The term “disruption” is frequently used in the insurance industry. It usually refers to digital disruptors such as insurtech, digital-only attackers, or e-commerce players entering the market and changing the way insurance products and services are delivered. However, insurance carriers can themselves win by using personalized marketing.
Not only can personalized marketing help insurers stay competitive against digital disruptors but it also offers a financial incentive. For example, personal auto insurance carriers in the United States could tap an additional $2 billion if they could retain just 10 percent of the $19 billion in direct premium written switches from one carrier to another every year; in the European Union, that would represent an additional €900 million, assuming a 10 percent retention increase of the €9 billion in gross direct premiums written switches.
Personalization—or reaching customers with targeted messaging, offers, and pricing at just the right time—is the future of insurance marketing. Marketers were once limited to a handful of undifferentiated, periodic marketing campaigns. But today a wealth of customer data, analytical tools, and marketing technology allow companies to run hundreds of personalized campaigns continuously to improve acquisition, cross-selling, and marketing return on investment. While personalized marketing is nascent in insurance, in other industries, we have seen the practice cut acquisition costs by as much as 50 percent, stimulate growth in revenues and customer satisfaction by 5 to 10 percent, and increase marketing returns by a factor of between five and ten.
Other industries, such as retail, have led the charge in using personalized marketing to drive growth and reduce their cost of acquisition. Leading retailers create a 360-degree view of their customers—where they shop (online or in-store), what they shop for, when they shop, how much they spend, what they view or click on—and target them with highly personalized offers based on that data. These retailers use personalization engines powered by machine learning to improve e-commerce websites and their marketing campaigns across channels. When a user searches for a product, the company can provide not only the search results based on the keyword but also personalized search results based on the user’s profile, actions on the website, and even contextual clues such as the weather at the user’s location.
This type of personalization is also possible in insurance. Imagine an auto insurance customer—we will call her Dianna—has been a longtime policyholder with her auto insurance company. But after a negative experience last year when filing a claim, she is considering shopping around. She has an older car with high mileage and a son nearing driving age. Dianna’s auto insurer knows all this. The insurer also knows that US auto insurance customers tend to start shopping for rates 60 days before their policies expire—and that one in three shoppers will switch carriers.
The carrier also knows that Dianna’s unsatisfactory service experience puts her account at risk. Taking the initiative, the insurer sends Dianna a letter at the 60-day mark. It is addressed specifically to her, acknowledges the negative prior experience, and provides a dedicated account manager and telephone number Dianna can use to expedite future queries. The letter also offers a discount for renewing early.
Often, the clues to a customer’s unmet needs are hidden in plain sight. A customer with an auto policy may be inclined to add renters insurance, for instance, if the insurer reaches them at the right moment—such as when a customer calls to change their billing address. Insurers can use these signals to trigger tailored outreach.
The advent of technology-enabled marketing personalization comes at an important moment for the insurance industry. The past ten years have seen an arms race among US property and casualty (P&C) carriers, with the top five now spending upward of a combined $4 billion on marketing annually, according to S&P Global Market Intelligence and Kantar Stradegy. Personalization programs allow insurers of any size to achieve greater returns and expand their business for much lower levels of investment than nonpersonal, mass-marketing approaches.
Stretching the marketing dollar
An effective personalized marketing program does not have to start with a multibillion-dollar investment; many of the key insights businesses need come from data they already have but don’t use.
For example, a leading pay-TV company wanted to convert more customers into regular viewers of on-demand content. So, using a recommendation algorithm, it piloted a multichannel campaign with messaging tailored to individual customers. Customers who simply browsed were invited to download one of the shows they had previewed. Those who viewed at least one program were invited to download a similar show. Marketers also provided customer-service representatives with a script, and when “browsers” called with a question about their bill or service, the agent wrapped up the call by offering customers the opportunity to download some on-demand movies for free. By tracking response data, as well as how customers were accessing content (for example, via laptop, tablet, or mobile), the company adjusted its messaging and recommendations over a period of weeks, resulting in a more than 10 percent lift in product engagement.
Creating an effective personalized marketing program
Of course, not every personalized marketing attempt is successful. Done well, personalized approaches create sticky relationships and new opportunities. Done poorly, they become blast messages that fill a customer’s spam folder and have little impact. Through our experience across industries, we have identified four core elements to an effective personalization program: data foundation, decisioning, design, and distribution (Exhibit 1).
Data foundation: Define the journeys that matter most
Leaders in personalized marketing begin by establishing which customer journeys have the greatest potential value. For one insurer, it could be auto customers who are moving to a new home; for another, it could be prospective teenage drivers. Making this determination requires understanding the most important customer journeys and prioritizing them by dollar value. In the retail context, that journey is often checkout; for telecom, it’s the subscription journey. An auto insurer may find greater lifetime value generation by engaging first-time car buyers than customers who are trading in their old vehicle and buying a new one. Knowing the biggest unmet needs in the customer decision-making journey provides the insurer with the tools to most effectively intervene with timely messaging and offers.
Decisioning: Identify customer signals and respond with the right triggers
Traditional point-in-time campaigns run, for example, every Monday at a preset time of day. But today, insurers can analyze customer behavioral data for signals that mark the ripest opportunities and then design appropriate responses—known as “triggers” because they are intended to prompt a customer action—over the course of the campaign. For an insurance customer, a signal of renewal risk may be, as illustrated in our opening example, a customer who has had a poor claims experience. The trigger may be a well-timed message, 60 days before policy expiration, acknowledging the issue and offering a discount for renewing early. In our experience, repeating and reinforcing a series of triggers can lead to a 20 to 40 percent lift in outcomes (for example, making a purchase, requesting a quote, filling out a form) over more traditional, episodic point-in-time campaigns.
To get the most from the approach, insurers need to track both the business impact and the desired behavioral changes; in other words, they must track not just open and click-through rates but also the number of resulting quote requests. Insurers should experiment with different types of messaging, refine the content, and measure the results again until they find what works best. In our experience, four to five trigger iterations per customer are usually enough to capture 80 percent of potential value.
Design: Empower a small group of the right people
Successful personalized marketing, of course, requires testing and adjustments. To start, companies may set a goal of releasing one to two new test campaigns per week—aiming to work up to one or two live campaigns per day across various channels. This testing and expansion process must be more than the side task of one or two individuals; instead, insurers should create small, interdisciplinary, agile teams of six to ten individuals—of whom one must be a senior executive with the management skills and institutional authority to remove roadblocks, streamline the approval process, and rally the organization. Other team members should possess a mix of analytics, content creation, marketing, and product experience and be empowered to dedicate themselves full time to the effort. The groups will work in innovative ways, designing new processes and reconfiguring existing ones to expedite repeated actions, such as legal and compliance reviews, as well as data access and integration.
One major bank was surprised to learn that putting just three talented analytics people in a room with a campaign manager, a project manager, and a small content team led to more than a dozen personalized campaigns over a three-month period. In our experience, working in this way can reduce cycle time for testing from six months to just a few weeks.
Distribution: Embed the new capabilities (technology, measurement, process)
While marketing analytics and automation tools are important, investing in new technologies without first determining what works and what doesn’t tends to automate underperforming processes. For example, investing in a cloud-based automated marketing platform to send out one email campaign every quarter would not fully harness the platform’s capabilities. But if an insurer combined the platform with a new process in which it employed campaign logic, it could send out several automated, personalized email campaigns each week.
The right business processes are what enable effective, technology-enabled personalized marketing programs. Insurers should test scenarios using available data and analytics and uncover the processes that work best for their organization. As capabilities mature, insurers can then tap a variety of marketing automation tools, experimenting until they find the vendors and solutions that best serve their needs.
Personalized marketing in action
To illustrate the four core elements of a personalization effort, let’s return to Dianna, the insurance customer introduced at the start of this article. In addition to nearing her auto renewal and grappling with a disappointing claims experience, Dianna is moving from Los Angeles, California, to Syracuse, New York, for a job promotion.
Since more than 40 percent of personal insurance carrier switching in the United States takes place at life events such as moving homes, Dianna’s carrier has focused its personalization efforts on customers in this position. It has identified the signals—such as tracking a visit to a real-estate website via cookies—that indicate Dianna is moving, which prompts a personalized campaign to keep her as an auto insurance customer or sell her a homeowners policy. The campaign could consist of a number of outreach methods depending on Dianna’s actions, including tailored social-media ads and timed emails with customized auto or home insurance offers in Syracuse (Exhibit 2).
Personalized marketing can help insurers increase retention of expiring policies and cross-sell new policies in the face of increasing competition from digital disruptors. Most insurers have the starting ingredients: capable analytics and marketing teams, rich customer data, and multiple channels to reach their customers. With thoughtful practices, dedicated teams, and committed leadership, insurers can realize significant impact in just 12 to 16 weeks.