When does a scaffold need a bespoke design?

The line between a compliance sheet and a bespoke design is not a judgement call made on site. It is set by whether the structure you intend to build — its bay lengths, lift heights and tie pattern sits inside published guidance. The problem is that the structures which cross the line rarely announce themselves — they arrive as a small request from a site manager on a Friday afternoon.

The usual crossings

• –Loading bays, and anything else carrying a point load the guidance has not assessed.
• –Sheeting or netting added to a structure that was designed bare — cladding changes the wind loading entirely.
• –Bridges, beamed sections and anything spanning an obstruction at ground level.
• –Temporary roofs, and structures carrying a covering intended to keep weather out.
• –Scaffolds tied into a facade that cannot take the tie pattern the configuration assumes.
• –Support and shoring work, where the scaffold is holding something up rather than holding people.

Who can produce the design

A bespoke design is the work of a competent person — in practice, a scaffold designer with the relevant engineering competence, producing calculations and drawings for that structure. Software does not replace that person. What software can usefully do is make the boundary visible early: flag that a requested configuration falls outside a compliance sheet while the job is still being planned, rather than when the squad is standing on site with the wrong kit on the wagon.

Recording the design once you have it

A bespoke design is only useful if the people inspecting and adapting the scaffold can find it. Held against the structure — the way the platform records it — it tells an inspector what they are looking at and tells the next squad what they are allowed to change. Held in an email thread, it tells nobody anything.

The decision made on site, under pressure

The question rarely arrives as a clean design decision at tender. It arrives mid-job, when a gang hits something the drawing did not anticipate: an obstruction, a façade that will not take ties where expected, a client wanting to load out a lift for materials. Someone has to judge whether this is still within the standard configuration or has become something needing bespoke design, and that judgement often gets made by whoever is standing there.

What makes that safe is having decided in advance who is allowed to make the call, and having a route to escalate that does not cost the gang a day. Where escalation is slow or unclear, the pressure runs one way — and it is worth checking whether the structure is still within the scope your design and compliance basis actually covers.

Common mistakes

• –No clarity on who may decide the structure has moved beyond a standard configuration
• –Escalation so slow that site is incentivised to avoid it
• –Loading changes treated as an operational matter rather than a design one
• –Client requests accepted on site with no design review
• –No record of who made the call, or on what basis
• –Assuming a bespoke design covers later alterations to the same structure

Worth knowing: the cost of referring is lower than it looks

The unspoken reason borderline structures do not get referred is time. A gang stood waiting is a visible, immediate cost, and the referral feels like it might produce a day's delay for something that will probably come back approved.

In practice the expensive outcome is the other one: a structure built outside its basis, discovered at an inspection or a client audit, requiring alteration or strike-and-rebuild with a following trade already on site. Framing the referral as cheap insurance rather than as an obstacle changes the calculation, and it is usually only believable if the referral genuinely is quick.

• –Make the referral route fast enough that using it is not a penalty
• –Name who can approve, and make sure they are reachable during working hours
• –Record the referral and the answer, so the decision is evidenced
• –Track how often it is used — near zero usually means it is being avoided

Where this connects: making referral cheaper than the workaround

Every firm has a referral route and most of them are used less than they should be, for one reason: referring costs a gang time and working round costs them ten minutes. Where that is the arithmetic, site will work round, and no amount of instruction changes it because the instruction is competing with a real cost that lands on the people being instructed.

Firms that get genuine referral rates do two things. They make the route fast — a named person, reachable, able to answer the same morning rather than the same week. And they treat a referral as a good outcome rather than an interruption, which sounds soft and is the difference between a chargehand who rings and one who decides. A referral rate near zero is not a sign that nothing borderline is happening.

• –Name who can approve, and make sure they are reachable in working hours
• –Measure how often referral is used — near zero is a warning, not a success
• –Record the referral and the answer, so the decision is evidenced
• –Treat a referral as the system working rather than as a delay
• –Make the route usable from site, not only from an office
• –Review borderline cases periodically to see what site is deciding alone