Stated Need Versus Real Need

In an interview, a warehouse supervisor tells you she needs to export the daily pick list to Excel. She is emphatic about it. You could build the export in two days. Instead you ask what she does with the file. She prints it, highlights four rows in yellow, and leaves it on a clipboard by the door, because the temporary staff who start at six have no logins and would not know which orders are urgent anyway. She does not need an export. She needs the urgent orders to be visible to people who cannot use your product.

A stated need is the solution someone has already designed inside the constraints they believe exist. It is honest, and it is useful, but it has been filtered twice: once by what they think is technically possible, and once by what they think you would agree to build. Both filters throw away information you need.

Getting from the first layer to the third is mostly a matter of asking what happens next. Not why, which people find slightly accusatory and which tempts them to justify themselves, but what happens next, and then what, until you reach a point where something outside your product happens. That point is usually where the real need is.

The dark mode row is deliberate. Sometimes the stated need and the real need are the same thing, and the honest answer after investigating is that they want dark mode because they want dark mode. Do not turn this technique into a ritual where every request must be revealed as something deeper. That produces its own failure, where customers stop asking because they know they will be interrogated.

Four questions that get you underneath

• What do you do with it once you have it? Follow the artifact out of your product and into the world. The interesting part is almost always downstream.
• Who else touches this? Needs that involve a second person are usually misstated, because the person you are talking to is describing their half.
• What did you do the last time you needed this and could not? The workaround is the real requirements document.
• If this worked perfectly, what would you stop doing? The answer names the cost you are actually removing, and sometimes reveals that the cost is small.

That last question is the one people skip and it is the most efficient of the four. If the honest answer is 'nothing, really, it would just be a bit nicer', you have learned something important in eight seconds.

One caution about going too deep. You can always abstract further. 'Export to Excel' becomes 'share information', becomes 'coordinate work', becomes 'run a business'. At some level of abstraction every need is the same need and you can no longer build anything. The right stopping point is the highest level at which you can still name several concrete, different solutions. Go one level past that and you have a philosophy rather than a brief.