Although financed largely by broadcasters, BBM was controlled for many years by advertising interests; of the nine positions on the original board of directors, three were filled by advertisers, three by advertising agencies, and three by broadcasters. Shortly after the creation of BBM, a similar organization called the Broadcast Measurement Bureau (BMB) was established in the United States. As a result of the efforts of Horace Stovin, chairman of BBM’s technical committee, the two organizations worked in concert for a few years, using the same mail-ballot technique and running their surveys simultaneously. This enabled advertisers to operate on either side of the border with equal facility. However, BMB was criticized for its methods, plagued by high costs, and thrown into disarray by the resignation of its president, Rugh M. Feltis. In 1950 it threw in the towel and left the U.S. station coverage field to the A.C. Nielsen Company, which used an interview-aided recall method.

By the end of Phenner’s presidency in 1951, BBM had increased the number of areas surveyed, introduced bilingual ballots in some areas, and more than doubled its broadcasting membership. However, a number of stations still refused to join, and in 1956 the CBC withdrew because of dissatisfaction with BBM’s surveys. The same year, BBM began producing time-period ratings for radio and television using a paneldiary method pioneered in Canada by International Surveys Limited (ISL). The new surveys were initially conducted every spring and fall, with each member of participating households keeping a week-long diary of listening and viewing by half-hour periods. At the same time, the circulation surveys were increased from every other year to twice a year. However, the CBC remained critical of BBM operations and subscribed instead to the A.C. Nielsen Company, ISL, and McDonald Research. A 1962 CBC report criticized BBM’s surveys for their “non-coverage, biased selection procedure, low response and poor quality of response.” By then BBM was also coming under strong criticism from both advertisers and private broadcasters, and there was a danger that it might collapse.

Under Bill Hawkins of CFOS Owen Sound, BBM began to put its house in order. It revised its constitution so as to increase the representation of broadcasters, and in 1964 the bureau became the first ratings service in the world to introduce computerized sample selection. It also increased the number of surveys, redesigned the bilingual household diary, and switched its premium from a card of safety pins to a 50-cent coin. In terms of winning back confidence in the validity of its surveys, the most important step was taken in 1967, when BBM decided to switch from household diaries, which had usually been kept by the harried homemaker, to personal diaries sent to selected members of households—including children, although their diaries were filled out by an adult. This change increased the response rate for mailed diaries to almost 50 percent and facilitated the acquisition of demographic data. Within a few years, BBM became the only audience measurement service for radio in Canada, and in television the competition was reduced to Nielsen. Between 1963 and 1968, BBM increased its membership from 357 to 534, or about 90 percent of the broadcasting industry, including the CBC.

Unlike the original household diary, the new personal diary was used for both radio and television, largely for reasons of cost. In theory, however, the most reliable diary is the single-medium personal diary. In addition, the use of dual-media diaries irritated radio broadcasters, who argued that they provided BBM with twice as much revenue as television broadcasters but only received the same benefits. In 1975, therefore, following several studies and considerable debate, BBM adopted separate diaries for each medium, including different samples and survey dates. This move greatly increased survey costs, however, so that in the mid-1980s BBM implemented household flooding or saturation sampling for both radio and television. Ironically, this development brought BBM almost full-circle back to its original household diary technique and illustrated the fact that audience measurement methods generally are determined as much by economic considerations as by the requirements for scientific validity.

In the mid-1970s, BBM began investigating electronic measuring systems. A committee was set up to develop a proposal for a meter-based system for television, and a contract was signed with Torpey Controls Ltd. for a prototype using existing circuitry and the vertical blanking interval. Despite successful test results, however, the cost of switching from diaries to meters was considered prohibitive, especially since diaries would still be required for radio and to supplement the data gathered for television. It was not until the advent of “electronic diaries,” or Peoplemeters, by the A.C. Nielsen Company and others in the early 1980s that BBM gave serious consideration to replacing its traditional diary system for television. Unlike the original Nielsen Audimeter, the Peoplemeter measured viewing rather than mere tuning and could track audience flow much more precisely.

In 1984, while still testing its new meter technology in the United States, Nielsen announced its intention to launch a Peoplemeter service in Canada. In response, BBM turned initially to Audits of Great Britain for help but then decided to invite bids from other companies as well, including Nielsen. In November 1985, A.C. Nielsen Company and BBM reached a tentative agreement by which Nielsen would provide BBM with Peoplemeter data from 1,800 Canadian households, which BBM could then market as it saw fit. The agreement later fell through, however, and in September 1989 Nielsen launched on its own a Peoplemeter service for network television in Canada. BBM tried to develop its own electronic television audience measurement (TAM) system in conjunction with Les Entreprise Videoway, but the tests results were unsatisfactory. Late in 1990, BBM and Nielsen resumed talks for a joint venture to extend Peoplemeters from the national network level to local and regional broadcasting. The following year, however, a proposed deal again fell apart, because of the concerns of local and regional broadcasters about costs and various technical matters. In 1996 BBM created a New Media division to mea sure interactive media, and in 1998 the bureau launched an advanced TV Peoplemeter service in Vancouver with plans to expand nationally. In 1999, ComQUEST Research Inc., a subsidiary of BBM formed a decade earlier, joined with Media Metrix Inc. to form Media Metrix Canada