I run punch lists and callbacks for a mid-size remodeling outfit. Twelve years, somewhere north of 300 emergency fixes — the kind where a lead calls at 4 p.m. and the walkthrough is at 8 the next morning. So I'll say this straight out: most "bad screw" complaints are driver-bit complaints wearing a costume.
And here's the other half of that, because it's the part people skip: there's a real slice of work — call it 15 to 20 percent — where the fastener spec is the entire job, and no bit, no driver, and no amount of technique will save you. Knowing which side of that line you're standing on is most of the skill.
I've been wrong on both sides of it. Here's what twelve years of callbacks taught me.
The screw is usually innocent
In my role running callbacks for a 60-person remodeling company, the first thing I look at on a stripped-head complaint is never the screw. It's the bit in the driver.
Three things, roughly in order of how often they're the culprit: bit size, bit condition, drive angle. In that order.
In my first year I did the classic rookie thing: grabbed whatever bit was already chucked in the impact driver and went to town on about 140 feet of baseboard. Cam-out on maybe thirty heads, two of them buried so deep I had to dig them out with a chisel and plug the holes. Cost me a Saturday and roughly $180 in replacement material. The screws were fine. I was not.
People think a stripped head means the screw was soft. Usually it means torque got delivered at a slight angle, through a worn bit, in a size that didn't quite match the recess. The screw takes the blame because it's the thing you can see afterward.
Here's the causation reversal that took me longer to accept than I'd like to admit: upgrading to a harder, more aggressive screw can make your failure rate worse, not better. The assumption is that a stronger screw strips less, so stronger must be better. The reality is that the screw stops being the weak link and the substrate becomes one. Now instead of a chewed-up head you've got a split cabinet stile or a blown-out hinge plate, and that's a much more expensive morning. Same energy, directed somewhere worse.
This is the actual reason star drive works. GRK Fasteners' Star Drive trim head construction screw uses a six-lobe recess, which gives you more contact points and spreads the load instead of concentrating it on four shallow corners. It genuinely cuts cam-out. But it only does that if the bit matches — and "matches" means you're not running a T15 in a T20 recess because the T20 is in the truck.
Which is why the humble screwdriver multi bit earns its place on your belt. Not for speed. For the fact that when five sizes are right there, you stop making the size you have work.
Where a 1/2 inch drive click torque wrench actually earns its keep
People buy torque wrenches because they want things tighter. That's the assumption. The reality is that torque instruments exist to stop you from over-tightening things that have a published value — they're a ceiling, not a floor.
A 1/2 inch drive click torque wrench belongs on threaded fasteners with a specified torque: structural steel bolting, trailer hitches, suspension work, machine assemblies, pump flanges. Typical 1/2" drive click wrenches cover roughly 20–150 or 50–250 ft-lb depending on the model. A properly calibrated click type holds about ±4% of the setting — check your own certificate rather than trusting the marketing copy.
On calibration: common practice is a 12-month interval or 5,000 cycles, whichever comes first, per ISO 6789-2:2017. The underlying tool standard in the US is ASME B107.300 (Torque Instruments). Neither of those is optional if the number matters.
What a torque wrench does not belong on is a wood screw. There isn't a published torque value to hit. Installation is governed by the manufacturer's instructions and by the fastener seating — the screw is home when the head bears. If you put a 1/2" drive wrench on a #9 trim head and start pulling, you won't torque it. You'll pull the head straight through the trim and into the drywall, and then you'll be calling me.
I've watched guys do it. Twice. Both times the reasoning was "it's a structural screw, so I should torque it like a bolt." That's the same word meaning two different things, and it costs people real money.
Match the fastener to the joint — and the nailer to the trim
Since people keep asking me what nail gun is used for trim, let me just put it here. Five guns cover essentially all of it:
15-gauge angled finish nailer, which is the default answer and what I'd buy first. 16-gauge straight finish nailer for tighter spots and where the angle doesn't work. 18-gauge brad nailer for small trim, cabinet face frames, and anything under about 3/4". 23-gauge pin nailer for delicate returns, pre-finished material, and mitered corners you don't want to fill. And a crown stapler for the plywood and cabinet-back work nobody ever sees.
Nails have their place and I'm not going to pretend otherwise. But anywhere the joint is going to get pulled apart — cabinet boxes into studs, exterior trim that moves with the seasons, anything a kid can hang off — you want a screw, not a nail.
For general work, GRK Fasteners GRK R4 multi-purpose screws are what's in most of our trucks. Framing touch-ups, blocking, cabinets, general construction, exterior work. Trim head screws are for the finish side — small head, countersinks flush, hole gets filled and forgotten. Different jobs, different fasteners, and mixing them up is how you get a callback.
One thing to be clear about, because I've watched it cause a failed inspection: "multi-purpose" is not the same word as "structural." Neither is "trim head." If the plan calls out a rated connection or an evaluation report, you need a screw that actually carries that report.
Where the fastener IS the whole job
Here's the 15 to 20 percent. This is the part where I stop being clever about bits.
If any of these are true, the fastener spec is your entire job and technique is irrelevant:
- The plan or the permit references a rated connector or an evaluation report number
- The connection is in the load path — ledgers, headers, hangers, anything holding something up
- You're into engineered lumber, LVL, or a engineered wood product with its own design values
- It's exterior, or it's in contact with today's copper-azole treated lumber, which eats standard zinc coatings alive
- An inspector has to sign off on it
In March 2024 I had a job where the plan called out a specific rated fastener for a deck ledger and one of our guys had swapped in a general-purpose screw because the box looked similar. The inspection was at 2 p.m. We had two hours. Normally I'd call the designer, get the submittal, and verify against the evaluation report. There was no time for that, so I pulled the spec sheet off my phone, matched it to the ESR, and drove forty minutes to the supplier. We made it. Barely. But that's a decision I made with incomplete information and I got lucky.
In hindsight I should have made the call at the truck instead of at the jobsite. The rule I've used since: if it's in the load path, nobody touches the fastener without the spec sheet in hand.
You can look up evaluation reports at icc-es.org — that's where the ESR numbers live, and they get revised, so pull the current one, not the one from the box your supplier has had on the shelf since 2022.
"But I've used the same cheap screws for 20 years"
To be fair, that's a real data point. I've heard it from guys who do beautiful work.
The reason it works for them is usually that they pre-drill, they don't over-torque, and they're driving by feel — which is exactly the point I'm making. They're already doing the technique part right. The screws get credit for something the operator is doing.
Two caveats. One, some of it is survivorship. The jobs that failed aren't the ones you remember. Two, the substrate changed under everybody. Today's treated lumber formulations are noticeably more corrosive to standard zinc coatings than what was going in twenty years ago, so "it always worked" is partly a statement about the lumber, not the screw.
What I tell my crew now
Check the bit before you blame the screw. Replace worn bits like you replace blades — they're consumable and they cost less than the callback. Pre-drill the ends of trim and anything within two inches of an edge. Know which screws in your truck are finish fasteners and which ones are rated, and keep them in different boxes.
And buy the torque wrench for the bolts. Leave it out of the wood.
Bottom line: the driver causes most of the problems people blame on the fastener — but "most" isn't "all," and the difference between a good carpenter and a great one is knowing which job they're on before they pull the trigger.
