There are two tools which are very useful when tackling a design, review, or evaluation exercise. These are the Ishikawa/Fishbone diagram and the KJ method (affinity diagram). They are often used in quality circles for problem solving, figuring out what should happen, and what is happening. Another useful resource is the description about root cause analysis.

While these are very useful tools when designing, reviewing, or evaluating processes, it is important to remember that they are mere tools and not magic bullets. It is easy to use either tool and end up with taurus excretus. Just because something was done with a fishbone or KJ method does not mean that it makes sense or was done well. It is easy to blindly follow the steps and fail. It is easy to be over confident. A diagram will be created and an activity accomplished, but it might not be the appropriate diagram or have the appropriate relationships.

Observations…

• It is possible to use these methods individually or in group brainstorming situations.
• We do not recommend groups larger than six or eight individuals when doing a group brainstorming session. Large groups are rarely effective or efficient when addressing an issue or trying to do problem-solving.
• The tools assume and require certain cognitive skills and expertise levels if they are to be used effectively and efficiently.
• If you have not used the methods before, do not start with something critical or high impact. Practice at least 3-4 times on smaller tasks before using them.

• Let the diagrams and groupings flow organically, do not force a pre-existing model; e.g., do not think that everything can be reduced to the common, mature manufacturing fishbone found in manufacturing: operator, machine, material, method.
• Do not have a pre-existing diagram or groupings in mind when you start an exercise.
• Avoid being too formal or too precise as this can intimidate people participating in the brainstorming activity.
• It is possible to use the concepts behind KJ which is a way to elicit attributes and qualities, or constraints, or topics related to the situation and use the fishbone method for representing the groupings.
• There is no right or wrong number of spines on fishbone.
• It is important to start the exercise with understanding what the outcome is supposed to be, and how this might be explained or quantified. This is what goes on the left or right hand side of the fishbone; with the spines representing the causality factors that contribute to the desired effect. The first analysis or exercise should not be restricted by constraints (e.g., budget); it should be a reasonable green field. This is also an affirmation approach; starting off with a what should be versus what is. The what is analysis can result in finger points and other problems.
• The major spines should be at the same level of granularity and contribution to the effect or outcome.
• Each major spine can then be dealt with as its own fishbone if it’s complicated, with sub spines and sub sub spines. This is why it has been nicknamed the fishbone diagram.
• It is important to have the spines balanced and all justified and rationalized with respect to the contribution to the effect. Thus, while there is no right or wrong fishbone for a situation, there are good and bad fishbone’s possible.
• The categories can be considered from a Zen perspective; all that makes it what it is and nothing else. What makes a chair at chair? Does a chair need arms? Does a chair need fabric? Those types of questions.
• When a fishbone gets too cluttered, that is when you go into a hierarchical diagram structure.
• A diagram should be easy to comprehend and if it appears as if there are many major spines, a common technique is to review the spines and see if there is really a meta-grouping or category that will collapse part of the fishbone.

• Often an exercise using fish bones in the KJ method is iterative and the diagram is sketched in parallel with the KJ idea generation as sometimes it can help think of ideas and points. This will depend on the situation and what is being analysed.
• Do not be surprised if the fishbone goes through many iterations as the exercise proceeds. Rarely does a fishbone diagram proceed without alteration.
• The fishbone diagrams can also be useful in the reports or documentation as they are a good way to visualize the components and the relationship the components have the final effect. They are very useful when designing maturity benchmarks because of this.
• It can often take 10 or 12 exercises before someone is comfortable doing a fishbone diagram with the KJ method.
• We have done this type of process analysis for what makes a good program, a good course design, a good course delivery, a good instructor, and a good student.
• Doing this type of exercise not only helps you understand the problem and get inspiration for potential ways to review it, evaluate it, and alter it, but if done in a small group setting, it also helps you understand how people think about it, what they see the relationships to be, and how much the people actually understand, depth and breadth. So it is important to be aware of how the fishbone is created, not just the final diagram.

Further reading

• https://asq.org/training/asq-quality-tools---fishbone-diagram-fdasq
• https://asq.org/quality-resources/affinity
• https://asq.org/quality-resources/root-cause-analysis/tools