Gap analysis is a technique used to evaluate the importance of particular product or service attributes determinant in consumers' decision making, then compare those ratings against their evaluation of a particular brand or company. The notion is that resources should be allocated to improving key driver product/service attributes for which satisfaction is low.

Quadrant Approach to Gap Analysis

The quadrant approach to gap analysis requires then graphing the results into a 2x2 matrix with four quadrants, each with strategic and tactical implications. Quadrant analysis is a simple, yet sophisticated method for management to prioritize areas to concentrate resources. Moreover, it is a efficient means of comparing a large number of attributes concurrently on a single visual.

The first step in conducting quadrant analysis is to determine the cross-hairs by which we differentiate the four quadrants. The cross-hairs are the points used to divide the high and low importance attributes as well as the high and low satisfaction or performance ratings. Typically, the mid-points of the measurement scale are used as the cross-hairs. People's ratings of importance of attributes, however, tend to polarize toward the high end, regardless of whether those attributes truly are that important to their decision making. Thus, if we were to simply use the mid-points of the scale, we would see a disproportionate number of attributes loading into quadrants II and III, thus preventing management from prioritizing areas to allocate resources.

Statistical Reasoning's proprietary technique/approach incorporates the grand means of the respective scales rather than the mid-points of those scales as the cross-hairs for the matrix, thereby depicting the relative positions of each of the product or service attributes, allowing management to prioritize investment of resources and where to divest resources.

To conduct gap/quadrant analysis, matching questions pertaining to respondents’ perceived importance of product/service attributes must be accompanied by some evaluation measure, such as satisfaction.  SR can help ensure that the required question variables are 'built in' at the questionnaire development stage.  Please contact Statistical Reasoning prior to beginning the data collection phase of the research project.