Analysis of intelligence always involves cognitive and social processes

Essentially, analysis is a cognitive activity.

Although intelligence analysts often utilize technological assistance and input from others, ultimately it is the human brain that organizes and interprets data to produce an assessment or prediction. The intelligence community and academia have done considerable work to identify the cognitive biases that can impair the validity of analytical conclusions, and the heuristics that can help analysts do their jobs efficiently and well.

Analytical work is a social process.

Of course, the conduct of intelligence analysis always involves relationships with those who assign analysis tasks and receive analysis products. Commentators and policymakers alike point out that when these relationships become politicized, the validity of the analysis's conclusions can be threatened. A less widely recognized fact is that the work of analysis is itself highly social. Analysts who extract meaning from a set of data in isolation are the exception. Instead, analysts typically draw heavily on the expertise, experience, and insights of their colleagues when developing and testing their conclusions. As the volume and variety of intelligence data increases, teamwork appears to be becoming more prominent in the production of analytical reports.

Different perspectives on analytical work have important implications for the design and leadership of analytical units.

The cognitive perspective puts the individual analyst center stage. Managers who hold this view tend to organize work to encourage and support superior individual performance. They may pay special attention to selecting talented analysts, training them well, and providing them with sophisticated technical and informational support. Of course, analysts still work in units where others do similar work, and individual contributions may be aggregated into a unit-wide product. But everyone works in parallel, and each analyst is responsible for his or her individual output. This unit is called a synergy group. While members of a cooperative group typically communicate with each other and consult extensively, this communication is primarily intended to help individual members competently discharge their personal responsibilities.

In contrast, the social perspective focuses more on the importance of team interaction in effectively assessing uncertain data, dealing with porous datasets, and managing relationships with those who provide data and those who receive analysis reports. Managers who hold this view are likely to form interdependent work teams whose members are collectively responsible for a major analytical effort. Since the work is done by a team, it may be larger in scale and potential than any individual task would normally do. Of course, members of work teams bring their own special expertise to work, and over time develop specialized team roles—but it is the entire team that produces the analytical product and is accountable for it.