
The agent-focused team identified more than 40 forms and reports that would benefit our agents if they were available electronically. Within two months, we launched our first extranet for general agents to provide those 40 documents. (Some still had only one Internet-connected computer using dial-up connections.) Initially, the extranet’s entrance was buried deep in our Web site and the contents were limited to replacing paper forms with electronic versions. But within a year, our still un-named extranet was easier to find and included some rudimentary claims reporting functions, all of our news publications and the capability for agents to update their contact information.
Even the limited capabilities of this first extranet proved popular and useful. So, in March 1999, we began in earnest to recruit more agents to use the system. We assembled another set of interdisciplinary teams in Project 2000, but this time they were more focused on the “means, methods and motivators” that would attract new users. By the summer, the extranet was re-launched as PennLink. The first PennLink was positively loaded compared with its predecessor and included an electronic agency manual, policy applications, e-mail links to our staff, online claims reporting, claims status, production status reports and other features.
During this period, we were much like evangelists with our general agents considering that many of them did not have e-mail addresses or Web sites. So, during this period we also began to offer IT assistance to general agents who needed a jump-start onto the information highway. In 1999, we designed, published and hosted the first three of which ultimately became 20 Web sites we designed and hosted for our general agents.
From March 2002 until today, our community of users and capabilities have continued to mature and grow. We didn’t always get it right and when we didn’t, it was always because we provided something we thought our general agents needed or wanted rather than asking them.
The insurance markets in our specialized niche were red hot from 2001 to 2003. Much of our IT attention was devoted to building increasingly sophisticated internal systems that allowed us to handle the spike in volume we experienced and to provide timelier, in-depth management reporting so that we could keep a close eye on underwriting profitability. But by late 2003, with our internal systems smoothly handling the high volume, we began to share the fruits of our labor with the general agents. PennLink was redesigned in 2003 to include a completely Web-based underwriting manual, a tabbed system to simplify access and the ability to drill down, slice and dice each general agency’s data in many different ways including, for example, industry segment analysis capabilities.
Today we have a myriad of methods for exchanging information with general agents. We’ve written code that allows our system to “grab” files from general agents’ systems once they sign-on to our servers and authorize it. Other general agents have more sophisticated systems of their own that automatically generate Adobe Acrobat and XML data files and e-mail them to us. Others use our penn-policies Web site to log-on to our servers directly and generate XML policies. Some send us image files and a few still use “snail mail” to send their policies.
Policies aren’t the only kinds of information exchanged between general agencies and carriers. Applications, endorsements, claims and inspection reports are all ripe for conversion. Every piece of paper we eliminate in a transaction between us and a general agent saves money. An image file is better and cheaper to process than a paper file. Exchanging pure XML data is more efficient than exchanging images. So, we constantly encourage and help general agents to systematically move up the scale of data sophistication. Many of them have learned quickly that the sooner the information is converted into electronic data, the faster it can be transformed into information they can use to manage their businesses profitably. That’s the ultimate aim of “franchise value” in any enterprise.
J. Ransley “Randy” Lennon is vice president of Information Technology of Penn-America Insurance Company, and its subsidiary, Penn-Star Insurance Company. He is responsible for the internal computerized processing systems and proprietary Internet services to Penn-America’s general agents.