As the examples I've described reveal, finding a target consumer requires a layered approach to get the best of each type of media planning strategy. In the case of our video game maker, a strong plan might look like this:

Strategy -- share of budget -- rationale

• Integrated sponsorship in a contextual gaming area(s): 30 percent -- serves as centerpiece of a campaign and allows unique "breakout" positioning for new launch but only reaches a small portion of audience. High cost per unique reached but high impact.
• Contextual ad network: 30 percent -- reaches almost everyone the sponsorship will miss who is interested in gaming. Lower cost per unique with massive scale.
• Behavioral ad network: 30 percent -- finds people with high gaming purchase behaviors based on key indicators of interest and provides consumer insights. Price per unique is middle of the road but delivers massive scale with accuracy.
• Mass reach program with added demographic targeting filters and smaller demo focused sites: 10 percent -- a lower cost way to find similar people but interest is uncertain.

The percentages will certainly vary by client and category, but approaching the planning process as a layered strategy with a full understanding of the benefits and downsides of each approach makes for a stronger plan. Allocating budget based on scale ensures dollars aren't spent delivering too high frequencies to small audiences who can't move significant product off the store shelves.