Performance management in the trades: what works when nobody is sitting at a desk

Annual reviews do not work in roofing. I have watched companies burn thousands of hours on quarterly forms nobody reads, scoring crew members on competencies they were never taught, then acting surprised when the foreman quits. The HR playbook was built for people who sit at desks. My crews do not.

TL;DR: Performance management for field workers is broken because it copies office HR. What actually moves work is three loops: a daily five-minute conversation between foreman and crew member, a weekly safety and quality scorecard the whole crew sees, and a quarterly career check-in. No annual reviews. No productivity dashboards. No paperwork bloat.

Why the office playbook fails on a roof

The standard performance management system has three parts. Annual review. SMART goals set in January. Mid-year check-in. All of it written down, all of it stored in an HRIS, all of it referenced exactly once a year when the raise conversation happens.

That system assumes the work is invisible. It assumes a manager cannot see what their direct report does every day, so the manager needs a quarterly mechanism to find out. On a roof, the work is the most visible thing in the world. The foreman watches the crew member install flashing in real time. There is no information gap to close with a form.

The system also assumes the worker reads. Not literacy. The office habit of reading email, reading goals, reading documentation. Field workers do not live in a document workflow. They live in conversations, radio chatter, and the quality of the work in front of them. A PDF goal sheet they sign in January has zero effect on what they do in July.

Worst of all, the system runs on a 12-month clock. By the time annual review feedback lands, the project is closed, the crew has rotated, and the moment that mattered is six months in the rearview. Telling someone in November that their layout work in May was sloppy is useless. They cannot fix May.

The productivity metric trap

The first thing every well-meaning operations person tries is squares per day. Or linear feet of seam per hour. Or man-hours per square. The logic is clean. Measure output, reward the high performers, coach the rest up.

Within a week, the crew has gamed it. They start with the easy sections. They skip details that do not show on the metric. They stop helping the slower guys because helping drops their personal number. The numbers go up. The quality drops. Callbacks climb. By month three, you have a productivity dashboard that looks great and a customer base that is furious.

This is not a character flaw. It is what any rational person does when you tell them their job depends on a number. The metric becomes the target, the target becomes the work, and the actual work, building something that does not leak for 25 years, falls off the page.

I track productivity at the company level for bidding and forecasting. I do not track it at the individual crew member level for performance. The two uses are not the same.

Loop one: the daily five minutes

This is the load-bearing piece of the whole system. Every day, before the crew breaks for lunch or at the end of the shift, the foreman has a five-minute conversation with one or two crew members. Not all of them every day. One or two, rotating, so each person gets a real conversation roughly twice a week.

The conversation has three parts. What went well today. What got in your way. What you want to learn next. That is it. No form, no rating, no follow-up email. The foreman remembers it or writes one line in a notebook.

The reason this works is that feedback delivered the day the thing happened lands. The crew member still has the muscle memory of the cut they made, the angle they set, the call they made about the deck condition. Telling them right then is not criticism. It is coaching. Telling them four months later is grievance airing.

The foreman has to be trained to do this. Most foremen got promoted because they were the best installer on the crew, not because they were a coach. The first time I ask a new foreman to run these conversations, half of them just give safety reminders or task assignments. That is not coaching. Coaching is asking a question and listening to the answer. I run roleplay sessions for foremen on this. It feels awkward the first three times. By the tenth time it is reflexive.

Loop two: the weekly scorecard

Every Friday afternoon the foreman fills out a one-page scorecard on the crew. Not on individuals. On the crew as a unit. Four items, no more.

1. Safety. Zero, one, or two near-misses. Any first-aid event. Any stop-work call.
2. Quality. Punch list items generated this week. Anything the QC walk caught that should have been caught by the crew.
3. Schedule. Are we ahead, on, or behind the production target the estimator built into the bid.
4. Crew health. One sentence on morale, conflict, or anything the office should know.

That sheet goes to me. I read every one of them. Not because the foreman cannot handle it, but because patterns only show up across crews. If three crews report the same supplier issue, that is not a crew problem, that is a procurement problem I need to fix. If one crew is always behind schedule but always quality-clean, that is an estimating problem, not a crew problem. The scorecard is how I learn what the field is actually fighting.

The crew sees the scorecard too. It gets read out at the Monday morning huddle. Public visibility on safety and quality at the crew level creates the right kind of pressure. Nobody wants to be the crew that put two near-misses on the board.

Loop three: the quarterly career check-in

Once a quarter, every crew member sits down with their foreman, or with me if they are a foreman, for 30 minutes. Off the roof, off the clock if it has to be, paid if at all possible.

The agenda is three questions. Where do you want to be in two years. What is one skill you want to add this quarter. What is one thing the company can do better.

That last question is the one most companies skip and it is the most valuable. The crew knows where the bleed is. They know which supplier sends bad material. They know which superintendent is making bad calls. They know which truck is going to break down next. They are not going to tell you any of this on an anonymous survey because they have learned that nothing happens. They will tell you face to face if you actually act on what you hear.

I keep a list of every thing a crew member has told me in a quarterly. Next quarter, I show them what we did about it. If we did nothing, I tell them why. That is the only way the question keeps producing real answers.

When you actually need paperwork

Documentation matters in three cases and basically nowhere else.

First, when behavior crosses into safety violation. Refusing fall protection. Coming to work impaired. Horseplay near an edge. That gets written up the day it happens, signed by the crew member, copy to HR. No exceptions, no warnings, no second chances at the documentation step.

Second, when you are building toward termination. The day you start thinking about firing someone is the day you start documenting. Date, behavior observed, conversation had, expectation set. If you have not been documenting and you suddenly need to fire someone, you do not have a case, you have a lawsuit waiting.

Third, when someone is on a promotion track. If I am moving a crew member toward foreman, I document the skills they have demonstrated and the gaps they still need to close. That document becomes the development plan. It is not a performance review. It is a roadmap.

Outside those three cases, the paperwork is overhead with no payoff. Stop writing it.

How to deliver critical feedback without losing the person

The hardest part of any of this is telling someone good their work was not good. Field workers take pride in their craft. Criticize the work badly and they hear you criticizing them, and you have lost them.

Three rules. Be specific. Be fast. Be private.

Specific means the exact piece of work, not a personality trait. Not “you have been sloppy lately.” Instead, “the cut on that scupper was rough and we are going to have to redo the flashing.” The first one is a verdict on the person. The second one is a fact about the work.

Fast means the same day, not the next week. The longer you wait, the more it feels like you have been building a case against them. Same-day feedback is coaching. Saved-up feedback is an ambush.

Private means not in front of the crew. The temptation to make an example of someone in front of the team is real and it is always wrong. You lose the person you corrected and you scare everyone else into hiding mistakes from you. The crew should see you supporting people in public and correcting them in private.

The practical takeaway

If you are running a field operation and you want to fix performance management this quarter, here is the order.

1. Kill the annual review. Tell your team you are doing it and replace it with the three loops. They will be relieved.
2. Train your foremen on the daily five-minute conversation. Roleplay it. Have them practice it on each other. Make it a job expectation, not a nice-to-have.
3. Build the one-page weekly scorecard. Four items, no more. Read every one. Act on the patterns.
4. Schedule the quarterly career check-ins on the calendar today, for every crew member, for the next 12 months. If it is not on the calendar it will not happen.
5. Build a documentation discipline only for safety, termination track, and promotion track. Burn everything else.

This is not soft. This is the harder version of performance management because it requires the foreman to actually know their people and the leader to actually act on what the field is telling them. It is also the version that retains crew, builds craft, and stops the callbacks. The paperwork version did none of that.

If you want to talk through what this looks like inside your operation, or you want me to come speak to your leadership team about it, book a conversation.