Safety Data Sheet Requirements on California Job Sites
Between 2015 and 2026 M Squared Safety evaluated 403,323 individual inspection items across 7,473 job site inspections in California. On the hazard communication line, one item fails more often than any other high volume item we record.
The number that matters
Across 5,034 checks of safety data sheet accessibility, 1,023 failed. That is 20.3 percent. One job site in five could not produce the sheet for a chemical in use, in the area where it was being used.
Hazard communication overall runs a 10.9 percent failure rate in our record, 1,635 findings across 15,057 checks, against an all-category average of 5.1 percent. It is one of the worst performing sections we measure, behind fall protection at 18.5 percent.
The two words that decide this finding: work area
8 CCR 5194(g)(8) requires the employer to maintain copies of the safety data sheets for each hazardous chemical in the workplace, and to ensure they are readily accessible during each work shift to employees when they are in their work area(s).
Almost every company we inspect has the sheets. Very few have them in the work area. A binder in the site trailer is a compliant document in a non-compliant location, and that is the finding.
The electronic SDS trap
The same subsection permits electronic access, in these words: electronic access and other alternatives to maintaining paper copies of the safety data sheets are permitted as long as no barriers to immediate employee access in each workplace are created by such options.
That last clause is where good intentions fail. A cloud binder behind a login the crew does not have is a barrier. An app nobody installed is a barrier. A tablet locked in the foreman's truck is a barrier. Going digital does not satisfy the standard on its own; removing every barrier does.
The test that matters: hand your phone to a worker in the work area and ask them to show you the sheet for the product in their hand. If they cannot do it in under a minute without help, an inspector will record the same failure we record.
What failed, how often, and how to fix it
Every figure below is from our own inspection record. The item wording is the condition assessed; the rate is the share of checks of that item marked as findings.
SDSs readily accessible
1,023 findings in 5,034 checks. 20.3 percent failure rate.
What the inspector is looking at: Whether a worker in the work area can produce the sheet for a chemical in use, without leaving the area and without asking a supervisor. Inspectors commonly pick a product they can see and ask the nearest worker to show them the sheet.
How to fix it: Put the sheets where the chemicals are, not where the paperwork is. If a crew uses solvents on the fourth floor, the sheets belong on the fourth floor. If you use an electronic system, test it from a worker's phone, on site, with no help.
What to document: Record where the sheets live for each active work area and update it as the work moves floors or phases.
SDS accessible to employees as needed
43 findings in 261 checks. 16.5 percent failure rate.
What the inspector is looking at: The same question asked a different way on some inspection templates. The two items together give 1,066 failures across 5,295 checks, which is the clearest single signal in twelve years of our data.
How to fix it: Treat this and the item above as one problem. Fixing access fixes both.
What to document: One access record covers both items.
Chemical inventory list maintained
22 findings in 184 checks. 12.0 percent failure rate.
What the inspector is looking at: Whether the list of hazardous chemicals known to be present actually matches what is on site today. Inspectors compare the list against what they can see.
How to fix it: Rebuild the list at each phase change. New trade on site usually means new chemicals, and the list is what goes stale first.
What to document: Date the inventory list every time it is revised. An undated list is treated as an old list.
Employees received HazCom training
84 findings in 2,122 checks. 4.0 percent failure rate.
What the inspector is looking at: Training records, and whether they cover the chemicals actually in use rather than a generic session from hire date. A new product introduced mid project triggers a new training obligation.
How to fix it: Run a short session whenever a new chemical hazard arrives on site, and cover the specific product, not the category.
What to document: Sign-in sheets naming the specific chemical covered, dated, with the trainer named.
List of hazardous chemicals maintained
89 findings in 2,415 checks. 3.7 percent failure rate.
What the inspector is looking at: The written program's chemical list, tied to product identifiers that match the sheets. A list using trade nicknames instead of product identifiers does not satisfy the standard.
How to fix it: Use the exact product identifier from the safety data sheet on the list, so the two documents can be matched line by line.
What to document: Keep the list inside the written hazard communication program, not as a loose spreadsheet.
Containers of hazardous substances labeled
82 findings in 3,113 checks. 2.6 percent failure rate.
What the inspector is looking at: Secondary containers above all. The manufacturer's drum is nearly always labeled. The decanted jug, the unmarked spray bottle and the transferred fuel can are where this fails.
How to fix it: Keep a marker and blank labels in the gang box. Any container a chemical is transferred into gets labeled before it leaves the storage point.
What to document: Include secondary container labeling in the daily walk, and note it as checked.
The pattern
Hazard communication failures cluster almost entirely on one item. Of 1,635 findings, 1,066 are safety data sheet access. Labeling, inventory and training together account for fewer than 200.
That is unusually good news. Most compliance problems are diffuse and need a program. This one is concentrated, and it is solved by moving paper closer to people.
Frequently asked questions
Where do safety data sheets have to be kept on a job site?
8 CCR 5194(g)(8) requires the employer to maintain copies of the safety data sheets for each hazardous chemical in the workplace and to ensure they are readily accessible during each work shift to employees when they are in their work area. The phrase that decides most findings is work area. A binder in the site trailer is not accessible to a crew working on the fourth floor.
Can safety data sheets be kept electronically instead of in a binder?
Yes. 8 CCR 5194(g)(8) permits electronic access and other alternatives to paper copies, as long as no barriers to immediate employee access in each workplace are created by such options. A login the crew does not have, an app nobody installed, or a tablet locked in a truck are all barriers, and the electronic system then fails the standard rather than satisfying it.
What is the most failed hazard communication item on California job sites?
Safety data sheet accessibility. Across 5,034 checks of that item we recorded 1,023 failures, a 20.3 percent failure rate. It is the most failed high volume item in our entire inspection record, ahead of annual fire extinguisher service.
Does a job site need a written hazard communication program?
Yes. 8 CCR 5194(e)(1) requires every employer to develop, implement and maintain at the workplace a written hazard communication program. The program must include, under 5194(e)(1)(A), a list of the hazardous chemicals known to be present using a product identifier referenced on the appropriate safety data sheet.
Do all chemical containers on site have to be labeled?
Yes. Under 8 CCR 5194(f)(6) the employer must ensure that each container of hazardous chemicals in the workplace is labeled, tagged or marked. This includes secondary containers such as a decanted solvent in an unmarked jug, which is where this item usually fails.
When does hazard communication training have to happen?
8 CCR 5194(h)(1) requires effective information and training on hazardous chemicals in the employee's work area at the time of their initial assignment, and again whenever a new chemical hazard is introduced into their work area. Training at hire alone does not cover a new product brought on site mid project.
Related
Every guide in this series reports what our own inspections found on that section, with the fix and the record to keep: fall protection requirements, where marking a floor hole cover fails at 41.1 percent; the written emergency medical services plan, which fails at 23.8 percent; the Code of Safe Practices and IIPP, failing at 17.7 and 13.9 percent; fire extinguisher requirements, where annual service fails at 18.2 percent and California caps extinguisher travel distance at 75 feet where the federal standard allows 100; ladder requirements, where securing the ladder fails at 31.7 percent; heat illness prevention requirements, where response procedures fail at 11.0 percent while shade fails at 0.7 percent; PPE requirements, where eye and face protection fails at 13.7 percent against 5.6 percent for head protection.
About this data
M Squared Safety Solutions, Inc. is a construction safety and Cal/OSHA compliance consulting firm based in San Clemente, serving Orange County, Los Angeles, San Diego, Riverside, San Bernardino and Ventura counties. The figures on this page come from our own third party inspection record, covering 403,323 evaluated items across 7,473 inspections between 2015 and 2026. They describe what we found on California job sites. They are not a Cal/OSHA publication and not a prediction of any individual inspection outcome.
We write hazard communication programs, build the chemical inventory, and set up sheet access that survives an inspection. Call 949-954-6581.