Drivers are shapes, not numbers.
Ask any operator what a supplier's lead time is and they'll pause. It depends. That pause is the whole insight — the number was never a number.
Every plan rests on a handful of inputs: how long supply takes to arrive, how much customers will buy, how much of what you make is good, what it all costs. Planning systems ask for each of these as a single number. The trouble is that none of them is a single number, and everyone who works with them already knows it.
The number was always a shape
Ask a buyer what the lead time on a part is. The honest answer is never "fourteen days." It's "usually around two weeks, but it ran to five once when the port backed up, and it can come early if we're lucky." That answer is a shape: a cluster of likely values, a long tail toward the bad outcomes, an occasional pleasant surprise. The single number you eventually typed into the system was a summary of that shape — and the summary threw away the part that mattered.
The same is true everywhere. Demand is a shape: a likely middle, a chance of a surge, a chance of a slump. Yield is a shape: mostly fine, with bad batches in the tail. Cost is a shape: stable until it spikes. Treat any of them as a fixed number and you haven't simplified the problem — you've deleted the risk from it and then planned as if the risk weren't there.
Stop asking "what is the lead time?" and start asking "what does the lead time look like?" The first question has a misleading answer. The second has an honest one — a distribution, with a body and a tail.
The average hides the expensive days
The most dangerous summary is the average, because it feels so reasonable. If lead time averages two weeks, plan for two weeks — what could be wrong with that? What's wrong is that the average is a quiet day. The days that cost you money are the tail days: the one delivery in twenty that arrives three weeks late, the one batch that yields poorly, the one week demand spikes past anything you stocked for. Those days don't move the average much. They do most of the damage.
Plan on the mean and you are, by construction, unprepared for exactly the events that hurt. You'll be on time for the ordinary delivery and caught flat by the late one. The mean optimizes for the day nothing happens. The business pays for the days something does.
The average is the day you didn't need a plan. The tail is the day you did.
Why the shapes have to travel together
There's a second reason single numbers mislead: the shapes are connected. The week demand surges is often the week lead times stretch and the spot cost climbs — the same disruption pushes several drivers at once. Plan each input on its own average and you miss the pile-up entirely, the bad case where three things go wrong together. Keeping the drivers as shapes, and keeping their relationships intact, is the only way to see that pile-up before it arrives.
This is where resilience starts
Once you accept that drivers are shapes, the rest of the category follows naturally. Mapping what those shapes imply, together, is the diagnosis. Choosing one plan that holds up across the whole spread of them — paying real attention to the tail rather than the average — is the decision. Everything VYAN does begins here, with the refusal to flatten a shape into a number that was never going to survive contact with a real week.
Drivers are distributions. The body of the distribution is the easy case; the tail is the expensive one. Plan for the shape, not the average — because the average is the day that never needed you.