Do We Really Need a Work Instruction for Every Little Thing?

Dave Crysler answers the question a client asked him in the middle of a gap review, the one everybody in the room was already thinking. There is no clean line between tribal knowledge and a work instruction, so most teams stall at the question and write nothing. Dave lays out the two questions that break the tie, the cost of leaving a gap undecided that almost nobody counts, and why a step that is already documented can keep failing anyway. This is the episode for anyone staring at a list of process gaps trying to work out which ones are actually worth doing something about. What You'll Discover: • Why nobody will give you a straight rule about what to document, and who benefits from that • The two questions that decide it: would this work at double the headcount, and what does it cost every time somebody gets it wrong • Why exposure decides what gets documented, not completeness • The hidden cost of a gray area, and the private calculation your team runs several times a day because of it • Why the person who knows the process best is the worst judge of what needs writing down • The billing step that was in the work instruction the whole time, and still kept getting missed • Why writing it down does nothing when the step depends on somebody remembering • The difference between a training problem and a system problem, and the one question that tells them apart • Why operator error, operator retrained is a prediction rather than a resolution • What designing it out actually means, and why building the tool first just automates the wrong process faster • The third answer nobody talks about, and why deciding to write a document is not the same as fixing anything If your team keeps working around the same thing, or somebody just told you the fix is a new SOP, start here. You have more than two options, and the one everybody reaches for first is usually the third best. Dave closes with four questions you can run against a single process on Monday morning.