If you manage contractors, or write documents for clients who manage contractors, you have read hundreds of method statements. You have probably also noticed that most of them tell you almost nothing about how the work will be done. They describe a scope, restate a set of generic safety platitudes, list some legislation, and arrive at a signature page. The useful ones are recognisably different, and the difference is not length or polish. It is whether the document contains an actual sequence of operations that somebody could follow. This is a practitioner's explanation of what a method statement is, what makes one good, and why the bad ones are bad in such a consistent way.
The message up front: a method statement is a work-planning and communication document that carries safety content. It is not primarily a safety document. Once you see it that way, both its value and its usual failure mode become obvious: it is valuable because planning work in sequence is genuinely how you find the clashes, the access problems and the isolations nobody had thought about, and it fails because it is usually produced to clear a client's document gate rather than to plan the job.
This is a general explanation of an industry practice, written for readers in mixed jurisdictions. What any specific site or contract actually requires is set by that contract, by the site's own safety management system, and by the law applying where the work is done. Preparing and approving method statements for real work is a competence matter, and nothing here substitutes for the judgement of someone who knows the work and the local legal framework.
1. What a method statement is, and the distinction that unlocks the topic
A method statement is a written description of how a defined piece of work will be carried out: the sequence of operations, the people and equipment involved, the arrangements for access, isolation and interfaces with other parties, and the control measures that apply at each stage. Its subject is the method. That is the whole idea, and it is right there in the name.
The distinction that makes everything else make sense is this one. A risk assessment identifies what could go wrong and what controls are needed to reduce the risk to an acceptable level. A method statement describes how the work will be done, in order, with those controls built into the steps. One is analytical and looks at hazards. The other is procedural and looks at the job. They answer different questions, they are read by different people at different moments, and a document that tries to be both usually ends up being neither.
That comparison is worth a proper treatment of its own, and it has one: see risk assessment vs method statement for the side-by-side. The two are also very commonly submitted and read as a pair, which is where the term RAMS comes from, and that pairing is covered in the RAMS guide. This article stays with the method statement itself.
On terminology: in some regions, particularly Australia and New Zealand, the common term is safe work method statement, often abbreviated SWMS, and it tends to be used for higher-risk construction work with a somewhat more prescribed structure. In the UK and across much of the Gulf the plain term method statement is normal, often bundled with the risk assessment as RAMS. Elsewhere you will hear work method statement, construction method statement, or simply method of statement of works in contract language. The naming varies; what a good one contains does not vary nearly as much.
The framing that changes how you write one
Stop asking "what does a method statement need to include to be accepted" and start asking "if I handed this to a supervisor who has never done this job, could they run it". The first question produces documents. The second produces methods. Almost every quality problem in this area traces back to which question the author was answering.
2. Who a method statement is actually for
Most bad method statements are not badly written. They are well written for the wrong reader. A method statement has three audiences, they want different things from it, and the author usually serves only one of them.
| Audience | What they need from it | What they get instead |
|---|---|---|
| The people doing the work supervisor, crew |
A usable sequence. What happens first, what happens next, what has to be true before each step starts, who does what, and what to do when something is not as expected. | Pages of legislation and PPE lists, with the actual sequence compressed into one paragraph or absent entirely. |
| The client or principal contractor | Assurance that this contractor understands the work, has thought about the site's particular constraints, and will not create a problem for other trades or for the operating building. | A polished document that demonstrates familiarity with safety vocabulary and nothing about this site. |
| The reviewer or approver | Visible evidence that the significant hazards of this job have been addressed by the method, not merely acknowledged, and that interfaces and permits have been identified. | A completeness checklist satisfied: every heading present, every hazard named, no line connecting hazards to steps. |
The honest core of the problem is that the second audience is the one that controls whether the contractor gets on site, so that is the audience the document is written for. A method statement written only for the client becomes a sales brochure with a signature page: confident, generic, legally literate, and operationally empty. It has been optimised for acceptance rather than for use, and acceptance is granted by someone who will never be on the tools.
The resolution is not to ignore the client. It is to recognise that a document which genuinely serves the crew also happens to be the most persuasive possible evidence to the client that the contractor knows the job. A specific, sequenced, site-aware method statement reassures a competent reviewer far more than a long one does. The two audiences are only in conflict when the reviewer is checking for presence rather than quality, which is a review problem rather than a writing problem, and I come back to it later.
3. What a method statement needs to settle, and why
Rather than a list of headings to fill in, it is more useful to think about what questions the document has to settle before work starts. If a question is genuinely settled, the reader can act. If it is answered in the abstract, it is not settled.
| What it settles | What that is for | How it goes wrong |
|---|---|---|
| The work, the location, and the boundary | Defines what is covered and, importantly, what is not. The exclusions are what stop a crew improvising into work nobody planned. | Scope copied from the contract, no exclusions stated at all, so any adjacent task gets absorbed into the same document by assumption. |
| The sequence of operations | The heart of the document. Establishes order, dependencies and the conditions that must hold before each stage begins. | Replaced by a description of the scope. Says what will be achieved rather than what will happen, in what order. |
| Access and egress | How people, equipment and materials get to the work face and out again, including routes shared with building occupants or other trades. | "Safe access will be maintained at all times." No route named, no lift or stair identified, no thought given to how a long item turns a corner. |
| Plant, equipment and materials | What will be used, so the client can check compatibility, certification and suitability, and so the crew brings the right kit. | A generic inventory that includes equipment the crew has no intention of bringing, and omits the one item the method actually depends on. |
| Isolations, permits and interfaces | Where the work touches the client's systems, the permit regime, or other contractors. The most common source of real incidents on operating sites. | Permits mentioned as a category, with no identification of which systems need isolating, who requests, or who authorises. |
| People, roles, competence and supervision | Who does the work, who supervises, what qualifications the method depends on, and who has authority to stop. | Job titles with no named supervisory arrangement, and competence asserted rather than tied to the steps that require it. |
| Controls mapped to steps | Carries the risk assessment's conclusions into the method at the point they apply, so a control is attached to the moment of exposure. | A generic control list at the front of the document. Every control present, none located anywhere in the sequence. |
| Waste, housekeeping and leaving the area safe | What happens to arisings, and the condition the area is left in at the end of each shift and at completion. | Omitted, or reduced to a single sentence, which is why so much friction with building occupants happens after the work stops. |
| Emergency arrangements | What the crew does if something goes wrong, in terms that fit the site rather than the contractor's head office. | Generic first-aid and evacuation text referring to arrangements that belong to a different building. |
| Hold points and stages | Defined moments where work stops for inspection, test or authorisation before continuing. Often the single most useful structural feature. | Absent. Everything reads as continuous, so there is no formal moment at which anyone checks anything. |
Read that table as reasons rather than as a form. The point is not that a document must have ten sections with those titles. Plenty of good method statements settle several of these questions inside the sequence itself, which is usually better than putting them in separate sections. The point is that if any of these questions is still open when the crew arrives, somebody will answer it on the spot, and that on-the-spot answer is where most of the risk actually lives.
The controls line deserves particular emphasis. Controls that arrive as a front-loaded list are the single clearest sign that a method statement was assembled rather than written. Where the control belongs is inside the step: at the point the crew opens the panel, at the point the load leaves the ground, at the point the guard comes off. The reasoning behind selecting those controls in the first place belongs to the assessment, and the ordering logic behind them is covered in the hierarchy of controls guide.
4. Sequence is the substance
If you take one test away from this article, take this one. Can somebody who has not done this particular job read the document and understand what happens first, what happens next, and what must be true before each step starts? That is the test. It is not a proxy for quality; it is the thing itself, because the sequence is what a method statement is.
Most method statements fail it, and they fail it in a specific way: they describe a scope rather than a sequence. "The contractor will replace the pump set, including disconnection, removal, installation of the new unit, reconnection, commissioning and handover." Every word of that is true and none of it is a method. It does not say what is isolated before disconnection or by whom, how the old unit leaves the plant room, what is verified before the new unit is energised, or what happens if the new unit does not fit the existing base.
A sequence, written properly, has a distinctive shape. Each step has a precondition, an action, a verification, and often a hold point. It reads as a chain of dependencies, not a list of activities. It is also where the genuine planning value appears, because the act of writing the steps in order is what surfaces the problems: the isolation that cannot be taken during occupied hours, the access route that will not take the equipment, the two trades who both need the same space on the same day. Those discoveries are the return on writing a method statement, and they are available only to an author who actually writes the sequence.
There is a related test worth applying to your own drafts. Look at the verbs. A sequence uses concrete ones: isolate, prove dead, lock off, lift, position, torque, test, witness, reinstate, hand back. A scope description uses abstract ones: ensure, undertake, manage, maintain, carry out, comply with. A document dominated by the second set has no method in it, however long it is.
5. Specificity, and the generic-template problem treated fairly
The standard complaint about method statements is that they are generic, and the standard remedy offered is that every document should be written from scratch. That is not right, and it is worth being fair about this, because the moralistic version of the argument is why practitioners stop listening.
Reusing a document for genuinely similar work is sensible and efficient. A contractor who replaces the same class of equipment fifty times a year should have a well-developed, hard-won, repeatedly improved method for that work, and rewriting it each time would make it worse rather than better, because each rewrite loses accumulated learning. Standardised methods are how organisations retain knowledge. The failure is not reuse. The failure is submitting a document whose named location, equipment, access route, isolations and interfaces do not correspond to the job in front of you, while presenting it as if they do.
Since a large part of this audience reviews contractor submissions rather than writing them, here are the practical tells. None of them is conclusive on its own; two or three together nearly always are.
| Tell | Why it gives the document away |
|---|---|
| No identifiable feature of the actual site appears anywhere | No plant room, riser, roof, level, switchboard reference or route is named. A document about your building should be unmistakably about your building. |
| Controls do not correspond to the hazards listed | Controls for hazards the job does not present, and no controls for the ones it obviously does. The list came from somewhere else. |
| Equipment listed that the crew has not brought, or cannot use here | Access equipment that will not fit the space, or lifting kit nobody has ever seen on the job. The inventory is aspirational. |
| The sequence would describe any job of that type | Remove the equipment name and the document still reads correctly. That is the definition of generic. |
| No interface with your permit system or your isolation authority | The document was written as if the work happens on an empty site with no operator. On an occupied building this is the most consequential gap. |
| Emergency and welfare arrangements belong to somewhere else | Assembly points, first-aid provision or contact routes that do not exist here. Nobody checked them against the site. |
| Legislation cited that does not apply in this jurisdiction | A strong signal the document has travelled between regions unedited, which means the method may have travelled too. |
| Revision history with no substantive changes across many revisions | Repeated resubmission to clear a gate, rather than revision because the method changed. |
The constructive way to use that list is not as an accusation. It is as a set of questions at review: which parts of this are your standard method, and which parts are specific to this job? A competent contractor answers that comfortably and often improves the document in the conversation. A contractor who cannot answer it has told you something important.
6. Writing, reviewing and approving
A method statement should be written by somebody who knows how the work is actually done. That usually means the supervisor, the engineer or the tradesperson who will run it, sometimes with help from a safety adviser on the control side and from a document controller on format. It should not be written by the document controller. When the person holding the pen has never done the work, the document can only be assembled from other documents, and assembly is exactly how generic text propagates.
This has a practical consequence that is often resisted: the best method statements are frequently less polished than the worst ones. A slightly awkwardly worded sequence written by a supervisor who knows the plant is worth more than a beautifully formatted document produced by someone who has only read the scope. If a review process penalises the first and rewards the second, it is selecting against quality.
On the review side, the reviewer needs to be competent to judge the method, not merely to check that the boxes are filled. Judging a method means being able to ask whether the sequence is feasible, whether the isolation strategy is right, whether the access arrangement works in this space, whether the hold points are in the right places. That is an engineering and operational judgement. It is a different skill from document administration, and on most sites it sits with different people, which is why so many reviews are administrative by default.
The honest observation about reviews
A review process that checks for the presence of sections rather than the quality of the method trains contractors to produce longer documents, not safer ones. Contractors are rational. If acceptance depends on completeness, they will deliver completeness, and the cheapest route to completeness is more standard text. Every client who complains about generic forty-page method statements should look at whether their own review criteria asked for anything else.
Approval means something specific and is worth stating plainly: the approver is accepting that the method described is a reasonable way to do this work on this site, and that the interfaces with the site's own systems have been correctly identified. An approval signature from someone with no knowledge of the work conveys nothing, and worse, it creates a false record that the method was examined. If nobody available is competent to judge the method, the honest response is to obtain that competence rather than to sign.
7. Briefing it to the crew
This is where the document either becomes real or does not. A method statement that has been approved and filed has changed nothing about how the work will be done. What changes the work is the crew understanding the method, and that happens in a briefing, at the workface, before the job starts, in the language the crew actually uses.
A good briefing does not read the document aloud. It walks the sequence, points at the actual plant, names the hold points, confirms who is doing what, and makes explicit what the crew should do if reality does not match the paper. It invites the crew to say where the method looks wrong, which is the most valuable information available at that moment, because the people who do the work almost always know something the author did not. The mechanics of running that conversation well are covered in the toolbox talk guide.
The point to be firm about: a signature confirming receipt is not evidence of understanding. Sign-off sheets are useful records of who was present. They are not records of comprehension, and treating them as such is one of the more comfortable self-deceptions in contractor management. If you want evidence of understanding, ask two or three people to describe the first three steps and what stops them proceeding. That takes a minute and tells you far more than a page of initials.
8. Changing the method when the work changes
Work rarely goes exactly as written. The valve is a different size, the access is blocked, the isolation cannot be taken today, the structure behind the panel is not what the drawing showed. This is normal, and a method statement that cannot accommodate it becomes an obstacle, at which point the crew works around it informally and the document becomes fiction.
So there needs to be a route to change the method with authority. That means somebody identified who can approve a change, a threshold for what constitutes a change significant enough to need reapproval rather than a note, and a practical way to do it in the timescales the work runs on. A change route that takes three days will not be used on a job that has to continue this afternoon. Designing the route to be usable is what determines whether it gets used, and it is worth more attention than the original approval workflow usually gets.
The cultural tell that this is working is straightforward: deviations get raised rather than concealed. On sites where changing the method is genuinely possible, supervisors stop work and call. On sites where it is not, they carry on and the paperwork is reconciled afterwards, or never. If you have never had a contractor stop and ask to change an approved method, that is not evidence your methods are unusually good. It is more likely evidence that asking is not worth the trouble.
9. Method statements in facilities and maintenance
Nearly all published guidance on method statements is written from a construction perspective: a project, a defined programme of work, a principal contractor, a site. Facilities and maintenance work is different in ways that matter, and the differences are usually handled badly because the construction model is simply imported.
Contractor mobilisation. In FM, the method statement is often produced once at the start of a term contract, covering a category of recurring work, and then treated as permanent. That is not inherently wrong, but the document that was written to win mobilisation approval is usually the most generic one in the file, and it quietly becomes the standing method for years of work.
Recurring planned work. This is the case where a standing, periodically reviewed method statement is genuinely appropriate. Quarterly plant servicing, annual statutory inspection, routine filter and belt work: the method is stable, the crew is familiar with it, and an annually reviewed document that is actually kept accurate is better than a new one written each visit. What makes this work is that the review is real and that site-specific variations are captured against the individual location rather than averaged away. Tying the method to the planned work itself, rather than to a document library, is the practical move, and it fits naturally with how work order types are structured in most maintenance systems.
Reactive attendance. Here the construction model breaks down completely. A leak at two in the morning does not wait for a method statement to be written and approved. Pretending otherwise produces one of two outcomes: the work is done with no method document at all, or a document is back-dated, which is worse because it corrupts the record. The realistic answer is a pre-agreed generic method for the category of reactive work, covering the parts that genuinely are standard, combined with a short site-specific confirmation at the point of attendance: what is actually being isolated here, what access is being used, what else is affected. That is a defensible arrangement, it reflects how the work really happens, and it is far better than a fiction.
The accumulation problem. This is the client-side failure that deserves more attention than it gets. An operating estate with a dozen service contractors, each submitting method statements for each work category across each site, accumulates hundreds of documents within a couple of years. Nobody has read most of them. Nobody knows which are current. Several describe plant that has since been replaced. Several were approved by staff who have left. The collection has volume and no usable content, and it creates a specific hazard: the existence of the file is mistaken for control of the work.
What I would advise on the accumulation problem
Ask for fewer documents and read them properly. A client who accepts one well-reviewed method statement per genuine work category, with named currency and a named reviewer, and who declines the rest, is in a far stronger position than one holding four hundred unread submissions. Volume of documentation is not evidence of contractor control; it is frequently evidence of its absence. If your system cannot tell you which method statement governs a job starting tomorrow, you do not have a method statement regime, you have an archive.
For the wider operating context this sits in, from planned work structure to contractor control, see the facilities maintenance management guide, and for where the method statement hands over to the authorisation layer on live systems, permit to work integration.
10. How method statements fail
The failure patterns are consistent enough to list, and every one of them is recognisable to anyone who has spent time reviewing submissions.
- Written after the work. Produced to close out a file rather than to plan a job. Contributes nothing to the work and corrupts the record of what was planned.
- Copied from another site. The method may even be sound, but the access, isolations, interfaces and emergency arrangements belong to a different building.
- No sequence. A scope description in place of a method. The most common failure and the most fundamental.
- Generic controls. A front-loaded control list with no connection to the steps where exposure actually occurs.
- No interface with the client's permit or isolation system. Written as if the site has no operator. On occupied buildings and live plant this is where real incidents originate.
- The crew never briefed. The document exists, is approved, is filed, and has never been read by anyone who will touch the work.
- Approved by someone with no knowledge of the work. Produces a signature, a record of review that did not happen, and false assurance on both sides.
- No route to change it. Reality diverges, the method is abandoned informally, and nobody says so.
- Length used as a proxy for rigour. Forty pages of standard text buries the three paragraphs that matter, and reviewers stop reading long before they reach them.
Notice that most of these are client-side as much as contractor-side. Documents written after the work, approved by the unqualified, never briefed and never revisable are produced in response to systems that ask for exactly that. Contractors respond to what is rewarded.
11. Standards, law, and what is actually required
A point worth making plainly, because a great deal of published material implies otherwise: the method statement is a widely used industry practice, strongest in construction and contracting, rather than a document universally mandated by name in law. Where legislation names a document, it typically names a risk assessment rather than a method statement.
In Great Britain, the general duty to assess risk sits in the Management of Health and Safety at Work Regulations 1999 (SI 1999/3242), where the risk assessment duty is Regulation 3, under the framework of the Health and Safety at Work etc. Act 1974 (c. 37). Note that Northern Ireland has separate instruments, and their years differ. What those regulations require is assessment; the method statement is the practical vehicle the industry developed to carry the resulting controls into the work, and its authority in most cases comes from the contract and from the client's own management system rather than from a named legal requirement. In the United States, federal OSHA regulations under 29 CFR similarly do not require a written method statement as such, and do not use the term. Written assessment and certification requirements do exist in specific places, for example the hazard assessment certification in 29 CFR 1910.132(d), which requires a written certification identifying the workplace, the certifying person and the date, but that is an assessment requirement rather than a method statement requirement. In the United Arab Emirates, occupational safety duties in the private sector sit under Federal Decree-Law No. 33 of 2021 on the Regulation of Employment Relationships, administered by MOHRE, with OSH duties in Article 13, alongside emirate-level frameworks such as ADOSH-SF, the Abu Dhabi Occupational Safety and Health System Framework, Version 4.0, administered by the Abu Dhabi Public Health Centre. US OSHA and UK HSE law have no legal force in the UAE; they are voluntary benchmarks.
On the standards side, the management-system standards do not prescribe a method statement either, but they do require the thing a method statement delivers. ISO 45001:2018, as amended by Amd 1:2024, requires the hierarchy of controls at clause 8.1.2, and the same requirement appears in ANSI/ASSP Z10.0-2019 at section 8.4 in the US. Both are voluntary; neither is law anywhere by itself, although ISO 45001 is certifiable and is the usual international reference in the Gulf. ISO 31000:2018, "Risk management. Guidelines", is frequently cited in this context and is worth being accurate about: it is guidance and is not certifiable, so there is no accredited organisational certification against it. The companion techniques document is IEC 31010:2019, "Risk management. Risk assessment techniques", correctly cited as IEC rather than ISO. For permit systems specifically, no international standard specifies a permit-to-work system; the widely used reference is UK HSE guidance HSG250, "Guidance on permit-to-work systems", 2005, written for the petroleum, chemical and allied industries, and it is guidance that creates no duties itself.
The practical implication of all this is not that method statements are optional. It is that their content is governed by what the work needs and by what the contract and site system require, rather than by a designation you can look up. That makes the quality of the method the only real measure, which is a reason to take the sequence seriously rather than the format.
Primary sources worth going to directly rather than reading second-hand: HSE for Great Britain, OSHA for the United States, and ISO for standards designations and current editions.
If you are working through the surrounding documents, the related material is split by intent: risk assessment covers the analysis, job safety analysis covers the task-step technique with JSA versus JHA on the terminology, permit to work covers authorisation on live systems, and what HSE means sets out the wider frame.
The idea to walk away with
A method statement is a plan for doing work, written in sequence, with the safety controls placed where they apply. It is not a safety document that happens to mention the work, and it is not evidence of compliance. Its value comes from the act of planning the work in order, because that is what surfaces the access problems, the isolation clashes and the trade interfaces before they become incidents. Its failure comes from being written for a client's document gate instead, at which point it costs money, produces paper, and changes nothing about how the job runs.
Everything else follows from that. Write it for the crew and the client is reassured as a side effect. Put the sequence at the centre and the specificity takes care of itself, because you cannot write a real sequence without naming the real place. Have it reviewed by someone who can judge the method and contractors stop padding. Make it changeable and deviations get raised. Brief it properly and it becomes how the work is actually done, which was the point.
Final thoughts
The most useful thing I can suggest to anyone on the client side is to pick one method statement from the file at random, for work that is happening this month, and apply the single test: could a competent supervisor who has not done this job run it from this document? If the answer is no, the problem is almost certainly not that contractor. It is the review criteria that accepted it, and that is within your control to change.
For anyone on the contractor side, the advice is simpler and slightly counterintuitive. Write shorter documents with real sequences in them. Name the plant room. State what you are not covering. Put the controls in the steps. Put hold points where somebody should actually check something. You will get fewer comments back, not more, because a competent reviewer recognises a real method immediately, and the crew will actually use what you wrote. That is the only test of a method statement that has ever mattered.
Disclosure
Alongside advisory work I also build a CMMS and CAFM platform, so I have a commercial interest in this category. Nothing above is a recommendation for it, and no vendor named here has paid for inclusion or had any editorial input. Weigh the analysis accordingly.
Drowning in contractor method statements?
Independent advisory on contractor documentation regimes, permit and isolation interfaces, and how planned work, method statements and authorisation fit together in an operating estate. 22+ years across utilities, oil and gas, manufacturing, government and facility operations.
Book a conversationRelated reading: Risk assessment vs method statement, RAMS explained, Permit to work, Toolbox talks, Facilities maintenance management.
Muhammad Abbas
CMMS / CAFM Manager & Independent Advisor · 22+ years across enterprise CMMS, EAM, CAFM and ERP implementations in utilities, oil and gas, manufacturing, government and facility operations.
Work with me