list test size rules

The larger the test group, the more accurate are the results. For example, a test of 10,000 names of a 50,000-name mailing list will give more accurate results than a 3,000-name test of the same list. As few as 2,000 names will usually be satisfactory for a test, although many owners of large lists have minimum rental requirements of 3,000-5,000 names, sometimes 10,000.

Some marketing “pros” say the test segment should be sufficiently large to produce a minimum of 25 to 50 responses.

The expert's thoughts on direct response - growth hacking - performance-based marketing activities - DIRECT MARKETING