At the time, television viewing was being measured by several different groups using varied techniques such as telephone coincidentals (calls made to viewers during television broadcasts), recalls (telephone calls made on days subsequent to broadcasts), and even door-to-door questionnaires. The common element that brought Cooper and Seiler together was that each found that distributing a viewing diary had distinct advantages in developing audience ratings for this new medium. Viewers could be measured from early morning to late night without being bothered by telephone. Moreover, audience composition, as well as household ratings, could be developed. Audiences outside normal dialing areas could be measured, and net weekly cumulative audiences could be produced.

The two researchers joined forces, incorporated, and established headquarters in Washington, D.C. At about the same time, John Landreth formed a company called Television National Audience Measurement Service. In 1951 he was directed to ARB and after a meeting with Seiler and Cooper became the third partner in the research endeavor.

ARB developed its own methodology for audience measurement. First, a random sample of homes was drawn from telephone directories of the area surveyed. These households were then contacted to determine whether or not a TV was present. One diary, with an explanatory letter, was mailed to the chosen respondents. Each television set in the house was monitored with a separate diary. The diary keeper in the home would record television viewing at 15-minute intervals day by day for seven days and then return the diary. It was determined that four weekly samples would be the basis for each market-research report. Diaries were tabulated manually and a simple report was prepared on a program-by-program basis during prime time. A Monday-through-Friday combination report was prepared for daytime programming.

In the early 1950s, ARB was ready to expand its operation. The Federal Communications Commission (FCC) lifted its 1948 freeze on new station license allocations in July 1952, and many new stations began telecasting. Advertising agencies needed a service to measure viewing in the increasing number of rapidly developing television markets. In 1952 ARB was measuring 15 TV markets. In order to position itself as the industry leader, the organization took a quantum leap and expanded to 35 markets. Ad agency support and usage of the company’s TV market reports followed and enabled ARB to be a pioneering leader in the exciting new field of audience measurement.

By the late 1950s, it became obvious that a better way had to be found to develop the diary data. Manual tabulation of the data from diaries was impossibly slow. ARB moved its headquarters to Beltsville, Maryland, and installed a UNIVAC tabulation method and report preparation. This new system almost put the company out of business.

The first reports produced by the system were woefully late; in some markets the reports made no sense. Gradually the company worked its way out of its dilemma. By the 1961–62 television year, ARB was on a better footing and had generally solved the problems it had endured. The new computer equipment gave the company the capability to expand its market reports to include needed data on specific demographic groups, making reports invaluable tools for advertisers and their agencies seeking to buy and sell spot television time.

By the early 1960s, homes owning a TV set had increased dramatically and hundreds of additional television stations had begun telecasting. Hundreds of thousands of diaries were being placed in American homes each year. By 1967 ARB had clearly defined 225 television markets. It produced television market reports called “sweeps” twice per year for every television market, and from four to seven times a year for the larger major markets. The sweeps provided comparative cumulative data for an entire week. TV’s advantage as an advertising medium was thus well documented and appreciated; hundreds of millions of dollars were pouring into station and network coffers.

At this time, as a result of demands from advertising agencies, a new and exclusive market definition called the “Area of Dominant Influence” (ADI) was introduced in ARB reports. The ADI was a collection of counties in which the viewing of particular stations in the market was dominant. Some station executives violently objected, complaining the new ratings did not reflect the true size of their station’s reach. To counter this, ARB continued to report total homes viewing the station and demographic characteristics of the total audience.

From its inception, ARB’s major competitor was the A.C. Nielsen Company. In the local market field, ARB was usually considered the innovative force, normally reacting quickly to what the ad agencies needed in the report. Although many advertising agencies subscribed to both rating services, ARB usually had a larger list of user agencies. In the larger TV markets, a majority of stations were subscribers to both rating services.

During the 1980s, the two services were caught up in a rapidly changing electronic media marketplace. Arbitron delivered reports on cable penetration and cable viewing within specific markets. It made a large investment in ScanAmerica, a unique service that combined viewing estimates with product purchase surveys. Additional investments were made to change methods of measurement. In larger markets, diary surveys were converted to an automated system that used a sample in which special equipment was attached to the television set. Viewing data from the meter was carried through telephone lines to an electronic data center. In the larger TV markets, metered research provided reports on a more timely basis; indeed, even overnight program ratings were now available.

These very sophisticated research methods were not only costly to install but also expensive to maintain. This resulted in substantial increases in the cost of market reports. TV stations had always borne most of the cost for the audience research. Both Arbitron and the Nielsen Company charged agencies a token amount for the complete package of all market reports produced.

In the competition Arbitron began losing market share. By the end of the 1980s, it had 19 metered markets, to Nielsen’s 29, and the number of TV stations subscribing to ARB’s market reports based on the viewing diary declined.

Finally, in the fall of 1993, Arbitron president Stephen Morris declared that his company was out of the television-measuring business, contending that the marketplace would no longer support two rating services. It was revealed that approximately 275 stations subscribed to both Arbitron and Nielsen local market reports. But Arbitron’s lists of exclusive station subscribers had dwindled to 180 clients while Nielsen could claim 359 exclusive subscribers. According to Morris, Arbitron would continue to provide specialized TV-audience research for television stations and advertising agencies. But for the first time in nearly 40 years, the sales offices of TV stations and the research departments of ad agencies were dependent on a single source of local market research reports.

As a company, Arbitron is still in existence. It continues to measure successfully radio-listening audiences, using the personal diary, and its research reports are widely used in the radio industry. In recent years it has developed highly successful programs of syndicated market research contracted to newspapers, cable television services, and Internet radio as well as for radio broadcasting. The most ambitious project has been the development of the Portable People Meter, a device that would record media use by individuals in any context or venue. Designed to track audiences for terrestrial, satellite, and Internet radio stations as well as broadcast, satellite, and cable television, this technology could once again alter the business of measuring media use and thereby the economics of the media industries.