Sunday, October 18, 2015

A Tradecraft Primer: Structured Analytic Techniques for Improving Intelligence Analysis

Brainstorming (p. 27-29)

US Government


     This report by the US government covers many structured analytic techniques for intelligence analysis including how to effectively use brainstorming. It allows analysts to come together to discuss a common topic or question, or individual analysts to generate  new ideas without concern for other analysts' opinions or judgments. However, an individual might not be able to develop an idea further or will be unaware of his or her own cognitive biases.
      The US government affirms that brainstorming permits individuals "to step outside their normal analytic mind-sets and suspend their typical 'good judgment' about the practicality of ideas or approaches". This process can maximize creativity helping analysts look at a topic from angles they would not have thought to before drawing attention to 'unknown' or under-analyzed areas.
     The only way brainstorming can produce effective results is if it is structured which involves two stages and several basic rules. The rules are as followed:

  • Do not censor an analyst's ideas no matter how unconventional
  • Look to the source of unconventional ideas to find the connection to the original topic
  • Give the analysts enough time (typically an hour is sufficient)
  • Involve at least one 'outsider' who has a different educational, cultural, or technical background as the other group members
     The two phases of structured brainstorming in this report are first the diverging stage followed by the converging stage. 

The diverging stage:
  • Give participants 'Post-it' notes and pens
  • Post question or topic on board/screen that everyone can view
  • Ask the group to write down their responses to the topic using key words that will fit on the post-it note
  • Stick the notes on the wall treating them all equally
  • "When a pause follows the initial flow of ideas, the group is reaching the end of their conventional thinking and the new divergent ideas are then likely to emerge"
  • Collect all notes after two or three pauses
The converging stage:
  • Ask participants to arrange the notes on the walls based on similarities or commonalities without talking
  • Select a word/phrase that characterizes all of the words in the group
  • Identify outlying words that do not fit into any groups and assess significance
  • Identify new concepts or ideas as well as ideas that might need more brainstorming
  • Vote on one or two areas that attention should be devoted to
  • Set group priorities to decide next analysis steps based on voting results

     This report did not discuss short comings for using brainstorming as a modifier for intelligence analysis. It did mention one sentence about cognitive biases regarding the individual's list of ideas but nothing about the risk of conformity or group-think. Although the diverging and converging stages are simple and easily reproducible, this report does not account for the weight of group dynamics in the brainstorming process as a whole which is very important to the efficacy of this technique.



  1. It's interesting that you note the lack of cons discussed by the author because I was thinking the exact same thing. However, the publication that contains this (and other) techniques is not really the forum for in-depth discussions on strengths and weaknesses of techniques, but more of a how-to. However, in the diverging and converging stages, the authors are essentially describing Nominal Group Technique, which eliminates many of the weaknesses of traditional brainstorming.

  2. I thought this is a great piece on how to more effectively impliment the technique, we essencially did this last year in our first intelligence class with Professor Wheaton. This seems like it would be possible to develop a means for conducting this technique remotely via the internet. There would just have to be a well designed user interface for having the group collectively sort ideas by type.

  3. I agree with Dan. It really explains very well how to implement brainstorming. This seems more like a nominal group technique. And I think it is very useful in terms of generating unique and quality ideas.

  4. I agree with you Andrew, that if structured this way it can eliminate the weaknesses. I was hoping it would go in depth as to why structuring the brainstorming this way is more useful than others and it is very similar to NGT.