Most turnaround managers obsess over mobilization. Getting people in the gate, getting them productive, hitting Day 1 milestones. Fair enough. Those first few days set the tone for everything that follows.

But here's what I rarely see anyone obsess over: getting people out.

I've managed demobilization from both sides. As an owner watching costs pile up in the final week. As a contractor trying to redeploy my people to the next job while the owner keeps asking for "just one more day."

The costs are real. And they're almost always invisible until someone in Project Controls starts closing out the event.

Here's a scenario I've seen play out dozens of times:

A 25-day turnaround finishes its critical path work on Day 22. Mechanical completion. Operations is happy. But 200 contractors are still on site "finishing up" and "standing by for punch list."

Those 200 workers cost roughly $150 per day in per diem. That's $30,000 a day. Over three days, you've burned $90,000 in per diem alone for workers who completed their scope.

But per diem is just the visible cost. Every direct worker drives indirect costs. Supervision, safety support, admin staff, and field engineers. You don't cut one superintendent because 15 pipefitters left. But if you're carrying 200 extra workers, you're also carrying 25-30 support staff who could have ramped down days earlier.

Those indirect costs run $75,000 to $200,000 per week on a mid-sized turnaround. Most of it never gets attributed to the demobilization failure. It just shows up as "labor overrun" in the final cost report.

Demobilization chaos isn't contractor malice. It's the same root cause as mobilization chaos: nobody owns the process.

Coordinators are focused on execution. Safety is focused on not having an incident in the final days. The turnaround manager is already thinking about startup. Not enough people are looking at the workforce curve and asking: who should have left yesterday?

And there's a natural bias toward keeping people. What if we find something? What if startup hits a snag? The cost of having someone and not needing them feels abstract. The cost of needing someone and not having them feels catastrophic.

So workers stay. Per diem accumulates. Indirect support stays fully staffed. And the event runs long not because the work ran long, but because nobody managed the exit.

First, assign ownership of demobilization before Day 1. Not "someone will figure it out later." A named person who tracks the workforce curve against the scope curve.

Second, require departure dates in the plan. Every contractor should know when their scope completes and when their workers are expected to leave. Build it into the contract.

Third, get daily visibility on who's still on site versus who's still producing. When those numbers diverge, you have a cost problem. The earlier you see it, the earlier you can act.

This is exactly why we built demobilization tracking into MODEMO®. Not because demobilization is complicated. Because it's invisible until it's expensive.