But what is considered to be the best depends on the situation. It could be based on reach , response , demographics , cost, etc. The restrictions could also include a variety of factors including cost, demographics, program type, network, program mix, etc.

For example, at a rather simple level, we could optimize on one of the following:

• Maximize average recall at a given cost.
• Maximize reach at a given cost.
• Minimize cost while achieving a desired recall or reach.
• Etc.

The process of optimizing an advertising plan still requires human judgment. One common mistake in optimization is allowing some automated process to make all the planning decisions. There will always be factors that cannot be considered by an optimizer which require the experience and judgment of a planner.

A detailed technical discussion of optimization, together with a statistical basis and examples for using an integrated method, see An Integrated Method for Optimizing Advertising Plans.