One of the most challenging problems in marketing today is proper attribution between spend and results. The first step in solving this problem is to establish a framework through which events can be correlated to an individual.
I would suggest that marketers borrow play from physicist and adjust events to a common frame of reference. A likelihood of exposure to a marketing activity can be thought of as a normal distribution of probability from point t=0 to some point in the future. When the normal distributions are layered onto the same axis, a vertical line can be drawn at any point in time and the relative probability for each media exposure can be derived.
Here's the catch... it's very difficult to adjust to a common frame of reference without strong multi-channel recognition technology and it takes time and experimentation to work out the actual shapes of the response curves for different media types.
