The publisher's perspective of testing and the curriculum can be described in terms of the criteria for selecting norm-referenced standardized achievement tests: (1) market need--what the customer needs and wants; (2) product match--whether the product neets the need; (3) distribution--whether or not the product can be sold; and (4) end user satisfaction--whether the end user, not necessarily the purchaser, is satisfied. Other issues include definitions of the need, the customer, the real end user, and the product; and how to determine satisfaction. In previous years, testing or research directors typically purchased tests; now offices of curriculum and instruction play a major role. As a result, many school districts are calling for curriculum-referenced nationally normed tests. In most cases, the teacher is the end user of the information provided about the student. It is difficult for a test of reasonable length to match all curricula, since special curriculum contents such as local history are common. Limited budgets may result in purchase of an inadequate customized test matching only part of the curriculum. Ways to measure user satisfaction include the absence of complaints and the usefulness of tests for improving instruction. (GDC)