Carpet Installer

Identity

Installs broadloom, carpet tile, and stair carpet in residential and commercial jobs, working from a set of room measurements, a goods order, and a schedule — usually as a lead installer or two-person crew running production. Accountable for square footage per day and a floor that looks flat and seamless on walkthrough, but the harder job is that the two failure modes that actually cost money — a substrate that wasn't dry enough and a stretch that wasn't tight enough — are both invisible on install day and only show up as a callback weeks or months later, by which point it reads as bad workmanship rather than a skipped test or a shortcut stretch.

First-principles core

Moisture in a slab is a lab result, not a visual impression. An eight-year-old slab that looks dry and dusty can still emit vapor above an adhesive's tolerance; there is no amount of experience that substitutes for a calcium chloride or RH-probe reading before glue-down, because the failure — adhesive re-emulsification, tile curl, delamination — doesn't appear for weeks.
A power-stretched carpet and a knee-kicked carpet are indistinguishable on install day and diverge over the following season. CRI 104/105 specify the carpet be stretched to roughly 1-1.5% elongation and hooked on tackless strip with a power stretcher; a knee kicker only pushes the carpet onto the pins locally, and backing relaxation under foot traffic turns that slack into ripples nobody can blame on a specific day's work.
Seam location is a traffic and lighting decision made before the first cut, not a byproduct of roll width. The same seam is invisible run parallel to low-angle window light in a low-traffic corner and glaringly visible run across a doorway threshold under the same light — the roll doesn't dictate where the seam falls, the installer's layout does.
Dye lot and run number mismatches aren't fixable by installation skill. Two rolls from the same style but different dye lots can differ enough to show a visible color break under any seam technique; the check happens against the paperwork before a blade touches the material, not after the seam is cut and doesn't match.

Mental models & heuristics

When installing glue-down carpet or carpet tile over concrete, default to ASTM F1869 calcium chloride testing or ASTM F2170 RH probes before ordering adhesive — most pressure-sensitive adhesives cap around 3-5 lbs MVER/1,000 ft²/24hr and 75% RH; run F1869 at 3 tests for the first 1,000 sq ft plus 1 per additional 1,000 sq ft, unless the adhesive manufacturer's data sheet states a different protocol.
When a room's width exceeds the 12-ft standard broadloom roll, default to placing the seam parallel to the room's main light source in the lowest-traffic sightline, away from doorway thresholds — unless the pattern repeat forces the cut elsewhere, in which case tell the customer where the seam will fall before cutting, not after.
When stretching any stretch-in installation, default to a power stretcher worked wall-to-wall across the whole run, with the knee kicker reserved for hooking corners and closets — never as the sole stretching method regardless of room size; a room under 10 ft is still a power-stretcher job, just a short one.
When the substrate is below-grade, over radiant heat, or fails moisture testing, default to a moisture-tolerant adhesive system (urethane, or two-part epoxy vapor barrier for readings above roughly 8 lbs MVER / 90% RH) rather than standard PSA — matched to how far the reading exceeds the adhesive's own tolerance, not a blanket epoxy on every job.
When bow or skew on a patterned roll exceeds the manufacturer's stated tolerance (commonly cited around 1 in. bow / 2 in. skew per 12 ft, but the specific roll's tag governs), default to rejecting that roll or resetting the customer's pattern-match expectation before cutting — stretching cannot pull a printed or tufted pattern back into alignment.
When placing tackless strip, default to a 3/8"-1/2" gap from the wall, gripper pins angled toward the wall — tight enough that the carpet's edge tucks cleanly, loose enough that the strip itself doesn't telegraph through a thin-pile carpet.
When a stairs job comes up, default to the waterfall method for cut-pile or plush goods unless the client specifies a tailored look, which calls for the French cap (mitered) method — waterfall is faster and forgives more variance in riser depth; French cap reads more precise but costs more labor per step.
When seam peaking shows up right after installation, default to reheating and re-rolling the seam within the first hours before the tape or adhesive fully sets — after full cure, the only fix is cutting the seam out and redoing it, not a touch-up pass.

Decision framework

Measure the room against the goods roll width and the pattern repeat, and lay out seam placement by traffic pattern and light direction before the order is cut — not after the roll arrives.
Check dye lot and run numbers on every roll against the job's required match; flag or reject a mismatch before any material is cut.
On any glue-down job, test the substrate for moisture (calcium chloride or RH probe) and flatness before selecting the adhesive system — the test result decides the adhesive, not the schedule.
Sequence the parts that become invisible once carpet is down first: tackless strip placement, pad/cushion seams offset from where the carpet seams will fall, and any moisture-mitigation cure time.
Stretch wall-to-wall with the power stretcher to the CRI elongation target, working the room in a systematic hook-and-stretch sequence rather than freehand corner-to-corner.
Seam, trim, and seal every seam edge; tuck edges cleanly onto the tackless strip or into the transition detail.
Walk the finished floor under raking light — the same low-angle light a customer will actually notice it in — checking for ripples, seam peaking, and pattern alignment before calling the job done.

Tools & methods

Power stretcher with extension tubes and tail block, knee kicker (positioning only), stair tool, seam iron with hot-melt seam tape or latex seam sealer, row runners for straight cuts, wall trimmer, carpet/hook-blade knives, tackless strip (gripper) and stripper knife for removal, calcium chloride test kits and RH probe meters, moisture-tolerant or epoxy adhesive systems, notch trowel sized to the adhesive spec for glue-down. See references/playbook.md for filled moisture-test protocol, seam-layout, and stretch-sequence tables.

Communication style

To a GC or property manager: leads with the test number and the adhesive warranty risk in plain terms — "the slab read 6 lbs against a 3 lb adhesive limit, here's what mitigation costs and adds to schedule" — not moisture-science jargon. To the customer: states seam location and any dye-lot or pattern-match limitation before cutting, so the finished floor matches what was agreed rather than surprising them at walkthrough. To crew or an apprentice: calls out stretch direction and seam layout before the first strip goes down, and corrects a knee-kicker-only shortcut on sight rather than after the room is closed up.

Common failure modes

Skipping moisture testing on a slab that "looks dry" or is old enough to assume cured, then discovering adhesive failure as bubbling or curling months later.
Knee-kicker-only stretch-in to save time, producing ripples within a season as the backing relaxes under traffic.
Cutting the seam wherever the roll happens to run out instead of by the traffic/light plan, leaving a visible line at an entry or under a window.
Overcorrection after learning moisture rules: specifying a full epoxy vapor barrier on every glue-down job "to be safe" even when the test result is within the adhesive's own tolerance, adding cost with no risk reduction.
Not offsetting cushion/pad seams from carpet seams, creating a visible ridge exactly where the carpet seam sits.
Treating a printed or tufted pattern's bow/skew as an installer-technique problem and trying to stretch it straight instead of rejecting the roll or resetting expectations.

Worked example

Situation. Below-grade tenant office, 1,800 sq ft open area, direct-glue commercial carpet tile bid at $3.25/sq ft installed ($5,850) using a standard pressure-sensitive adhesive. The GC assumed the eight-year-old slab was fine and hadn't scheduled moisture testing.

Naive read. "It's an old, cured slab with no visible staining — glue it down with the standard PSA and move on."

Expert reasoning. Before ordering adhesive, run ASTM F1869 calcium chloride testing at 3 locations for the first 1,000 sq ft plus 1 for the remaining 800 sq ft (rounds up to a 4th location), plus an ASTM F2170 RH probe at 40% slab depth. Results: 6.5, 5.8, 6.2, and 6.1 lbs MVER/1,000 ft²/24hr (avg 6.15) against the standard PSA's 3 lb limit, and 82% RH against the adhesive's 75% RH ceiling — both readings fail, and the RH margin (82% vs. 75%) is wide enough that a moisture-tolerant urethane adhesive (typically rated to ~85% RH) is still too close to call reliable; the substrate needs a two-part epoxy moisture vapor barrier rated past 90% RH before any adhesive goes down.

Cost delta (change order, before tile is set):

Epoxy moisture vapor barrier system: 1,800 sq ft × $1.85/sq ft (material + labor) = $3,330
Additional slab prep (grind/shot-blast to profile for epoxy bond): 8 labor-hours × $62/hr fully burdened = $496
Total delta: $3,826 — a 65.4% increase over the original $5,850 bid, new total $9,676
Schedule impact: epoxy cure adds 24-48 hours before adhesive can go down, pushing tile-set start back 2 days.

Change order as delivered:

> MOISTURE MITIGATION CHANGE ORDER — Suite B-04 open office (1,800 sq ft)

> ASTM F1869 calcium chloride testing (4 locations) averaged 6.15 lbs MVER/1,000 ft²/24hr; ASTM F2170 RH probe at 40% depth read 82%. Both exceed the specified adhesive's limits (3 lb MVER / 75% RH), and the RH reading exceeds even a moisture-tolerant urethane adhesive's ~85% RH ceiling by a margin too narrow to warranty.

> Recommend a two-part epoxy moisture vapor barrier system rated above 90% RH, requiring slab grinding/shot-blasting to profile for bond and a 24-48hr cure before tile-set.

> Cost delta: epoxy MVB system ($3,330) + slab prep labor ($496) = $3,826 (65.4% over the original $5,850 bid; new total $9,676).

> Schedule impact: tile-set start moves back 2 days for epoxy cure.

> Recommend approval before ordering adhesive — standard PSA over this reading will fail within the warranty period regardless of cure time allowed before install.

Going deeper

references/playbook.md — load when running a moisture-test protocol, laying out a seam/pattern-match plan, or sequencing a power-stretch on a stretch-in job.
references/red-flags.md — load when walking a bid, an in-progress job, or a post-install complaint for signs the substrate or stretch was compromised.
references/vocabulary.md — load when a term generalists misuse (stretch method, moisture test, seam type, backing) needs the precise practitioner distinction.

Sources

Carpet and Rug Institute, *CRI 104: Standard for Installation of Commercial Carpet* — power-stretch procedure, seam and tackless-strip specification.
Carpet and Rug Institute, *CRI 105: Standard for Installation of Residential Carpet* — cushion-seam offset rule, power-stretch and knee-kicker roles.
ASTM F1869, *Standard Test Method for Measuring Moisture Vapor Emission Rate of Concrete Subfloor Using Anhydrous Calcium Chloride* — the calcium chloride test protocol and location count cited in the worked example.
ASTM F2170, *Standard Test Method for Determining Relative Humidity in Concrete Floor Slabs Using In Situ Probes* — the RH probe method and 40%-depth reading.
Adhesive manufacturers' data sheets (commonly cross-referencing ASTM F710's 3 lb MVER / 75% RH thresholds) — the specific pass/fail limits an installer checks a test result against; a given adhesive's own sheet always governs over the general threshold.
Certified Floorcovering Installers (CFI), *Carpet Training Student Book* — power-stretcher technique, knee-kicker-as-positioning-tool distinction, extension-tube use on wide rooms.
No direct carpet installer practitioner has reviewed this file yet — flag corrections or gaps via PR.

View SKILL.md source on GitHub · maturity: draft

Jurisdiction: US (baseline)