In general: means used to evaluate the behavior of a person, the characteristics of a person or thing, or the effect of an action. Tests require a method and scale of measurement and a controlled environment in which to observe or measure. For example, to test a light bulb, the method used might be to place it in a lamp that is plugged into a socket and turned on. The scale of measurement relates to the increase in light seen coming from the light bulb. The control involves using a lamp that works, has access to electricity, and is visible to the observer.
Advertising: introduction of a new product or promotion on a small scale to measure consumer response. Testing of this nature is usually limited to consumer marketing. In business-to-business marketing, reactions are usually gathered through personal interviews conducted before a product is introduced to the market. A product test may include the initial in-factory assessment of the product, a small test by a group of potential users, and, finally, a market test which is a simulation of a national launch.
The design of an effective market test includes decisions regarding where to test, how long to test, and what to do in response to test results. A test location is generally chosen based upon the presence of demand for that product and the degree to which consumers in that location/city are representative of the total market. The duration of the test depends upon the money budgeted, the length of time it usually takes for a consumer to make a second purchase (or first purchase if the test sample was free), and the probability of a competitor's entering the market. See also test market.
Direct marketing: evaluation of one or more elements of a marketing strategy or promotion on a small scale prior to full-scale introduction. Tests usually consist of a standard element called the control (such as a control package or offer, etc.) and a test element. The control represents the package that has outperformed every other package used to date. The objective of the test package is to assess its ability to beat the control. The control and the test are sent to sample groups of prospects or customers selected using a statistical sampling technique. The validity of a test is dependent upon the degree to which the sample selected is representative of the promotion universe.
The elements most frequently tested are product, offer, list, copy, and package; however, virtually every aspect of a direct-marketing promotion gets tested at one time or another, from the tilt of the address label to the taste of the glue on the reply envelope.
Direct marketing is uniquely well-suited to testing, because the results of a promotion change are easily measured in terms of response. Most direct marketers include a test of some sort in every mailing or promotion they do. One basic rule of testing is that only one element at a time should be changed, so that a difference in response can be clearly attributed to that change. It is commonly understood that what doesn't work for one marketer might work for another; therefore, the results of a test should not be universally applied across all marketers. See also a-b split; dry-testing; key code; nth name selection; panel; pyramiding; rollout; rule of 300.
