Sample Memo on Reshaping a CI Project

November 2, 2000

[The Financial Institution] is seeking to develop common intelligence on up to twelve major national and regional competitors.

Each of these competitors may be facing us in up to four separate market niches. [For the sake of discussion, we will assume they are each in all four niches]. For each competitor and niche, we will be seeking CI on the following areas:

• marketing strategy

• promotional thrust

• advertising expenditures

First, based on our conversations with others on your staff who will be using this CI, I would suggest that the CI project include, in addition to the above, the following considerations:

• a summary of the overall strategic thrust of the parent organization, as that thrust directly impacts or drives the direct competitors' marketing, goals, budgets, and so on

• a separation of what the competitor has done in the past year from what it should or could be expected to do in the next year

Having said this, I should alert you to the fact that a project like this can quickly become so complex as to be unmanageable in any limited period of time. In this case, for example, we might have the following template for each competitor (see Figure 9.1)

Figure 9.1: Competitor templates.

A model such as this generates 8 separate cells to be completed for Market Niche D alone, and thus 32 for this single competitor. Taking 12 competitors then produces 384 separate cells of intelligence, with some, but not always much, overlap among them.

As you know, time and cost are trade-offs in CI. Therefore, the more data sets that have to be generated, the higher the cost and usually the longer it will take. And the more cells we create, particularly when it splits a cell into smaller parts (such as by size, date, geography, etc.), the more likely it quickly becomes that we will not be able to get all the cells filled.

A project like this can quickly and innocently become unmanageable. Let me show you how. Take an apparently simple request to subdivide present and future advertising expenditures into radio, TV, print, and direct mail expenditures. That one change would immediately add 8 cells (4 for the year 2000 and 4 for the year 2001) to each market niche for each competitor. That alone suddenly doubles the number of niches, to 768 separate intelligence tasks.

For these reasons, we should consider using one or more of several strategies:

• Severely limit the level of detail sought on each of the twelve target competitors;

• Develop significant detail only on those niches and below for which the cost and time to produce/acquire it can be justified; and/or

• Select a few, very key competitors and, for each of them, select only a limited set of market niches. Then, develop the targeted intelligence on them. From there, evaluate the utility of such intelligence to us in light of the time and cost. Then we can decide whether expand the research to cover more competitors, more niches, and more subsets in the future.