As viewed from the executive suite, agency technology can look complicated, confusing and expensive. In many agencies it can also be quite disappointing, requiring a fair amount of manual processing to create even the most basic reports.
Outside of our agency walls, we live in a world of easy access to even the most arcane information. Want to settle a debate over whether Alexander Hamilton was a United States president? Just click into the magical world of Google and you can settle that argument in less than 10 seconds.
The answer lies in quantitative information, a fact available through a simple data lookup. The magic is that you can type (or speak) only his name into your smartphone and receive a result showing the aggregated answer from multiple sources. With a quick scan of the results, you’d find that Hamilton, in fact, was not a president but the first secretary of the U.S. Treasury.
Let’s translate this example to your agency and refine the question: Is Alexander Hamilton Inc. a customer of our agency? For most agencies, finding the answer to that simple question is not a simple proposition. Manually searching for “Alexander Hamilton” in the p-c management system, the employee benefits system, the surety system and the CRM system may not do the job.
If your agency systems don’t have a usable search function, you might have to pore over customer lists while accounting for potential misspellings, creative naming conventions, archaic codes and the like. This manual processing of unmanaged data makes it difficult to produce any answer with certainty.
Q: Is Alexander Hamilton Inc. a customer of our agency?
A: Based on three hours of analysis, I’m pretty sure the answer is no.
It doesn’t have to be this way. While the complexity of individual lines of business and the limited choice of available agency-management systems force us to spread our data among multiple sources, agencies can regain control over their data.
Agencies can regain control over their data through a data warehouse - a database that receives all of the key data from your different management and transactional systemsTweet
Simply put, a data warehouse is a database designed and maintained by the agency that receives all of the key data from your different management and transactional systems. This central database can then be used to report and eventually provide analytics across the entire organization. By pulling all of your agency data together, you can create an environment that allows for across-the-board reporting. Unfortunately, although the concept of a data warehouse has been around for decades, our industry has yet to fully adopt it.
Yet motivated agencies can go further down the techno-rabbit hole and build a query-capable dashboard. Type in an insured’s name and see an overview of the customer showing current coverages across all lines of business, endorsement and claims activity, upcoming renewal actions, cross-selling opportunities and relevant accounting information.
Want more? Tie the dashboard into external sources of information like SEC filings, a real-time Web search on the customer or a feed from the press release section of the customer’s website.
Did the customer just announce a significant expansion? You can read all about it.
And there’s more. Now that you have a complete picture of the customer, why not provide access to this dashboard to the customer? Your systems have compiled a unique view of your customer’s risk management, unavailable through any other source. Share the love.
What Is a Data Warehouse? This part gets technical, but don’t stop reading—we won’t go into the weeds here. From the perspective of your technology team, a data warehouse encompasses fun terms like ETL, star schema and data federation. From the perspective of agency leadership, a data warehouse represents a big project, a healthy price tag and probably some outside help.
Most agency IT teams have a strong handle on computers, servers and networks. Very few have existing resources with a deep background on data structures and manipulation. It’s a different animal. Get some help on this one.
Like all technology projects, a data warehouse will take a phased approach. Before you start, you have to know where you want to go. Defining the end first allows your team to clearly identify what data points to collect from your various systems and to more accurately design the warehouse database structure. Once the warehouse is designed, the heavy lifting begins. The sole purpose of the data warehouse is to aggregate data from systems that were never designed to work together. Each originating system will require a different approach to pull the data out, normalize it, and inject it into the warehouse. This process will take time, so prepare to be patient. The results will reward the effort.
It’s a common mistake to use the terms “data warehouse” and “data analytics” interchangeably. While a data warehouse allows for the easy retrieval of quantitative information (how much premium do we have placed with Chubb?), it does not provide qualitative information (based on current placement trends, how can I restructure a program to provide a greater return?). Data analytics are the next level up. You will want to go there, but do it one step at a time. Without a data warehouse, your agency would struggle to make any sense of trending data. Start with the warehouse, but keep an eye on the horizon.