KJ Method to Get New Ideas
Posted on by Gentaro "hibariya" Terada
KJ method is a way to integrate classified and arranged information and encourage to conceive new ideas. This will be useful to make better solutions for difficult issues. Today I introduce the outline of that.
1. Explore and Record
To begin with, collect and record information as much as possible from all the participants by any method such as brainstorming. Summarize each idea into short title and write it to a little paper (e.g. Post-it). As a rough standard, there may be several dozen to one hundred plus several dozen ideas during 2 hours session. To keep an essence or context of each idea, keep casual expression. In other word, a title should not be abstracted too much.
2. Grouping
And then, group all ideas. First, Spread all ideas in one place and read (look) one by one by any order you like. In a matter of time, you will find some ideas which have sense of affinity each other. Move them into one place and make a group. All ideas will become several groups of ideas. And create (find out) a title for each group considering “why these ideas make this group?". A title will be a summary of the group. It's good idea to change the color of a title from other ideas.
In the same way, group all those small groups and create a title for each middle group. Also, create large groups. Each title of the large groups should be a subject of the discussion.
3. Illustration
Next, try to find the best layout of the groups and ideas which makes sense by reading the ideas carefully. If you can create sentences clearly from your layout, then it suggests the understanding which affects the layout is relatively fair.
Once the layout is created, then illustrate by diagrams what does it mean by connecting related groups and ideas. For example, you can use <-->
for mutual relationships, >--<
for adversary relationships and -->
for causal relationship. And you can also surround related groups.
Note that grouping doesn't have to be “logical at a glance”. In other word, relationships of group don't have to be semantically tree.
4. Documentation
Finally, try to describe the groups in writing. It's good idea to start with a group which was pondered. In the middle of writing, please distinct clearly from description and interpretation.
Sometimes you may find room for improvement of the illustration which you created on before phase. Because documentation and illustration can complement each other's weak side.
5. Free Competition of Information
Each group you created can be a BAD (Basic Abductive Data) which stimulates ideas you want. You can pile up these BAD by considering BADs cumulatively. Worthless BADs will not grow. Useful BADs acquire stability by mutually enhancing each other. Useful BADs also acquire comprehensiveness by bringing higher ideas. In this way, thanks for free competition of information, new ideas for tentatives grows naturally.