Permits to work and contractor control.

Contractor control has a single characteristic failure. A contractor is approved, their insurance and accreditations are checked, the paperwork goes in a folder, and eighteen months later they attend a site with a certificate that expired in the spring. Nobody noticed, because nothing was watching the date. The distinction that matters afterwards is not whether the contractor was approved — it is whether they were approved on the day they were on your site.

Approval is not a binary

A single approved flag against a contractor is the shortest route to trouble, because approval is almost never general. A firm competent on ventilation is not thereby competent on lifting equipment. A contractor with public liability cover at one level is not necessarily carrying the level your client's lease requires. An insurer examining an incident will ask what the approval actually covered, and "they were on our approved list" is a weaker answer than it sounds.

• –Record trades and disciplines against the contractor, not a single approval status
• –Record the cover levels you verified, not just that cover existed
• –Record any accreditation relied on — scheme, number and expiry
• –Record geographic or site coverage where it is limited

Expiry is the mechanism, not the paperwork

The reason a contractor register with expiry tracking beats a folder of PDFs is not tidiness. It is that a date field can be watched and a folder cannot. The annual re-approval exercise that most estates run catches the certificates that expired since last year, which means the exposure window is up to twelve months wide. Watching expiry turns that into a warning before the date, which is the only point at which the problem is cheap to fix.

The other reason is turnover. The manager who approved a contractor in 2023 and understood exactly what they had checked is not necessarily the person fielding the question in 2026. A record that holds the basis of an approval — what was checked, by whom, when, and until when — survives that handover. A conviction that "we checked them properly" does not.

What a permit to work is actually for

A permit to work is a formal authorisation for a specific piece of work, in a specific place, within a specific time window, under stated conditions. It exists because certain activities carry a risk that generic site rules do not manage: hot works, confined space entry, work on live electrical systems, work at height in particular configurations, isolation of fire systems, roof access. The permit forces a conversation about the controls before the work starts, and it creates a record that the conversation happened.

It is not the same thing as contractor approval, and the two are frequently conflated. Approval says this firm is generally competent and insured. A permit says this specific job, today, has been risk assessed and authorised under these conditions. A contractor can be perfectly approved and still be doing hot works on your roof with nobody having considered what is underneath.

• –Approval is about the firm and lasts until something expires
• –A permit is about one job and lasts hours or days
• –A permit names the person authorising, the controls required, and the conditions for handback
• –An expired or closed permit is itself a record — it evidences that control ended deliberately

Paper permits versus electronic ones

Paper permit systems are still very common and they are not automatically inferior. A paper permit in a folder on site is physically present at the point of work, which is where it is needed, and it does not stop working when the signal does. What paper does badly is the estate-level question: how many permits are open right now, across all sites, and is anything still open that should have been handed back on Friday.

Electronic permit systems answer that and introduce their own failure modes, principally that a permit issued from an office by someone who has not seen the work is easier to issue carelessly. If you are moving from paper, the thing worth protecting is the conversation, not the form.

Being clear about what FacilityOptix does here

FacilityOptix does not issue permits to work. There is no hot works or confined space authorisation workflow and no electronic permit sign-off. Teams running a formal permit regime operate it alongside FacilityOptix rather than inside it, and it is better to know that before a demo than during one.

What it does hold is most of the documentary record a permit regime depends on: the contractor register with competence, insurance and accreditation evidence behind any authorisation, and the work order the attendance was made against, with its own photos, notes and sign-off. When the question afterwards is who attended, under what authority, against which job, and what they recorded while they were there, that is where the answer comes from.

A workable minimum

If you are building a contractor control regime from a standing start, this is the order that gets you the most protection soonest.

• –One register of every contractor you actually use — including the ones used occasionally, which is where the gaps hide
• –Insurance and accreditation expiry dates recorded as dates, with a warning ahead of them
• –Trades and coverage recorded, so approval is specific rather than general
• –Every attendance tied to a work order, so there is a record of what they were there to do
• –A permit regime for the high-risk activities, operated by someone who has seen the work
• –A periodic review where expiring documents and unused contractors are both dealt with