How to Run a Skills Drill That Techs Do Not Resent

Why this matters

Drills fail for social reasons more often than technical ones. A crew that has sat through one condescending, badly-run drill will treat every future one as theater, and once that judgment is made it is very hard to reverse. The cost is not the wasted hour. It is that the one category of skill that can only be maintained by drilling, the rare and consequential kind, quietly rots, and you find that out on the day it matters. Getting drills right is mostly about removing the four specific things that make experienced people resent them.

The four reasons techs resent drills

Name them before designing anything, because every step below exists to remove one:

1. It drills what they already do daily. Nothing tells a crew you do not respect their skill faster than rehearsing something they perform eight times a week.
2. It is a test with consequences wearing a drill costume. If a poor showing lands in a file or a review, people stop participating and start protecting themselves, and you learn nothing real.
3. It happens on their time. Before shift, through lunch, at the end of a long day. This one is not a perception problem, it is a legitimate grievance.
4. It is run by someone who could not pass it. An owner or office lead running a hands-on drill they have never performed loses the room in the first two minutes.

Step 1: Drill only what is rare and consequential

The only skills that require drilling are the ones the work does not maintain by itself. Filter every candidate through both questions:

• Does normal work provide reps? If yes, it does not need a drill. It needs a standard and a spot check.
• Is the consequence severe? Injury, fatality, major property damage, a customer-safety event.

Only the rare-and-severe intersection earns drill time. Emergency shutdown sequences, a gas-odor response, a rescue from height, a water-plus-electricity entry decision, an equipment failure that has to be handled in seconds. See related: Spaced Practice and Why One Training Day Does Nothing.

Everything else that feels drill-shaped is usually a checklist problem or a reps problem, and drilling it burns the crew's patience on the wrong target.

Step 2: Make it a rehearsal against a standard, not a quiz

A quiz asks whether you know. A rehearsal asks you to perform the sequence until it is automatic. Design it as the second.

• Announce the standard in advance. "The shutoff is reached and closed inside 60 seconds, from wherever you are standing." No surprise criteria. Hidden standards are what make it feel like a trap.
• Publish the scenario type ahead of time, even if not the exact scenario. People rehearse beforehand, which is a good outcome, not cheating.
• Run it more than once in the session. Whoever goes first is at a disadvantage, and a single rep is a measurement rather than practice.

The framing sentence worth saying out loud at the start: we are not finding out who knows this, we are making sure everyone can do it without thinking. That sentence, said honestly and backed by step 6, does more for participation than any other single move.

Step 3: Pay for it and put it on the schedule

Drills go on the clock, in working hours, on the posted schedule with the same weight as a job. If a drill can be bumped for a service call every time things get tight, it will be bumped every time, and the crew will read that correctly as the shop's real priority.

Keep it short enough to defend. Twenty minutes is a realistic ceiling for a shop that has to protect billable hours, and 20 minutes done four times a year on one topic beats a two-hour annual session by a wide margin, because the value is in the repetition spacing.

Step 4: The person running it goes first, cold

Whoever calls the drill performs it first, unrehearsed, in front of everyone. This single move removes most of the resentment, and it is the one most often skipped.

It works for three reasons. It proves the drill lead can do the thing. It sets a realistic bar, because their cold run will be imperfect and everyone will see that imperfection is expected. And it converts the exercise from an evaluation into something the whole crew including the lead is doing together.

If the owner cannot perform the drill, the owner does not run it. Hand it to whoever can, and participate as a participant.

Step 5: Run it in real conditions, with the safety action leading every rep

Rehearsing at a table teaches recall. Rehearsing in the actual space, with the actual gear, in actual clothing, teaches performance. Move it to a real environment wherever it can be done safely.

Every hazard drill starts with the protective action, and the sequence never varies between reps:

• Energized equipment: de-energize, lock and tag, verify dead with a meter proven on a known live source before and after the reading. The rep is the whole sequence, not the meter reading alone. The failure being trained out is a meter nobody proved, which reads zero on a dead circuit and on a dead meter alike.
• Gas odor: stop work, operate nothing electrical including a phone or a switch, evacuate to distance, shut off at the exterior valve only if reaching it is safe, call from outside. No diagnosis inside the building, ever, at any rep.
• Stored pressure or mechanical energy: isolate, relieve to zero, verify at a gauge or bleed point, then open. Never open on the assumption it was relieved.
• Height and rescue: anchor set and inspected before the first step up, and the rescue plan rehearsed with the actual equipment, because a suspended worker is a time-critical situation and a plan on paper is not a plan.
• Water plus electricity: kill power at the source and verify dead before entering standing water. The rep starts at the panel, not at the water.

Never simulate the hazard itself. Do not introduce a live circuit, real gas, real pressure, or real fall exposure to make a drill realistic. Simulate the condition, perform the protective sequence for real. A drill that injures someone has failed at the only thing it was for.

Step 6: Score the system on the first run, the person only after

On the first run of any new drill, the finding belongs to the shop, not to the individual. If three of five techs cannot find the shutoff in 60 seconds, that is a labeling, layout, or standard problem before it is a competence problem.

Debrief in that order: what about the setup made this hard, then what about the sequence was unclear, then what individuals should do differently. Most first-run failures resolve at the first two.

Keep drill results out of performance reviews. The moment a drill has career consequences, you stop measuring capability and start measuring who prepared for the drill. If someone genuinely cannot perform a critical sequence after repeated drills, that is a separate, real conversation held privately, not a drill outcome posted on a board.

Step 7: End with exactly one fix, owned and dated

Every drill produces one change, with a name and a date on it. One, not a list. A drill that produces a list of eight improvements produces zero improvements, and the crew notices by the third drill that nothing ever changes.

The fix can be tiny: relabel a valve, move a piece of equipment to a reachable spot, correct one step in the written sequence. Small and done beats comprehensive and pending, and visible follow-through is what earns you attendance at the next one.

Step 8: Rotate who designs the drill

After the first two or three, hand the design to a different tech each time. The designer has to think through the scenario, the standard, and the setup, which is more training than any participant receives. It also breaks the pattern where drills are something management does to the crew.

Worked example: a six-person shop drilling emergency shutoff

The shop has never lost anyone, and nobody has needed the emergency shutoff sequence in about three years. That combination is exactly the rare-and-severe profile, so it earns a drill. Four sessions a year, 20 minutes each, is 80 minutes per person per year, or roughly 0.04 percent of a full-time schedule. That is not a number anyone can argue with, which is the point of keeping it to 20 minutes.

Run 1, first quarter. The lead tech performs it cold first and takes about 90 seconds against a published 60-second standard, mostly because the valve access was obstructed by stored material. Then each of the other five runs it twice, so 11 reps total in the session including the lead's.

Results: 2 of 6 people beat 60 seconds, 3 landed between 60 and 90 seconds, and 1 could not locate the shutoff at all without help. Read as a system finding, not a personnel finding: 4 of 6, or 67 percent, failed the standard, and the same obstruction slowed everyone. The single fix, owned by the lead with a date one week out: clear and mark the access path. Nothing else on the list.

Run 2, second quarter. Access cleared and marked. Lead goes first cold again, 48 seconds. 5 of 6 beat the standard, and the sixth lands at 68 seconds. The single fix this time is a wording correction in the written sequence, which two people had read differently.

Run 3, third quarter. 6 of 6 beat the standard, best time under 30 seconds. Designed and run by a different tech, who introduces a variation: the drill starts from wherever each person happens to be standing rather than from a fixed spot, which is more realistic and slows two people back into the 60 to 70 second range. That is a good outcome, not a regression, because the earlier runs had been rehearsing one route rather than the skill.

What the year cost and produced. 80 minutes per person, 8 hours total across six people for the year. Two physical fixes to the site, one correction to a written procedure, and a crew that went from 33 percent meeting the standard to consistently at or near 100 percent under a harder version of the same drill. The obstruction found on run 1 had been sitting there for months and was visible to everyone every day, which is the usual way with these: the drill did not create the knowledge, it created the occasion to act on it.

What changes the answer

A crew that already resents drills from past experience. Do not open with a hands-on drill. Start by asking them to design one, and run theirs first. Giving the design away is the fastest way to reset the relationship, and you will get a better drill than you would have written.

A one-person or two-person operation. There is no group dynamic to manage, and the drill becomes a solo timed rehearsal. It still works, and it is still worth scheduling, because the decay problem is identical. Trade drills with a peer shop quarterly if you want an outside observer.

Sites with a customer present. Never drill in an occupied customer space without explicit arrangement. Use your own yard, shop, or a friendly site. A drill that alarms a customer costs you more than the training was worth.

A genuinely new hazard. Where nobody in the shop has performed the sequence, the first session is teaching, not drilling. Bring in whoever does know it, teach it properly, and start the drill cycle at the next quarter.

How to verify you got this right

• Ask a tech, privately, whether the drills are worth doing. An honest no, and the reason behind it, is the most valuable feedback available and you will only get it if scores stay out of reviews.
• Check whether the last three single fixes were actually completed. Uncompleted fixes are the fastest way to convert a working program back into theater.
• Watch for hesitation in the reps. Someone who performs correctly but visibly thinks through it has knowledge, not a reflex, and knowledge is what fails under real pressure. That means more reps, not a longer session.
• Look at who designed and ran the last three. If it was the same person every time, step 8 is not happening and you are one departure away from the program ending.

References

• OSHA, emergency action plan and employee training requirements
• OSHA, control of hazardous energy (lockout/tagout) verification practice, and fall protection rescue planning
• NFPA 70E, test-before-touch and verification of absence of voltage
• See related: How to Train Safety Without Checkbox Fatigue, Training a Team to Handle a Mistake Without Panicking