The practicality of systematic observations is readily apparent. One researcher used 200 hours of videotape recording of diaper changing at a day care center to help with the redesign of disposable diapers.
A large food retailer tested a new slot type shelf arrangement for canned goods by observing shoppers as they used the new shelves.
Toy manufacturers regularly put children in one way mirrored room to watch how they play with product prototypes.
Observation is the second method of collecting data. It is the process of recognizing and noting people, objects and occurrences rather than asking for information. For example, instead of asking consumers what brands they buy or what television programs they view, the researcher arranges to observe what products are bought and what programs are watched. Very few of the ‘one-shot’ research projects done in marketing are of this type perhaps no more than 1 percent. Some of the major continuing services, however, such as the Nielsen Television Index of television viewing, do use observation methods. The development of the Universal Product Code and scanning at supermarket checkout counters, moreover, now take observation a much more important method of collecting data than it has been.
About 50 percent of all customer purchases in American supermarkets are now recorded by passing each item over a laser scanner that automatically reads the Universal Product Code printed on the package. This activates a computer memory that identifies the current price, calculates the amount due, and prints out an itemized sales slip that shows each items purchased with the price. For practical reasons such data are usually accumulated and reported on a weekly basis. As all the data are stored in the computer from the instant the product code is ‘read’ data can be made available within a few days rather than weeks or months required for other methods of collecting grocery product sales data.
Lasers read product codes accurately on curved surfaces and even when the printing is poor, but malfunctions occur. Other errors occur when several of the same items are purchased and the clerk may enter the cost manually or when large units, such as 25 pound packages of dog food, are not lifted but are recorded manually. Sometimes they do not use the Universal Product Code. These things mean that scanning does not provide perfect data, but it is believed to provide more accurate data than any other method of collecting sale information and it provides an entirely new database for more sophisticated market analysis than has been available.
New technology is providing additional ways of observing other customer information to combine with the scanning data, described above. Several companies now have consumer panels, each member of which carries a card that identifies him or her. When these individuals make purchases at the supermarket, they give their cards to the clerks who put the cards in a scanning machine and record the purchaser as well as the item purchased. Demographic data on the customer are in the computer data bank and can then be correlated with purchases.
A firm called Information resources Inc., now has a panel that, in addition to the above has a microcomputer on each panelist’s TV set tracking each commercial that plays in panel members’ homes. Cable TV permits the firm to show different commercials to some homes and, thus, to test the relative effectiveness of various commercials.
Some of the advantages of the observation method of collecting data as compared to the questionnaire method are obvious. If the researcher observes and records events, it is not necessary to rely on the willingness and ability of respondents to report accurately. Furthermore, the biasing effect of interviewers or their phrasing of questions is either eliminated or reduced. Data collected by observation are, therefore more objective and generally more accurate. Several comparisons of observation and questionnaire data were presented.
Unfortunately, the observational method also has a number of weaknesses that keep it from being widely used. Researchers have long recognized the merits of observation as opposed to questioning, yet the vast majority of marketing research projects continue to rely on the questionnaire. Probably the most limiting factor in the use of observation is the inability to observe such things as attitudes, motivations, and plans. Only as these factors are reflected in actions can they be observed and then they are confounded with so many other factors as to make their identifications difficult, if not impossible.