“If there is Big Data, is there Little Data?” – Philosoraptor

Ok, so there is no such thing as a velociraptor that is also a philosopher and is called a philosoraptor. That’s just an Internet meme that has been going viral for some time (a meme is a concept that spreads rapidly from person to person, usually through social media and other Internet forums like Reddit or 4chan, and it is distinguished by its creativity). However, big data is a real concept that’s disrupting businesses everywhere, the life and health insurance industries included.

But what is big data? How do we as advisors or businesses use it? Will it make your life easier, as most technologies claim to do? And how can you get started?

We consulted with experts in this matter, who are also completely immersed in the insurance industry, to answer these and other questions. Here is their advice on big data.

business report

1) What is big data?

According to Jeffrey J. Hogan, Northeast regional manager of Rogers Benefit Group in Connecticut, “big data is the common denominator of our value proposition. Big data is any and all information that will help a plan sponsor determine how to mitigate risks on his/her plan. This includes traditional reporting from employers and more. We also typically include preventive usage reporting as part of our big data dump,” Hogan explained in a brief interview with LifeHealthPro.com.

Meanwhile, another industry expert defined big data as a “term adopted by marketers and organizations that encompasses the idea that client information can be used to refine their own marketing or sales buying habits. Big data analytics will show and prove that certain demographics buy or respond to marketing differently than others, and allow an organization not to waste efforts on those that statistically will not convert,” said Nic West, director at PinneyInsurance.com, a national insurance brokerage that specializes in servicing the needs of the tech savvy, next generation agent.

And while this is all very exciting for some, others are mildly skeptical about the transformative power of this technology. In a recent press release, Strategy Meets Action (SMA), an insurance strategic advisory firm, released two new research reports on big data, “Big Data in Insurance: Insurer Priorities, Projects, and Investment Plans for 2014 and Beyond“ and “Big Data in Insurance: Emerging Trends in Organization and Technology.”

The lead author of one of the reports, Mark Breading, says that big data may be “simultaneously over-hyped and underestimated.” It won’t change the insurance world overnight, “but the potential is enormous and benefits are already being reaped,” said Breading, also an SMA partner.

big data

2) How to use big data?

In West’s case, he uses big data analytics to market and sell. “Blast marketing is a thing of the past. Now, it is all about finding the message that fits exactly the interests of the recipient by using data analytics from previous attempts to refine the message and audience,” he said.

The claim for more efficiency in all industries is being driven in part by new technology, and big data is definitely behind it. According to Hogan, the Patient  Protection and Affordable Care Act (PPACA) has created incentives for employers who sponsor health plans to figure out how to use data to coordinate and manage their health plans effectively. “Intelligent agents, brokers and consultants now are partnering directly with the C-Suite to teach them both how to manage risk and how to strategically design their plans and incentives to change behavior among members. You can’t manage or even accept risk unless you have data to inform your every move,” explained Hogan in an interview with LifeHealthPro.com.He says that big data will emancipate and revolutionize the health care marketplace.

And there are other ways that companies are using big data. More than pulling reports related to their groups, the health care industry, for example, uses it for gathering information such as detailed utilization information including prescription usage by tier, preventive usage, claims stratification, premium and claims by month, etc., according to Hogan.

The goal of big data is to make the industry more efficient, lowering risk, costs and providing better access to information, while also fueling innovation. “Suddenly, big data is driving a demand for the best, most efficient, nimblest payers and providers. It’s so incredibly exciting! The health care marketplace is de-centralizing and the various players including plan sponsors, payers and providers are taking on and managing their own risk,” Hogan said. 

And it seems that more than just providing consumer problems, big data is also creating a demand for better, more informed and more competent agents and brokers, he added. “Those that are stuck in the mud will — and are — losing out. Big data creates tremendous opportunity for strategists to become trusted advisors,” the benefits expert cautioned.

With 25 percent of insurers investing in big data projects in 2014, a dramatic increase from 9 percent in 2012, according to SMA’s report, it’s no surprise that you’ll be hearing a lot more about it.

And, big data is also used for customer acquisition, which is why more than two-thirds of insurers are either using or planning to use big data by 2016.

business intelligence

3) Will it really make life easier?

Ok, so this question is a little bit harder to answer. After all, what is really an “easier life” if you have to navigate through tons of information and reports?

West offered an example of how to use it in marketing and new client acquisition: “Using big data analytics to refine marketing and sales processes will absolutely increase sales and revenue and will provide a tailored message to the recipient that they will likely share and refer others to. Once set up, the campaigns can be running and you or your sales team will only be dealing with SQLs (sales qualified leads).”

Meanwhile, on the health insurance spectrum, Hogan described that insurers can or are already using it to determine risk, or what he calls “disease process data mining on every claim.” He says, for example, that insurers looking to set their insureds path to wellness will include an analysis of their preventive usage data. That analysis might include questions such as: What percentage of the group’s population is getting their annual physical exam, their baseline cytological exam, their pap or mammography, their colorectal screening, their child screening?  

“Clearly if people aren’t getting these things done they likely don’t have a primary care physician who is responsible for the coordination of their care. They likely end up being a catastrophic claim in the long run. They may have a disease process that they are trying to ignore. They aren’t being managed by a medical professional. Coordinated care is key for employers who take on risk,” Hogan explained.

cloud

4) How to get started and how does it work?

As with any technology, there’s always a learning curve and lots of trial and error. In fact, some are still developing applications or cloud-based programs to handle the large amount of information that needs to be processed.

According to SMA, many technology platforms will be used, but cloud-based solutions are being considered by 68 percent of insurers under $1 billion and by 38 percent of those over $1 billion in written premiums. However, Denise Garth, SMA partner and lead author one of the SMA reports said that, “While big data adoption is increasing, organizational and technology challenges will hold insurers back and place insurers at risk in the competition for big data talent and the capabilities to leverage emerging technologies that are creating massive new data sources.”

Hogan explained that in order for advisors to being using big data for their businesses, they have to become a project planner who analyzes data, conveys it to their company and then makes incremental recommendations or plans for acting upon the data. Hogan’s firm specializes in selecting agents, brokers and consultants who have adopted these practices or techniques and helps them empower and support these relationships by receiving and analyzing the data.

On the other hand, West explained that he got started by “looking at my sales trends, product sales and demographics like most others, in Excel.” He recommends that you start small, find one or two trends in your book of business and send out messages specific to that audience. Then, based on the responses, refine those messages.

West also provided a very specific approach to get started. In his words:

  • Excel. I would download EVERY field in your database into Excel and start sorting and filtering. 
  • I would look at per 5 year agent brackets, per $25k income levels, gender, number in household etc. to see who is buying what products. 
  • I would take that data and create marketing campaigns to re-engage or sell additional products to those that fit the criteria but don’t own. For example, if you see in your data that a working male under age 50 that makes over $75k is statistically a disability insurance owner (based on your past sales), I would query your database to find all those clients that fit that mold that do NOT own disability insurance and put them into an email marketing campaign to start garnering interest.
  • Also, a CRM is a necessary requirement of big data analytics and advanced marketing and sales. You need a CRM to store and track data and it should allow you to export data for future automated marketing use. We use www.dataraptor.com as it allows for the automation of both marketing and sales, all using big data to spot trends.

See also:

How advisors can (and should) use Big Data as a differentiator

Survey: U.S. insurers gearing up for growth, tech advancement

The 2014 100 best sales & marketing ideas