Number of clients to whom it is appropriate as set up and training costs are high.

Large populations must exist, as statistics is the science of large numbers.

Adequate controls must exist where there are no controls it is impossible to use statistical techniques.

Population being tested must be homogenous in materiality. Some system of control must exist in each of them.

Too many variables cannot be tested at once.

Items must be separately identifiable therefore sequential numbering is essential.

Expectation of error must be low, i.e. that the internal control system must be reliable.

Risk factor; the level of risk allowable and the degree of risk attached to the item being tested.