Mavix Coaching · Contractor cost worksheet

Cost of Doing Nothing

What miscommunication, turnover and people problems cost your company every year, and how much new work you'd have to win to earn that money back.

The business

Sets the revenue math

A
$
B
%

If you don't know yours, use CFMA's 2025 benchmarks: 6.7% for all contractors; for commercial builders 5.9% at $10M to $24.9M, 4.7% at $25M to $49.9M, and 3.9% at $50M to $99.9M.

Rework

Work torn out and done again

C

Only rework you pay for yourself, not change orders your customer paid for. If you're a GC that subs out most of the work, you usually sit at the low setting.

Where these settings come from

The 5% typical setting is the industry's most-cited benchmark. A study of 359 projects in the Construction Industry Institute database states that direct rework costs often total about 5% of construction cost (Hwang, Thomas, Haas and Caldas, 2009).

Research tied to CII reports rework anywhere from about 2% to 20% of cost, depending on the type of project and how carefully rework is tracked. The 3% and 8% settings are reasonable low and high points inside that range. They are judgment calls, not published figures.

Sources

D Share caused by people miscommunicating
26%Locked

The slice of rework that starts with someone not getting the right information, not asking, or not answering: the crew working off an old sheet, an RFI nobody answered, a sub closing up before inspection.

Why it's locked at 26%

Almost nobody tracks rework by cause, so you can't pull this from your books. Instead of making you guess, this uses the only widely cited survey that splits rework by cause.

In FMI and PlanGrid's Construction Disconnected survey (2018), US contractors attributed 26% of their rework to poor communication among project stakeholders. The single biggest communication failure was people not responding to questions and requests (30.9%).

The same survey put another 22% on poor project data, for a combined 48%. We leave that 22% out on purpose. It's largely a document-control and software problem, and counting only the people part keeps the number conservative and hard to argue with. The rest of rework comes from design errors, changes requested by the project owner, workmanship and other causes that leadership training doesn't fix.

Limits

Self-reported by about 500 US respondents, in a survey sponsored by a software company. Treat it as what contractors say causes their rework, not a measured audit.

Source

E Rework from miscommunication, per year
—

Craft turnover

Field workers who quit

F
people

Quits only, not layoffs at the end of a job. For a gut check: construction loses about 1.8% of its workers to quits every month (BLS), roughly one in five a year.

G
$

Default is $39.69 an hour (AGC and NCCER 2025 average) times 2,080 hours.

H Cost to replace one craft worker
25% of wageLocked

Recruiting, onboarding and safety orientation, overtime to cover the gap, and a slower crew while the new hand learns.

Why it's locked at 25%

The Center for American Progress reviewed 30 case studies on what it costs to replace a worker. The median was 21% of annual pay, and about 20% for jobs paying up to $75,000 (2012). Most of those studies counted recruiting and training but not lost productivity, so 20% is a floor.

We set craft at 25%, a small step above that floor, because construction adds costs most of those studies didn't: jobsite safety orientation, a crew slowed down while the new hand learns, and injury risk. Travelers' claims data shows 44% of construction injuries involve first-year workers (2026), so every quit puts a higher-risk worker in that spot.

Locking it keeps the math conservative. Nobody in the room can say the number was pushed up to make a sale.

Limits

No published study measures craft replacement cost in construction specifically. The CAP figures are cross-industry and dated. The common claims that turnover costs "6 to 9 months of salary" or "200% of salary" have no traceable source for typical roles, so this calculator doesn't use those claims.

Sources

I Craft turnover cost, per year
—

Leader turnover

Foremen, supers and PMs who left

J
people
K
$
L Cost to replace one leader
75% of salaryLocked

Search time, the jobs that keep running while the seat is empty, the punch list and schedule the replacement inherits, and the customer relationship that walked out the door.

Why it's locked at 75%

Gallup estimates that replacing an employee costs one-half to two times their annual salary, and calls that a conservative estimate (2019). We use 75%, in the lower part of that range.

Leaders cost more to replace than craft because they're harder to find and their work doesn't stop when they leave:

  • Superintendents were the hardest salaried role to fill in AGC and NCCER's workforce surveys: 81% of firms with openings struggled to hire one in 2025, and 75% in 2026. PMs and supervisors came close behind at 76% and 74%.
  • FMI's 2023 Talent Study found only 19% of firms have succession plans for key strategic roles, and FMI notes it can take five to seven years to develop a successor for upper-level roles.
  • The same study found firms expect to lose 28% of their field managers to attrition and retirement within five years.

When a super walks mid-project, someone else runs his jobs while you search for a replacement, and the replacement inherits a schedule and a customer relationship they didn't build.

Limits

No study measures the cost of replacing a construction superintendent or PM. Gallup doesn't publish its method in detail. 75% is an assumption inside Gallup's range, the same figure used in the worked models in the Mavix research report.

Sources

M Leader turnover cost, per year
—

Leader time lost to people problems

Paid hours spent refereeing

N
people

Foremen, superintendents and PMs.

O

Settling fights between trades, chasing down who said what, handling a crew blowup, calming a customer who got surprised, re-explaining instructions that didn't land.

Where these settings come from

In FMI and PlanGrid's Construction Disconnected survey (2018), construction professionals reported spending 4.7 hours a week on conflict resolution, on top of 5.5 hours looking for project information and 3.9 hours dealing with mistakes and rework.

The 2 and 8 hour settings are reasonable low and high points around that average. They are judgment calls, not published figures.

Limits

Self-reported, and some conflict is a normal part of running work, so not every hour is waste. It also overlaps a little with rework. That's one reason the other locked numbers are set conservatively.

Source

P
$/hr

Wage plus taxes, benefits and truck. A $110,000 super is roughly $55 to $70 an hour loaded.

W Working weeks a year
48 weeksLocked
Why it's locked at 48

Fifty-two weeks, less about four for holidays, vacation and weather days. It's a simple working assumption, not a research figure, and it keeps the time math from counting weeks nobody is on site.

Q Leader time lost, per year
—

First-year injuries

The new hands who replace the quits

R1
claims

Travelers reports 44% of construction injuries involve first-year workers, and the average construction injury costs 114 lost workdays (2026).

R2
$

Pull it from your loss runs or ask your insurance broker. Leave it at zero if you don't know; the rest still holds.

R First-year injury cost, per year
—

What it's worth to fix

Test an improvement and compare

U
%

An illustration, not a promise. Start with 20% and decide what's realistic for your company.

X
$

Contract value. This is usually the comparison that makes the number real.

Built from the Mavix Coaching research report on the cost of poor communication, leadership and turnover in construction. Locked figures are set conservatively and each one shows its source. Industry research isn't broken out for private contractors in the $10M to $100M range, so your own numbers always beat a benchmark. Nothing typed here is saved.