What Is a Good EMR for Construction? Benchmarks and Cutoffs

What Is a Good EMR for Construction? Benchmarks and Cutoffs

A good EMR for construction is any number below 1.00. EMR is short for experience modification rate, the score insurers use to price your workers' comp. NCCI (the National Council on Compensation Insurance) runs the rating plan in many states. It says the plan works by comparing the experience of individual employers with the average employer in the same classification. So 1.00 is average, a lower number earns a discount, and a higher number adds a charge. BiltOn is a Safety Intelligence platform that gives general contractors (GCs) and their subcontractors verified site records that help protect that number, and its Predictive Safety Management tools help you act on risk before a claim is filed. This guide shows the benchmarks, the cutoffs owners use, and what each point costs.

What is a good EMR for construction?

A good EMR is below 1.00. NCCI defines a credit mod as "a mod lower than 1.00" and a debit mod as "a mod higher than 1.00". A credit means your insurer charges you less than the base rate. A debit means it charges you more. The score is built from your own claims, so it compares you with firms that do the same kind of work. A drywall crew and a high-rise concrete crew are not graded against each other.

The target for a good EMR for general contractors is the same. Owners often want proof that you beat the average, not just match it. Use the table below as an EMR benchmark guide. It is a guide, not a rule.

Sources: NCCI, ABCs of Experience Rating (2025); Safety+Health, Screening for Safety (2013). The band edges between 1.00 and 1.25 are our simplification of those sources.

In short, below 1.00 is good, and lower is better. A mod of 1.00 keeps you in the game. Once you pass 1.25, both your premium and your access to bids are at risk. For the how-to, read our guide on how to reduce your EMR with verified safety data.

What is the EMR formula, and why do the rules change by state?

The EMR formula compares the losses you had with the losses expected for a company of your size and type of work. NCCI says the experience period is generally based on three years of payroll and loss data. The policy you are in now is not used. In NCCI's own example, the newest policy year is also left out because its loss data is not yet valued or reported. A ballast value in the formula helps prevent the mod from shifting too far above or below unity. Unity is NCCI's word for 1.00.

Not every state uses NCCI. New York uses NYCIRB, the New York Compensation Insurance Rating Board. Its plan states that it applies exclusively to the New York Workers' Compensation and Employers' Liability Manual. The IRMI list of state bureaus also names separate bureaus in California, New Jersey, Pennsylvania, and other states. So always check which bureau sets your number. Here is what drives it:

• Claim count matters most. NCCI splits each loss in two. The part up to a dollar line, called the split point, is primary loss, which "reflects frequency." The part above the line is excess loss, which reflects severity.

• The split point varies. NCCI says it is approved as part of each state's rate or loss cost filing. NYCIRB says its value varies by size of risk. In NCCI's example, the split point is $18,500, so a $50,000 loss has $18,500 of primary loss and $31,500 of excess loss.

• Many small claims can hurt more than one big one. NCCI's guide says ten $5,000 losses produce a higher mod than one $50,000 loss, even though the totals match.

• One huge claim is softened. NCCI notes that "one very large loss does not imply a pattern of claim frequency." A state limit also caps how much one claim can add.

• Small medical claims count less. Only 30 percent of each medical-only loss is included in the mod calculation.

Because the plan looks back about three years, one claim can stay in your number for about three renewals. That makes the first days after an injury important. Whatever the insurer pays becomes a loss in the formula, so the facts on file matter.

How does EMR affect your workers' comp premium?

Your mod multiplies your base premium. NCCI says the mod is applied to the employer's manual premium at the policy's renewal. In NCCI's own example, $100,000 x 1.25 = $125,000. The same math works at any size. A mod of 1.25 means you pay 25 percent more than the base. A mod of 0.85 means you pay 15 percent less.

Illustrative example only. We computed these numbers using the multiplication method in NCCI's ABCs of Experience Rating. They are not a quote.

The gap between 0.85 and 1.25 is $200,000 a year. On this premium, each 0.01 on the mod is worth $5,000. Injuries are common enough that this matters. The Bureau of Labor Statistics (BLS) reports 167,100 injury and illness cases in construction in 2024, a rate of 2.2 per 100 full-time workers. Across all private industry, the rate was 2.3 cases per 100 full-time equivalent workers. At 2.2 per 100, a firm with 100 full-time workers would expect about two recorded cases a year. That is our arithmetic. Not every case becomes a paid comp claim, but each one is a chance.

Premium is the first cost of a high mod. The second is lost work, which the next section covers. To go deeper on cutting the premium itself, see lower workers' comp premiums: the safety data ROI for GCs.

What EMR cutoff do owners and GCs use to prequalify subcontractors?

No single national cutoff exists, but 1.0 is the most common line. In 2013, Safety+Health magazine reported that "employers typically require an EMR below 1.0." The same article quotes the Oregon OSHA administrator. He did not set an automatic no-go point, and he said owners should weigh the type of work. Real prequalification rules vary, so the table below shows examples from public documents.

Two lessons come out of this table. First, 1.0 shows up often, but it is a custom and not a law. Second, in some places the law limits cutoffs. A sub should aim for a number below 1.0 and keep a current carrier letter ready. A GC should ask each owner for its rule in writing. Our guide on how to reduce your EMR with verified safety data covers how to move the number once you know your target.

How does BiltOn help protect your EMR?

BiltOn helps by giving you a verified record of who was on site, when, and for which employer. Every paid claim becomes a loss in the mod formula. So the record behind a claim is worth real money. BiltOn verifies every worker at the gate by 3D facial recognition. Each worker is tied to an employer, a trade, an orientation status, and a set of credentials. Every check-in and check-out is timestamped by site, including shifts that cross midnight and sites in different time zones.

The attendance and hours record by worker and employer can be exported. It syncs two ways with Procore and Autodesk, both partners. It is kept in a tamper-evident form that was piloted in the work behind NYC DOB Buildings Bulletin 2024-007. That record supports claims defensibility, which means you can show what happened on site with proof that holds up. Read more in claims defensibility: building a site record that holds up.

BiltOn also captures leading indicators, the early signs of risk that show up before a claim. These include pre-task plans tied to a verified identity and inspections captured at the point of work with a photo and timestamp. They also include incident and near-miss management. BiltOn scores risk from those verified leading indicators. That is Predictive Safety Management in practice. Clients report about a 30 percent average reduction in EMR, premium reductions of about 7 percent, and roughly three fraudulent claims prevented per project. These are client-reported results, not a promise.

Carriers are moving in the same direction. In November 2025, Zurich North America announced a pilot of camera-enabled remote safety observation. It compared nine New York City building projects with the technology against twelve without it. Zurich reported more than $2 billion in cumulative project value and a drop of more than 50 percent in claim frequency. Zurich now requires the technology on every construction wrap-up project it insures in New York. That is a different tool from BiltOn, but the signal is the same. Insurers want verified field data. Learn more in why carriers are repricing GCs.

Customer proof point: SD Builders. SD Builders, a New York City GC, replaced badge swipes with facial verification turnstiles. The company reports zero false workers' comp claims since the change. It also dismissed one claim outright for a date on which the worker never badged in. That is claims defensibility in action: a timestamped record answered a question that a paper sign-in sheet could not.

As Omer Slavin, Co-Founder and CEO of BiltOn, puts it: "See every site, verify every worker, prove every record."

Request a demo and we will show you how BiltOn puts this to work on your sites. Book a demo at bilton.tech/demo.

Executive takeaway

A good EMR for construction is below 1.00, because 1.00 is the average for your class of work. The score comes from about three years of your payroll and losses, and the newest year is left out. NCCI sets the plan in many states, while New York uses NYCIRB, so check which bureau sets your number. The mod multiplies your base premium: on a $500,000 base, a 0.85 costs $425,000, a 1.00 costs $500,000, and a 1.25 costs $625,000. Many owners ask for a mod below 1.0 to prequalify, though some public rules limit cutoffs. Verified records of who was on site help you defend claims and protect the number, and BiltOn is built to produce that record.

Frequently Asked Questions

1. What is a good EMR for construction?

A good EMR is below 1.00, which means your losses are better than average for your class of work. A score of 1.00 is average, and anything above it is a debit. Many owners ask for a mod below 1.0 when they prequalify subcontractors.

2. What is the EMR formula?

The formula compares your actual losses with the losses expected for a company like yours, using about three years of data. It counts claim frequency more than one large loss, and a ballast value keeps the score from swinging too far from 1.00. Your rating bureau sets the exact numbers.

3. How much does EMR change my workers' comp premium?

Your insurer multiplies your base premium by your mod. On a $500,000 base premium, a mod of 1.25 means $625,000 and a mod of 0.85 means $425,000. That is a $200,000 gap in this example.

4. What EMR do owners require to prequalify?

There is no national rule. Safety+Health reported that employers typically require an EMR below 1.0, and some owners, such as Michigan Technological University, ask exactly that. Others, such as SRP, set no numeric cutoff, and Virginia law bars a required mod in public construction bids.

5. Do all states calculate EMR the same way?

No. NCCI runs the plan in many states, but New York uses NYCIRB, and states such as California and Pennsylvania have their own bureaus. Details such as the split point differ, so check the bureau for your state.

Works Cited

• BiltOn. (2026). How to Reduce Your EMR With Verified Safety Data. BiltOn Blog, bilton.tech.

• BiltOn. (2026). Lower Workers' Comp Premiums: The Safety Data ROI for GCs. BiltOn Blog, bilton.tech.

• BiltOn. (2026). Claims Defensibility: Building a Site Record That Holds Up. BiltOn Blog, bilton.tech.

• BiltOn. (2026). Why Carriers Are Repricing GCs. BiltOn Blog, bilton.tech.

• Commonwealth of Virginia. (n.d.). Va. Code Ann. 2.2-4302.1, Requirements for bidders on construction contracts. Virginia Law, law.lis.virginia.gov.

• International Risk Management Institute. (n.d.). Rating Statistical Bureaus. IRMI, irmi.com.

• Michigan Technological University, Facilities. (n.d.). Contractor Prequalification Form. Michigan Technological University, mtu.edu.

• National Council on Compensation Insurance. (2025). ABCs of Experience Rating. NCCI, ncci.com.

• National Council on Compensation Insurance. (2020). Insights From NCCI's Experience Rating Plan Review. NCCI, ncci.com.

• New Mexico Department of Transportation. (2021). 18.27.5 NMAC, Contractor Prequalification Rule. New Mexico State Records Center and Archives, srca.nm.gov.

• New York Compensation Insurance Rating Board. (2024). Experience Rating Plan (effective October 1, 2024). NYCIRB Digital Library, digitallibrary.nycirb.org.

• New York Compensation Insurance Rating Board. (n.d.). Experience Rating Plan, Jurisdiction Where This Plan Applies. NYCIRB Digital Library, digitallibrary.nycirb.org.

• Safety+Health Staff. (2013). Screening for Safety. Safety+Health magazine, National Safety Council, safetyandhealthmagazine.com.

• SRP. (n.d.). Contractor Safety Management Program (CSMP) Standard, Version 1.0. SRP, srpnet.com.

• U.S. Bureau of Labor Statistics. (2026). Number and rate of nonfatal work injuries and illnesses in private industries. BLS, bls.gov.

• U.S. Bureau of Labor Statistics. (2026). Employer-Reported Workplace Injuries and Illnesses, 2024. BLS, bls.gov.

• Zurich North America. (2025). Zurich Selects Arrowsight to Help Transform On-Site Construction Safety Across the U.S.. PR Newswire.