New safety manager 30/60/90-day plan

Most people reading this were not hired as a safety manager. They were an operations lead, a supervisor, a quality manager or the person who did not step back quickly enough — and safety arrived on top of the job they already had. The advice written for a new EHS director at a company with a department does not survive contact with that situation.

This plan assumes you have no budget, no data you trust, and no credibility yet, and that the fastest way to lose the last one is to start changing things before you understand them.

Days 1–30 — find out what is actually true

Change nothing structural this month. You are gathering facts, and the facts are mostly not in the documents.

• Walk every area, on every shift you can reach. Night shift runs a different plant than day shift does. If you only ever see one of them you are managing half a site.
• Find the records and find out whether they are current. Injury records for the last three years, training records, inspection records, safety data sheets, any written programs. "We have that somewhere" means you do not have it.
• Establish whether you have to keep OSHA injury records at all. Partial exemptions exist by employer size (1904.1) and by industry classification (1904.2). Do not assume either way — find out, and write down the answer with the reason.
• Meet every supervisor individually. Not a group briefing. Supervisors decide what actually happens, and they will tell you things in private they will not say in a room.
• Ask the question that gets a real answer. Not "is it safe here" — nobody answers that honestly. Ask: "What is the job here you would least like your own kid to do?" You will get the truth, and usually within the first three conversations you will hear the same task twice.
• Find out what happened to the last person who reported something. This single fact tells you more about the safety culture than any survey.

Days 31–60 — fix the obvious, then start exactly one system

Credibility is earned in weeks, on small visible things, and spent on the large invisible ones later. This month you do both — but only one of the large ones.

• Close the fast, visible items. The blocked exit, the extinguisher that has not been checked since last year, the missing machine guard, the eyewash nobody can reach. These are cheap, they are noticed, and leaving them open while you write a strategy document tells everyone what you actually are.
• Pick ONE system to make real. Not three. The usual right answer is incident and near-miss reporting, because everything downstream — priorities, corrective actions, any argument you will ever make for budget — depends on knowing what is actually happening. If your site has an obvious dominant hazard, that system can come first instead.
• Make reporting genuinely easy, and then respond visibly. The first few reports are a test of you, not of the form. What happens to report number one determines whether there is a report number twenty.
• Get recordkeeping current if it applies. The 300 log up to date; the 300A summary posted 1 February to 30 April (1904.32(b)(6)). This is not glamorous and it is the thing most likely to be visibly wrong.
• Do not launch a training programme yet. You do not yet know what the gaps are.

Days 61–90 — make it repeatable without you

A programme that depends on you chasing it is not a programme, it is you. This month converts effort into routine.

• Put the inspection on a schedule. A monthly walkaround that happens because it is scheduled beats a thorough one that happens when you remember.
• Give every corrective action an owner and a date, and review the open ones on a fixed cadence. Weekly or monthly matters far less than never moving.
• Write down the one thing that actually needs writing. Whichever program your real hazards require — energy control, respiratory protection, fall protection. Write the one you need, properly, rather than a general manual you will never use.
• Brief leadership with numbers, not adjectives. How many hazards reported, how many closed, how long closure takes, what is overdue. "Safety is improving" is not a briefing.
• Plan the next 90 days before this one ends, so month four is not another standing start.

Print the one-page plan Clear Type into the form below, then print — what you enter prints with it. Or print it blank and fill it in by hand. Nothing you type is saved, sent to EHSTEC, or stored on this device; refresh the page and it is gone.

What not to do in the first 90 days

• Do not rewrite the safety manual. It is the most visible way to spend three months producing nothing anyone will read. Write the one program your hazards actually require, later, once you know which.
• Do not run a full compliance audit in week two. You will generate two hundred findings, close none of them, and teach everyone that your findings do not matter.
• Do not buy software before you know your process. Tooling makes an existing process faster; it does not supply one. That applies to EHSTEC as much as to anyone else.
• Do not announce an injury-rate target. A target attached to a lagging number you do not yet trust mostly buys you under-reporting.
• Do not promise a culture change. See below.

What you cannot fix in 90 days

• Culture. You can start changing what happens after someone reports something, which is the part of culture you actually control. The rest follows it, slowly.
• A workforce that has learned reporting gets punished. That took years to teach and it is not unlearned in a quarter. Every visible non-punitive response chips at it.
• Production pressure and staffing. If the schedule is the hazard, that is a business decision and your job is to make the trade-off explicit rather than to absorb it quietly.
• Anything requiring capital. Budget cycles are longer than 90 days. Spend this quarter building the evidence that the case will need.

Where EHSTEC fits — and where it does not

Not in month one. Month one is walking, reading and listening, and no software helps with that. By month two, when you have picked the one system to make real, the question becomes whether reports reach you and whether anything happens to them — and that is what EHSTEC does: hazards and near misses captured from a phone, each one becoming a corrective action with an owner and a due date, inspections on a schedule rather than on memory, and the numbers for your month-three leadership briefing produced from the work rather than assembled the night before. See what EHSTEC costs, or start with the free plan.