House of Quality matrix resolution

House of Quality matrix area

Determining the appropriate level of details in the QFD House of Quality is a double-edged sword. On one hand, people in the team can lose oversight if the resolution becomes too fine, where the wealth of details makes the House of Quality so complex that we can no longer visualise the rationale behind our design decisions. People, also including the overseeing project sponsors, will perceive a diminishing return-on-investment from our efforts, as we grow the amount of work that is required to manage and evaluate a larger number of increasingly trivial planning details. Yet, on the other hand, if requirements in the House of Quality are over-simplified then people may (rightly) suspect that important details are being left out. The QFD team members need sufficient information to give them confidence in forward-applying the planning outputs they obtain from the House of Quality.

So, what is an appropriate number of requirements to manage within a single House-type planning matrix? A very few design requirements, say 2 or 3 only, are so easily resolved that it would hardly make it worthwhile setting up a planning matrix. As for higher numbers, several factors can play in, such as the level of resolution and many-to-many relationships; but as a rule of thumb for a single planning matrix

5×5 requirements are simple
10×10 requirements are easy
20×20 requirements are comfortable
30×30 requirements are many
40×40 requirements are too many

Like in any other project management methodology that contains an excessive number of requirements to be evaluated and mutually resolved, the large QFD master project could be broken down into a hierarchy of interrelated sub-projects. Or, we could adjust the resolution by combining requirements, as when balancing the translation,  or simply omitting the lesser important ones from the matrix evaluation (but not from the customer requirements list). Reduce the focus if you must. Remember, 20% of designer actions will create 80% of the new product value. Minimal efforts can therefore still produce a successful result; but it requires that you understand the VOC and you target what are truly the most important requirements, without losing sight of their context.

In many practical commercial QFD projects, reality is that some team members are non-scientific or they are new to QFD and may not naturally have the fluency in reading a busy algorithmic House of Quality. These people will have a lower threshold, before switching off to a perceived academic level of complexity. It is the people in the QFD team who produce a quality and innovative design.

Losing the participation of project team members is significantly more critical than it is getting the odd planning refinement perfectly right.

The House of Quality should be worked effectively and efficiently; but its level of details should only be refined to a point that is before people start to fear that they cannot cope with the perceived complexity. Just remember, a simplified smaller House of Quality, although appearing unrefined to the purist, still maintains a merit in getting everyone around the table and collaborating on producing a targeted design. The fact that data is worked at a lesser resolution makes it important to confirm with all team members whether the output from the House of Quality agrees with their experiences and intuitions. And if not all can agree, then we must look at the data again, to resolve any issue. We should not consider the House of Quality work completed until everyone agrees that its output makes good sense.