How to reduce escalations from the shop floor

You reduce escalations by sorting them first, then fixing the right ones. Not every escalation is a problem. Some are exactly what should happen, where a real safety or quality risk goes up the line. The ones to cut are the calls that came up only because the person at that tier could not make them. Build the judgment at that tier and those escalations stop.

Owners often try to cut escalations by simply discouraging them, telling people to handle more themselves. That backfires. People either keep escalating quietly or they stop escalating the things that genuinely should come up, which is worse. The goal is not fewer escalations at any cost. It is the right escalations only.

Sort your escalations before you fix anything

Every escalation falls into one of three groups:

  • Should escalate. A genuine risk or a call above this tier's authority. Keep these. Make them faster, not fewer.
  • Could decide, lacks authority. The person knew the right move but was not sure they were allowed. This is a boundaries problem.
  • Could not decide, lacks judgment. The person did not know the right move for the situation. This is a judgment problem, and it is usually the largest group.

You cannot fix escalations until you know which group yours fall into. Most factories never sort them, so they treat all three the same and fix none.

The fix is different for each group

For the authority group, set clear decide-or-escalate limits, so people know what is theirs to call. For the judgment group, you build the judgment. The call a person makes in the moment, Decision-at-the-Edge, has to be built and seen for the specific situations that keep coming up the line. The hub guide on why supervisors escalate defines it.

How to reduce them, in order

  1. Log escalations for a few weeks and sort each into the three groups above.
  2. For the authority group, write simple decision-rights limits per tier and tell people plainly.
  3. For the judgment group, name the recurring situations and find, per person, where they know the move but hesitate versus do not yet know it.
  4. Build the judgment on those situations in the flow of work, and keep logging escalations to see the right group shrink.

The judgment part is the Diagnostic Gap Model: close each person's specific gap on the calls that keep escalating. The hub guide covers it.

How you will know it worked

The total escalation count falls, but more importantly the mix changes. The "could decide" and "could not decide" groups shrink, while the genuine "should escalate" ones remain and get handled faster. That mix is the real signal, not the raw count.

Where this fits

This is one piece of building decision quality on your floor. The full picture is in the hub guide on why supervisors escalate and how to fix it. PDGMS HCD is the system that builds and proves it.

See how it works on your floor →

Related questions