You know your operation is not running the way it should, but you do not yet know where the break is, and the break decides the work.
Operational trouble comes in six shapes that need completely different fixes: the work is invisible and lives in people's heads, the work is visible but not repeatable so outcomes swing with who does it, the whole thing rests on one irreplaceable person, the same problems keep coming back because nothing catches them, the technology is wrong or unused, or the data cannot be trusted. This sorts out which one you have, confirms it is really an operations problem, tells a one-off breakage from a recurring pattern, and hands you a plan. It is a sorting instrument, not a fix. It takes about ninety minutes and asks for honesty about how much of your operation lives in people's heads and a willingness to name the problems that keep coming back.
If something is actively on fire today, a key departure mid-week, a payroll that will not run, a system down, handle that first. This is the work you do once the fire is out, so the next fire is less likely. Come back when you are steady enough to look.
You cannot diagnose an operation you have not laid out. Name the handful of things your organization does over and over to deliver its mission and keep the lights on, and for each, mark how much of it is written down versus living in someone's head. Then write down where it keeps breaking, and mark anything that has happened more than once, because a recurring problem matters most.
First, make sure you are in the right place. Then place your trouble in one of the six shapes, because the shape decides the guide.
Good, you are in the right place. Answer each statement the way it really is. There is no shame in a low score; that is the whole point of looking.
Two honest checks separate a well-run operation from one that only believes it is. First, do your systems actually exist and get used, or do you have binders no one opens? For each system you believe you have, ask for the proof: the document a new hire could actually follow, the tool people actually use, the data everyone actually trusts. Where you cannot produce it, treat the system as not real. Second, is the problem in front of you a one-off, or the latest instance of something that keeps coming back? That second question is the important one, because a recurring problem is not really about the problem, it is about the fact that nothing in your organization catches and fixes it.
| What you believe you have | Real and used / On paper only / Absent |
|---|---|
| A process you think is documented | |
| A tool you think you use | |
| Data you think you trust |
Then whatever else you found, add Build a Way to Find and Fix What Breaks to your plan, and consider making it first, because without it every other fix you make will quietly erode. This is a route, not a stop.
Turn the finding into an ordered plan that names your first guide and follows the order operations are actually built: see the work, make it repeatable, make it survive people, build a way to improve it, get the tools right, own the data. Where you have more than one, build in that order, because you cannot standardize work you cannot see, and you cannot choose the right tool for a process you have not defined. Give the first guide a real start date, and if you have nothing written down, protect against the worst case by documenting your single most critical process this week.
List the guides your finding names, in the required order. If problems keep recurring, Find and Fix may jump to first.
Your answers stay in your browser on this page and are never sent anywhere. Use Save or print to keep your copy.
Two moments here reach past these pages. If the operational problem is really that you, the leader, are the single point of failure and are running on empty holding it all together, treat that as its own priority and lean on a trusted person, because no system survives its builder breaking. And if a piece of what you find is a genuinely technical job, migrating data off a failing system, securing sensitive information, an outside technology professional is the right help, and this work will point you there when you reach it. Naming these is not the guide failing. It is the guide being honest about its edge.
Once the operation is on firmer ground, it is worth stepping back to look at the whole organization, not just the part that was breaking. There is a short whole-organization check-up waiting for you when you are ready. That is not a sales pitch. It is the next honest thing to look at. The whole-organization check-up →