Ten storeys on Fenchurch Avenue, a glazed elevation with anodised fins every 1.5 metres, and three contractors quoting three different methods at three prices with a spread of about 400 per cent between them. That happens because façade access is not really a cleaning decision. It is a decision about the building, made mostly by people who left the project fifteen years ago, and the cleaning contractor arrives to find out what they chose.
What Does the Roof Decide Before Anybody Prices the Job?
The roof decides it. Everything downstream – method, frequency, price, whether the job is viable at all – follows from what is up there and whether the paperwork on it is current.
What you’re looking for is a short list. A building maintenance unit, meaning a permanent roof-mounted machine with a trolley on a track and a cradle hung off it. Or a monorail track under a soffit. Or davit sockets and counterweight positions for a temporary suspended platform. Or, most commonly on a 1980s and 1990s City office, a run of abseil eyebolts along the parapet with a date stamp on somebody’s certificate.
Frequently what you find instead is a plant deck laid out in 1988 with no thought given to access at all, a parapet you can’t get a rope over cleanly, and a facilities manager who inherited the building in 2019 and has never been up there.
Eyebolts, and the certificate that expired in 2021
Anchor devices for rope access need periodic inspection and recertification by a competent person, and BS 7883 covers the installation, inspection and testing of permanently installed anchors. Twelve months is the working assumption for the interval. A lapsed certificate is not a technicality that a contractor can wave through, because the anchor is the thing keeping two men off the pavement of Bevis Marks.
A run of twenty eyebolts costs a few hundred pounds to test and takes a morning. I have seen that few hundred pounds hold up a £9,000 façade clean for six weeks, and I have twice seen eyebolts fail the pull test outright, which turns a cleaning enquiry into a structural conversation.
When Is Rope Access the Right Call on a City Mid-Rise?
Most of the time, at this height, on these streets. A two-man IRATA team turns up in a van, rigs on the roof, and works the elevation without touching the highway, without a licence from the Corporation, and without anybody below noticing. On a building between six and fifteen storeys with a reasonably regular façade and sound anchors, nothing else comes close on cost per elevation.
There’s a persistent argument that the Work at Height Regulations push you away from rope and towards a cradle, on the basis that collective protection ranks above personal protection in the hierarchy. I think that reading is lazy and it suits the people who hire out platforms. The hierarchy asks you to avoid working at height where you can, then to use equipment that prevents a fall, and rope access is a fall prevention system with redundancy built into it – working line and backup line, two independent anchors, a man who cannot detach from both at once. It is explicitly recognised in the regulations as a method with its own conditions attached. To treat it as the last resort misunderstands what the hierarchy is for.
What stops a rope team on a given morning
Wind, first and always. A team will come off the ropes well before the wind reaches anything a MEWP operator would worry about, because a suspended man on a 40-metre line becomes a pendulum, and in the City the wind does peculiar things – accelerated between towers, dumped downward off flat elevations, calm at the kerb and unworkable at the tenth floor.
Then geometry. Deep recesses, atria, brise soleil, plant screens that project half a metre past the glass line. A rope hangs vertically and does not care about your architect. Anything requiring the operative to work more than a couple of metres out of the vertical fall line is a re-rig, and re-rigs are where rope access loses its cost advantage against everything else.
Does a Cradle Still Make Sense at Ten Storeys?
If one is already installed and maintained, it wins on almost everything. A cradle carries kit – buckets, poles, applicators, restoration equipment, two operatives and their consumables – and lets people work with both hands and their feet on a floor. For anything beyond a wash, meaning glass restoration, sealant inspection, cerium oxide work or scratch polishing, that platform is worth more than the speed of a rope team.
Where it stops making sense is when it isn’t there. Rigging up a temporary suspended platform for a routine clean – davits, counterweights, and a rig-out and rig-down either side of the works, plus the survey that precedes all of it – carries a mobilisation cost that swallows the job. On a Great Tower Street building with no permanent provision, a temporary platform for a twice-yearly clean is difficult to justify to anyone holding the budget.
LOLER, and the six-month rule that catches buildings out
Equipment used for lifting people falls under LOLER, and thorough examination by a competent person is required at intervals not exceeding six months. That applies to a BMU sitting idle on a roof just as it does to one in daily use. Miss it, and the machine is unusable until it’s examined – which typically means an engineer visit, a lifting inspection, wire rope checks, and a lead time.
The pattern I see is a building that runs its BMU service on a twelve-month cycle because that’s what the facilities schedule inherited, and discovers the gap the week the cleaning contractor arrives. The roof decides it, and the roof has a paper trail.
One more thing that separates the two on a live building. Rope teams work outside occupied hours without much trouble – a 05:00 start on a Saturday, nobody in the offices, no interference with the entrance. A cradle on a track makes noise through the slab and needs a banksman at ground level in a busy street, which in EC3 means a permit conversation with building management every time.
Why Do MEWPs Fail on City Streets Specifically?
Reach is rarely the problem, whatever the objection sounds like when a hire company hears the postcode. A truck-mounted boom will get to fifty metres and a 35-metre spider lift will fit through a gate. The problem is what the machine needs underneath it.
Outriggers concentrate the entire weight of a vehicle and its boom into four pads. City pavements sit above vaults – old coal cellars, service voids, retained basements extending under the footway, in some cases eighteenth-century brickwork carrying a modern flagstone. Those structures were not designed for point loads of several tonnes. Spreader plates help and are not automatically enough, and on Austin Friars I’ve been on a job where the Corporation required a structural verification of the vault below before a spider lift was allowed to set up on the footway.
Then add the street itself. Narrow carriageway, red route corridors nearby, servicing vehicles for four buildings, a highways licence, traffic management, and quite possibly a night shift. The plant hire is the cheap part of that day, and on a narrow street it is not even the constraint.
Where MEWPs are the obvious answer
Low-rise elements and podium levels. Setback roofs at the fourth floor that a rope can’t reach and a BMU doesn’t cover. Atrium glazing internally, where a spider lift on a slab with a known loading capacity is straightforwardly the right machine. And any building where the anchors have failed testing and the client needs the windows cleaned next Tuesday regardless.
Out in Croydon, where Wellesley Road offices sit behind their own forecourts with hardstanding on either side, MEWPs make sense far more often than they do inside the Square Mile. It’s the same building type and a different access problem entirely.
How Should a Facilities Manager Compare the Three on Cost?
Not by the day rate, which is the number every quote leads with and the least useful one available. Compare cost per clean over a five-year cycle, including the fixed costs the building carries whether or not anybody cleans anything.
The variables that move that number are the ones nobody puts in the tender document: how many elevations can be worked from a single rig position, whether the ground floor and podium need separate provision, how many days a year the wind takes off the schedule, and what the contractor has to do about a plant screen that was added in 2014 without anyone revisiting the access strategy.
A BMU carries a servicing and examination regime all year round, plus an eventual refurbishment cost measured in six figures. Rope access carries anchor testing and a comparatively low variable cost. A MEWP carries traffic management, licensing and lost productivity in one lump every visit, with nothing between visits. At four cleans a year on a 10-storey City office, a competent rope operation will usually come out lowest, a well-maintained existing BMU close behind and better for the fabric, and a MEWP a distant third unless the building’s geometry forces it.
Frequency changes the answer more than height does
Two cleans a year and a temporary suspended platform is absurd. Twelve cleans a year and it starts to look reasonable, because the mobilisation amortises and the productivity gain compounds. Somewhere around six visits annually the lines cross, and that crossover point is the calculation nobody does before they sign a three-year contract.
Here’s what I’d tell any FM in EC2 or EC3, knowing it costs my trade money: get your eyebolts pull-tested and your BMU examination current before you go to tender, and tell the bidders you’ve done it. A contractor pricing an unknown roof prices the risk in, and that contingency is often twenty per cent of the quote. The building I know best on Fenchurch Avenue has a full anchor schedule laminated in the plant room, and every contractor who has ever quoted it has come in cheap.