What Does Data-Driven Insurance Mean to Insurance Agencies?

There is a lot of talk about the increasing importance of big data in business, especially in the insurance industry. We see articles about “data-informed decision making” and “siloed data” and “gathering the right data”, not to mention the many regulations regarding protecting customer data. It seems that data-driven business represents the key to our prosperous future and perhaps even our survival.
I'd like to step back from all of that and think through how to use data and analytics to drive growth in an insurance agency in a pragmatic way, in real time. 

In the first place, insurance agencies have always been data-driven. We are consumed, to some degree, with loss ratios, production numbers, growth rates, and all kinds of other data. Our lives are already full of data. Do we really need any more data? Or do we need a different context for thinking about it?

Size matters

Working with senior executives of Fortune 500 companies as well as entrepreneurial insurance agency owners, I see both using similar data sources to analyze results and make decisions. 

I also see a difference in style: the fortune 500 Executive is loath to make a decision which the data does not clearly support. Given a hard choice between a gut feeling or data-based option, that executive is inclined to rely on an answer being provided by the data. On the contrary, the entrepreneur tends to use data as a precursor to intuition, as a risk reduction device, and as a support for gut-level decision making.

"Both kinds of executives use data to make decisions, but how they use it within their business models is fundamentally different."

Another difference in the way Fortune 500, big-company thinking styles use data, compared to insurance agency owners, is that the large company uses data for extensive planning and then for comparison of results to plan. The agency owner tends to use data more for spotting opportunities, and for pivoting and changing plans based on those opportunities.

There really is a fundamental difference in the use of data, and neither one of them is necessarily correct. Both methods have their merit.

Using data to sustain growth

As agencies grow larger, they tend to adopt more of the large business approach to the use of data. They do this because they're making bigger bets and the cost of failure is consequently greater.

The change in the way that agents use data and think about it tends to accelerate as the agency passes $1 million in revenue. At that point in an agency's development, not only are the stakes for decisions bigger, but the complexity of the business itself is now at a whole new level, and the challenges to continued growth increase. This is where the solid strategic use of data becomes helpful because the planning process for the agency becomes more important.

Crossing the $1M line

It's relatively easy to grow a business in the early years, but as complexity and size increase, maintaining a consistent rate of growth becomes a bigger challenge. This challenge requires better and more sophisticated planning. This planning, in turn, requires a better use of data: it requires increasingly making decisions based on data, rather than hunches.

This crossover point is a significant one for the insurance agency owner, regardless of whether they are in health insurance, auto insurance, or any other specialty; because it represents the need to change from entrepreneur gunslinger to entrepreneur data-driven decision-maker. 

As the business owner transits this period in their development, they are faced with two challenges: 

• The first, as I said earlier, is that the bets they’re making are bigger, and the consequences for getting them wrong are also more painful. 
• In addition to that, the agency owner is increasingly insulated from the market realities due to the demands they are under in growing the business. Typically, at this point, the owner is no longer primarily a producer, they are also Sales Manager, Business Manager, and a variety of other roles, which means divided focus and decreased connection with the feel of the business that they once had. Good data, well captured and competently analyzed, can take the place of this feel in the decision making process of the owner. 

So increased size simply demands better systems: better data systems, and better decision systems. Those better decision-making systems require an increased capability on the part of the people making the decisions, an improved process for making the decisions, and also much better data on which to base them.

As the agency continues to grow, it should recognize that future growth in a more complex environment with an increasing rate of change demands even better use of data and even more sophisticated data-informed decision making.

The right clients

For example, data around the cost of writing and servicing a given type of account will increasingly drive decisions about who you will accept as a client, it doesn’t matter if you’re selling car insurance or protection against tropical storms. In earlier days, the agency just needed revenue and cash flow. Now, it needs clients who create revenue greater than the cost the client creates. 

Increasingly, in order to grow revenue and profits over time, the agency must become the buyer, not just the seller of insurance. By this I mean, the agency has to increasingly define, both from a financial loss market segment and other bases, who is a right client for the agency; and based on data, reject those who do not fit - otherwise the agency's growth and its profitability will be negatively impacted.

These are all data-based decisions, or should be.

As the agency seeks to make those decisions, it needs to understand in an increasingly sophisticated way what the relative costs are for human involvement in the various parts of account service and customer experience, versus the benefits and cost-saving of digital insurance and automation.

The best price

The analysis of that data must drive not only its account selection criteria, but also its operational process decisions. In fact, as agencies progress, particularly in commercial lines, their continued profitability increasingly demands the use of complex software systems that automatically select the right markets for the accounts it wants to write, assuring not only price competitiveness, but book balance, loss ratio maximization and E&O minimization.

As we go forward into the 2020s, commercial lines clients are increasingly demanding evidence that their agents are delivering their assurance to them at the lowest possible cost for products and services. Price transparency is an increasingly relevant issue for agencies. Coronavirus and the economic travail it is creating will do nothing but accelerate and exacerbate that, and the scrutiny from social media will only become more intense.

As a traveler's insurance company regional vice president said to me recently, “Price is table stakes.” The problem for commercial lines agencies is the cost of preparing multiple submissions and proposals every year for every client can quickly eat up agency profits. So, how does one deal with that?

The right carriers

In order to maintain an acceptable level of profit, insurance agencies are increasingly realizing that they must select insurance carriers for representation for business placement based on the cost of doing business with that carrier. This requires that the agent not just take the carrier's word for it, but actually analyze their own experience with pass-through rates, speed to issue, and ease of use of systems as primary decision criteria around carrier representation along with market competitiveness, commission payments, and profit-sharing agreements.

Agencies eagerly await comparative rating systems for commercial insurance. Within the next 36 months, those systems will be widely adopted by agencies seeking to lower their costs and maintain their forward momentum and growth. There are already systems available in early stages, and companies like Tarmika are rapidly making grounds using machine learning, artificial intelligence, and industry experience to develop systems that promised a lower cost increase speed for agents. 

Agencies must constantly analyze their cost data and production results in order to make decisions about which systems to implement.

We’re not yet in a future of connected cars and AI, but our very human intelligence can use the available data to make better decisions.