The company called it Autograph. Rather than price insurance according to traditional, easily measured risk factors such as a driver’s gender, age, and driving record and vehicle make and model, Progressive was experimenting with using global positioning systems (GPS) and wireless technology to record the actual use of policyholders’ cars including times during and conditions under which they were driven. Progressive was committed to technological leadership in an industry that was growing increasingly excited about employing technologies such as GPS and cellular modems for calculating policy premiums.
The company prided itself on running cutting-edge experiments such as McMillan’s data-heavy program that could become the basis for customized, differentiated services. Customers were responding to Progressive’s eighteen-month Texas pilot, which was drawing to an end, with enthusiasm. Autograph systems had been installed in 1,100 cars and users had saved on average about 25 percent. “I use some cars intermittently, and when I do, I pay,” explained one Houston resident with four cars insured by Progressive-3 The positive response to the pilot had led Progressive to consider rolling the program out nationally.
But ome wondered whether conditions specific to Texas”a preponderance of rural and suburban driving, which were less expensive to 1 Marcia Stepanek, “Q&A with Progressive’s Peter Lewis,” BusinessWeek, September 12, 2000. 2 Anne Eisenberg, “What’s Next: paying for Car Insurance by the Mile,” The New York Times, April 20, 2000, 67. 3 Eisenberg (2000). Professor Frances X. Frei and Research Associate Hanna Rodriguez-Farrar prepared this case from published sources. This case derives from an earlier case prepared by Professor Frances X.
Frei and Research Associate Hanna Rodriguez-Farrar, “Innovation at Progressive Pay-AS-youGO Insurance,” HBS NO. 601-076, which it replaces. HBS cases are developed solely as the basis for class discussion. Cases are not intended to serve as endorsements, sources of primary data, or illustrations of effective or ineffective management. Copyright C 2002 President and Fellows of Harvard College. To order copies or request permission to reproduce materials, call 1-800-545-7685, write Harvard Business School Publishing, Boston, MA 021 63, or go to http://www. hbsp. arvard. edu. No part of this publication may be reproduced, stored in a retrieval system, used in a preadsheet, or transmitted in any form or by any means”electronic, mechanical, photocopying, recording, or otherwise”without the permission Of Harvard Business School. This document is authorized for use only by Sh u-Wei Hsu in Customers and Markets taught by Tao (Tony) Gao University of Massachusetts – Lowell from August 2014 to February 2015. 602-1 75 Innovation at Progressive (A): Pay-as-you-go Insurance insure”were responsible for Autograph’s success.
It was suggested that perhaps the company should expand the pilot to several other states before attempting a national roll out. Others worried that Progressive was becoming istracted by the functionality of new technologies, and losing sight of the insurance business. Progressive and the Insurance Industry4 Progressive had sold auto insurance exclusively since its founding in Cleveland, Ohio in 1937. Recalled Peter Lewis, founder Joseph Lewis’ son and chief executive officer since 1 965: My father… sort of fell into it. He was a lav?»yer but he got an idea.
Car insurance at that time was rather novel. It was only sold in the carriage trades…. He said, ‘I’m going to sell it to factory workers and let them pay for it Cover] time…. I’m going to have a whole set of olicies that are very different. ‘ The company started out as a maverick company. 5 In 1 956, Progressive started writing auto insurance for high-risk (non-standard) drivers, who typically made up 20 percent of the insurance market in any given year. In 1 971 the company went public and moved its headquarters to Mayfield Village, Ohio. By 1987 it exceeded $1 billion in premums. The U. S. roperty-casualty (P/C) insurance industry, although it comprised thousands of companies all vying for a share of the multibillion-dollar market for personal and commercial insurance coverage, was dominated by only a handful of companies. The ten largest P/C insurers (based on premium volume) accounted for nearly 44 percent of net written premiums, approximately $290 billion in 1999. The two largest insurers”State Farm Group and Allstate Corporation” captured close to one fifth of the market. The insurance business was one of shared risk. Insurers set aside a portion of the premiums collected from policyholders to cover losses.
These premiums, called earned premiums, were insurers’ primary revenue source. The second largest component of their reven ues derived from investment income. Insurers invested funds set aside for loss reserves, unearned premium eserves, policyholders’ surplus, and shareholders’ equity. Insurers’ expenses included commissions paid to agents and salespeople, usually deducted immediately from collected premiums, losses (also called claims), and claims-related and lossadjustment expenses including adjusters’ and litigation fees. Insurers also incurred underwritingrelated expenses, such as salaries for actuarial staff.
The underwriting profit (or loss) was determined by subtracting these expenses from earned premiums. Of more than a dozen property-casualty product lines, auto liability and auto damage accounted for nearly 40 percent of premiums written (Exhibit 1 Auto insurers had historically found it difficult to make money on the insurance side of the business, but maintained profitability by investing premiums. (Exhibit 2 breaks down revenues and expenses for auto insurers. ) Auto insurance was sold through both dedicated (single company) and independent (two or more companies) agents as well as directly to the consumer via telephone and Internet. Exhibit 3 tallies 4 Some of the industry information in this section draws on Standard & Poor, Industry Surveys Insurance: Property-Casualty, January 11, 2001. 5 Stepanek (2000). 2 Markets taught by Tao (Tony) Gao University of Massachusetts Lowell from numbers and types of agents for selected competitors. ) Progressive, the largest writer of auto insurance in the independent agency system (with 10 percent of the market) wrote more than 80 percent of its auto premiums through independent agents.
Auto insurance premiums were traditionally rated on attributes of vehicle (age, manufacturer, and value), customer (age, gender, marital status, place of residence, driving record), and types of coverage and deductibles selected. Insurers used this information to establish a driver and vehicle “class” and loss experience with that class to set a rate. Progressive operated in every state except Massachusetts and New Jersey and competed with industry behemoths such as State Farm and Allstate as well as with smaller companies such as Berkshire Hathaway subsidiary GEICO. Exhibit 4 reports market share data for 1999; Exhibits 5-8 provide selected financials for Progressive and its competitors. ) In 1956, with the formation of Progressive Casualty Insurance Company, Progressive was one of the first insurers to enter the non-standard (high-risk) market, in which it quickly became a dominant player. Progressive’s strength lay in its ability to finely segment its customer ase. Lewis maintained: We’re very good price segmenters. We built our business by out-segmenting the competition…. We’re now the largest insurer of motorcycles in the world.
We got into the motorcycle business in 1969. At the time, people who write motorcycle insurance did it based on the size of the motorcycle. We figured out how to also write it based on the age of the driver. What happened when we started adding that bit of information to the equation is that we got all the old drivers and the other companies got all the younger ones. The older ones are better risks. Competitors at the time got their clocks cleaned so they raised their prices. See, they didn’t have the information to show them how their mix of customers was shifting. We did.
We were the first to offer discounts for four-doors and add surcharges for convertibles. Using information, we’ve been able to out-segment everyone else. Now we’re always looking for new ways to use information to segment prices. 6 Progressive’s price segmenting consisted of data mining and extensive statistical analysis of customer behavior. As an example of how Progressive differed from its competition in this regard, consider the following example. Two elderly drivers with identical driving records each have a moving violation (elderly driver A failed to yield while elderly driver B was speeding). rogressive’s competitors are likely to treat these two violations equally in terms of increase in insurance. At Progressive the average impact on insurance for these ?’0 drivers would be the same as the competition but importantly, one driver would be priced above the competition and the other below. Through extensive analysis, Progressive has found that failure to yield should result in a higher premium increase than speeding. Progressive’s ability to segment depended upon its sophisticated underwriting software, which allowed agents to set rates at finer levels than its competition.
While most insurers would simply reject an application from a 19 year old driving a motorcycle with a history of accidents and a poor driving record, Progressive had a rate for that driver because of its ability to factor in other aspects giving a more accurate price for the risk. All insurance companies looked at driver and vehicle location factors to set premiums, but most companies merely looked to their customer history based on zip codes o ascertain the risk. Progressive’s software, however, looked for correlations between 6 stepanek (2000). Markets taught by Tao (Tony) Gao University of Massachusetts Lowell from Innovation at Progressive Pay-as-you-go Insurance drivers, 12 vehicle characteristics, risk, and 16 variables in a credit-scoring model, and extended most risk models to include factors such as typical weather and number of intersections per mile of road. 7 In 1993, after its growth in non-standard drivers leveled out and competitors entered the market, the traditional underwriter of nonstandard and high-risk policies oved into the standard and preferred sectors. Low-risk policies, an eighth of Progressive’s business in 1 995, accounted for almost half of its business by 1999 (Exhibit 9). Between 1 993 and 1999 Progressive advanced from ninth largest to fourth largest auto insurer in the United States. By 1998, Lewis declared: We’re in the big league now, up there with State Farm and Allstate. The question is, can we win the pennant? Today four out of a hundred cars in the United States are insured with Progressive. People laugh when talk about 1 00 percent market share. But if we get better than everybody else in very aspect of the business, why would anybody buy from another company?
Of course, I’d settle for 25 percent. people tell me that 25 percent can’t be done either, but people have been telling me things like that my whole life. 9 (Exhibit 10 plots growth rates for Progressive, the auto insurance industry, and the property-casualty industry. ) A History of Innovation Freedom to experiment was ingrained into Progressive’s culture early in the companys history. Lewis recalled his father’s approach, “The obsession he had, was having the freedom to experiment, to figure out how [we could be better]. 0 Lewis recalled making some crucial investments upon becoming CEO: The first thing did was hire a guy from Travelers [Insurance] who had been in their data processing unit. I recognized that getting in the lead on this stuff would be a competitive opportunity…. Our attitude is that we will try almost anything that makes sense and well stop it when it stops making sense. Weve spent a lot of money on dumb ideas, but we had the flexibility to stop it early and the conversation and the idea-storming [went] on constantly.. It’s ingrained in our culture to experiment, but to do so responsibly.
We reward people for taking risks, but punish them for not spotting bad ones early and pulling the plug. 1 1 In 1988 Lewis and the rest Of the insurance industry got a ‘Wake up call” from consumers.