Key Takeaways
- The recall problem: an annual review written in January is a judgment about work that happened in June, reconstructed from memory.
- Safety leadership is a defined duty: 29 CFR 1926.32(f) makes the competent person someone authorized to take prompt corrective measures.
- The rule names the behavior: 1926.20(b)(2) requires frequent and regular inspections by competent persons, and 1926.21(b)(2) requires instructing each employee.
- Outcomes mislead: a quarter with no recordable incident can reflect a quiet quarter as easily as a well-run crew.
- Cycle length: a project runs on its own clock, so the review cycle belongs to the project.
Your superintendent sits down in January to write reviews for 6 foremen. The work being judged happened between March and November, across 3 projects, and the only contemporaneous record is a folder of daily reports nobody indexed. What gets written is an impression of the last 8 weeks.
That is what happens when the record of a year's work is spread across daily reports, toolbox talk sheets and one superintendent's memory. Performance management software earns its place by holding that record as it is created, so the January conversation has evidence in front of it.
The 3 Things a Foreman Review Has to Measure
A foreman's job splits into 3 families, and they behave differently under measurement. Schedule adherence is countable, with milestone dates and crew deployment against plan. Quality outcomes are partly countable, through first-inspection pass rates and punch item closure. Safety leadership is the one most review forms handle badly.
What you measure | Where the evidence lives | How often it should be logged |
|---|---|---|
Milestone hit rate | The project schedule | Per milestone |
Crew deployment against plan | Daily reports | Weekly |
Safety observations and corrections | Site inspection records | Per inspection |
First-inspection pass rate | QA sign-offs | Per inspection |
It is handled badly because most forms reach for an outcome. A quarter with no recordable incident can mean a well-run crew or a quiet quarter, and those two look identical on a form that counts incidents. Construction recorded 2.2 total recordable cases per 100 full-time equivalent workers in 2024, and 1.3 of those involved days away, restricted work or transfer. Events that infrequent tell you very little about any single quarter.
The Regulation Already Names the Behavior
What the form should measure is written into the standard that governs the role. 29 CFR 1926.32(f) defines a competent person as somebody capable of identifying existing and predictable hazards in the working conditions, and authorized to take prompt corrective measures to eliminate them. That definition describes the foreman on almost any site, which is why the review can be written against it.
Two more sections turn it into observable work. 29 CFR 1926.20(b)(2) requires an accident prevention program to provide for frequent and regular inspections of the job site, the materials and the equipment by competent persons designated by the employer. Alongside it, 1926.21(b)(2) requires the employer to instruct each employee in the recognition and avoidance of unsafe conditions.
So the measurable behaviors are inspections run, hazards identified, corrections made and crews instructed. Each of those leaves a record on the day it happens. A review that reads those records is judging what the foreman did, and a review that counts incidents is judging what happened to them.
Why the Annual Cycle Fits Projects Badly
Those records also arrive on a project clock, which the calendar year cuts across. A goal set in January, such as crew deployment at 95% of planned headcount, belongs to a project that may finish in August and be followed by one with different conditions. Judging both against the same January sentence flattens the difference.
A new foreman shows the mismatch most clearly. Their first 90 days on a project is when coaching changes the outcome, and an annual form reaches them 10 months later. The same structure that carries an annual review can carry a 90-day one, and the second is where the superintendent's time pays most.
Before you redesign the form, settle 4 things with your superintendents:
- Decide which of the 3 families each goal belongs to.
- Write what good looks like for each, in a sentence a foreman can read.
- Name where the evidence for it is logged, and by whom.
- Set the review cycle to the project, and keep a short one for new foremen.
Take your last completed project and try to answer 3 questions about one foreman from the records alone. How many safety observations did they log? What was their first-inspection pass rate? Which milestones moved and why? Where the records cannot answer, the review was always going to be memory.
Hold the record as the work happens.
KC Performance runs annual, quarterly or fully custom cycles with goals and competencies in one record.
Continuity Is the Other Reason to Write It Down
Crew continuity is the second argument for a written record. Total separations in construction ran at 3.2 per 100 in August 2026 on the BLS measure, with project completion and seasonal cycling behind as much of it as resignations. A foreman who moves between projects takes their reputation with them, and that reputation is the only thing that moves, because nothing in the daily reports travels with them.
That is a problem when the next superintendent inherits the crew. Without a record they start from the same place the last one did, forming an impression over a season. Competency mapping for operational roles covers writing the capability definition that a record like this is scored against.
Written Standards Beat Calibration Meetings
A written standard is also what keeps 2 superintendents judging the same way. Where 2 of them rate the same foreman differently, the usual response is a calibration meeting, which treats a definition problem as a scoring problem. Writing down what a competent inspection log looks like settles it earlier and more cheaply.
Those standards do not need to be long. One paragraph per family, naming the evidence and the level expected, covers a foreman role completely. Most contractors need 3 or 4 of these, because a foreman is a foreman whether the project is a school or a distribution center.
What KC Performance Changes for a Contractor
Cycle length is the first thing KC Performance changes, and it is why performance review software for small companies and large contractors alike has to run annual, quarterly or fully custom cycles. A 90-day review for a new foreman and an annual review for an experienced one use the same record, so nothing moves to a side spreadsheet. The project clock and the review clock can finally match.
Goals, competencies and 360 feedback live in that same record. A foreman measured on milestone adherence and one measured on safety observations are both using 1 system with different content, and the superintendent sees the whole picture in one place. Succession planning and improvement plans run from there too.
Development is where the record earns its keep. A gap identified in a review becomes a training assignment through the KC LMS integration, with a completion record behind it, so an improvement plan carries evidence behind it. Performance management for frontline supervisors follows that link from review to pay decision.
Pick your highest-volume foreman role and write 1 standard for each of the 3 families this month. Test them on a single completed project before you change the form for everybody. The form will be obvious once those 3 standards exist.
Frequently Asked Questions
1. What should a foreman review actually measure?
Behavior that leaves a record, in 3 families: schedule adherence, safety leadership and quality outcomes. Safety leadership is the one most forms get wrong, because they count incidents. The regulation already names the behavior, since 29 CFR 1926.20(b)(2) requires frequent and regular inspections by competent persons and 1926.21(b)(2) requires instructing each employee.
2. Why is counting incidents a poor measure of a foreman?
Because incidents are infrequent and partly a matter of chance. Construction recorded 2.2 total recordable cases per 100 full-time equivalent workers in 2024, 1.3 of them involving days away, restricted work or transfer, so a single quarter with no recordable case can reflect a quiet period as easily as good management. Measure the inspections and corrections instead, which happen weekly.
3. Should reviews run annually or on the project?
On the project, with a short cycle for new foremen. A goal set in January belongs to a project that may end in August, and judging the next one against the same sentence flattens the difference. A new foreman's first 90 days is when coaching changes the outcome, and an annual form arrives 10 months too late.
4. What is a competent person under the construction standards?
29 CFR 1926.32(f) defines a competent person as one who is capable of identifying existing and predictable hazards in the surroundings or working conditions which are unsanitary, hazardous or dangerous to employees, and who has authorization to take prompt corrective measures to eliminate them. On most sites that description fits the foreman.
References
- Legal Information Institute. 29 CFR 1926.20, General Safety and Health Provisions. law.cornell.edu
- Legal Information Institute. 29 CFR 1926.21, Safety Training and Education. law.cornell.edu
- Legal Information Institute. 29 CFR 1926.32, Definitions. law.cornell.edu
- Bureau of Labor Statistics. Table 1, Incidence Rates of Nonfatal Occupational Injuries and Illnesses by Industry, 2024. bls.gov
- Bureau of Labor Statistics. Job Openings and Labor Turnover Survey, Table 3. bls.gov