What this page is. The questions we actually get asked on first calls,
answered directly, with the conditions attached. Where we can’t evidence something, we
say so rather than guess. Every answer below is written to be quoted straight into a tender
return or forwarded to a colleague.
Who it’s for. The project manager who carries the project P&L
and has ended up owning the welfare compound, and the bid or estimating team answering
an energy-monitoring clause against a deadline.
The answers in brief. Gaia supplies a control unit, not cabins, and you
keep your existing cabin supplier. DataMate monitors and reports, free. AutoMate adds control
and is £25 per cabin or bay per week. A qualified electrician does the install, never
your site team, and every cabin leaves with its own commissioning certificate. Savings are
measured against your own site’s benchmark, at circuit level, in kWh, week
against week, before you pay anything. The guarantee is published with its condition
attached.
What to do next. If you would rather see this on a live project than read
about it, the one button below puts DataMate on one of your sites, free. If you would rather
read first, the index further down lists every question on this page, and the full answers
follow beneath it in the same order.
If you’d rather see this on a live site than read about it:
You keep your existing cabin supplier, always. Our preferred route is to ship our units
to that supplier so they fit them during cabin setup, which means cabins arrive on site
already fitted. That's what we call a Gaia Enabled cabin: their cabin, our
box in it.
It's better for you that way. Nothing to arrange on site, and no install to fit around
your programme.
Who pays the electricity on a JCT contract, and why does that kill energy projects?
On many contract forms the contractor doesn't pay the welfare electricity bill;
the client does. Where that's true, a project P&L argument collapses on that
project, because the saving lands on somebody else's account.
So we ask it on the first call rather than discover it three months in. It's a
qualification question, not a detail.
Where the client pays, the conversation moves to carbon reporting and standardisation,
usually a client-side or bid-team discussion rather than a site one. Still worth having. Just
a different conversation, with different people in the room.
Why does a drying room cost more to run than the site office?
More high-load circuits, running at the same time.
A drying room board typically carries heaters, dehumidifiers, fans and
sockets, with heat and dehumidification working together, one driving moisture out
of the clothing and the other removing it from the air. A standard office or canteen carries
one climate appliance, a water boiler, sockets and lights.
Then add that nobody switches a drying room off on a Friday. It becomes the most expensive
room in the compound, and it's the one nobody is looking at.
We build one from your own site, and every saving is valued at what the
equipment actually draws. Two things run together.
Actual draw, not rated power. For every load we control we hold its
measured average consumption in the conditions it was running in: the load, the cabin,
the outside temperature, whether the site was manned, and whether it was inside working
hours. When one of our rules switches that load off, the saving we credit is the time it
was off multiplied by that measured average,
never its rated power multiplied by hours. A heater cycles, so its actual
average draw is materially lower than its rating. And the average is taken across
every hour in the comparison, including the hours the load drew nothing, so those
idle hours sit in the divisor and account for the periods a thermostat would have switched
it off anyway. That produces a smaller number than the rated-power arithmetic
would, and it is deliberate. The averages are refreshed daily as more sites are
monitored, so the figures behind your savings sharpen while you use the system.
We prove it week against week, before you pay anything. We monitor
until consumption has settled, then take the seven days ending the day before the control
goes on as your baseline, against the seven days beginning the day after, 14 of 15
consecutive days, the changeover day itself excluded from both periods. Same cabins, same
site, equal periods, so the difference is measured, not modelled. Both periods are read
straight off the same meter, with no scaling or normalisation applied.
The proof is your own site; the rate comes from everybody's. The two
weeks that prove the control works are your cabins and your meter, nothing pooled. What an
hour of switching-off is worth is measured against the largest record of
welfare-cabin energy behaviour in UK construction, and it grows every week, every
site we monitor, matched on those five conditions. An average built from one site is a small
number of hours and a shaky figure; built from all of them it is the most reliable estimate
of what that equipment really draws that anyone in this industry has. And we can tell
you how many measured hours sit behind any figure we give you, including when a
particular combination is thin.
Two things we don't do. We don't treat the benchmark as a fixed snapshot
taken before the automation goes on; the band averages behind every credited
saving are refreshed daily as every site we monitor adds to the record. And we don't verify
savings against your energy bills; welfare cabins
rarely sit on separate metering, so there is usually no bill isolated to the cabins. Savings
are measured against your site's own benchmark, at circuit level, in kWh.
The cabin heated an empty room all weekend: how do I stop that?
The heater circuit runs through one of the four contactors in the AutoMate unit, so it can
be switched at the distribution board rather than relying on somebody
remembering.
Door and occupancy sensors establish whether anyone is actually in the room, and timers
cover the known pattern, so the heating follows whether the room is being used, not
somebody remembering.
The same thing happens during the day, not only overnight,
tied open for a reason that made sense at the time. A heater cannot see an open door; the
sensors can.
The same detection covers this weekend example and a door propped open at midday alike.
You don't set any of it up yourself. You ask us, and we handle it.
Do I need an electrician?
No, because we do it, or your cabin supplier does. Never your site
team. Installation is included at every stage and never charged.
This is real electrical work. The circuit cable is disconnected from its
MCB, extended into our unit, and a new cable from the contactor terminates back in the MCB.
It is carried out by a qualified electrician and tested to BS 7671, with a
New Installation Certificate handed over on completion.
On your first sites that electrician is ours. At scale it's your cabin supplier's, fitting
units during cabin setup.
To be plain about it: this is not a plug-in device, and it is not
a job for a site team. The answer is no because somebody qualified does it for you, not
because it's trivial.
What stops a bad install, and what do I get to prove it was done properly?
Every cabin leaves with its own commissioning certificate. Not one per
site, one per cabin. It records the cabin ID, the installer, the date, the circuit
mappings, the load-test results, which circuits were deliberately excluded and why, and an
explicit “safe to operate” status.
That is the document to give your H&S or site-access team, alongside the BS 7671 New
Installation Certificate for the electrical work itself.
The system refuses to control some things, by design
Whatever a site asks for, the commissioning process will not put any of these on a
contactor: 32A loads, three-phase circuits, industrial loads, life-safety circuits,
incorrect cable groupings, or any circuit that cannot be positively identified.
They are all still measured; a CT goes on every one. Refusing to
control a circuit is not the same as ignoring it.
Every circuit is tested, not trusted
Before a cabin goes live each controllable circuit is switched and measured, and the
reading is checked against what the appliance label claims. A mismatch gets diagnosed rather
than accepted: wrong pole, wrong CT, a CT fitted backwards, miswired, or an appliance
that is not what the label says. Sockets are always questioned, because a
ring main tells you nothing about what is plugged into it.
This is the honest reason a Gaia install so often turns up faulty or mis-labelled kit
nobody knew about: we test the board rather than trusting its labelling.
And if something looks wrong, the install stops
The first anomaly gets a warning and an explanation. The second stops the
install. It does not proceed on a best guess; the reason is logged and the
job escalates to a Gaia engineer as a person.
The same applies if a cabin has more controllable loads than contactors:
lights come out first, then sockets, never the water boiler or the
dehumidifiers, and if that is still not enough it escalates rather than resolving
itself.
Is there a second installation when I move to AutoMate?
No. DataMate and AutoMate are the same box.
DataMate is that box in monitoring mode. Moving to AutoMate means switching control on. No
second installation, and no new hardware.
What if my cabin supplier won't fit it?
You don't need your cabin supplier involved to start; we install free on your
first sites.
When you roll out at scale we work with your supplier directly, and we already
ship boxes to a major plant hire firm who fit them at manufacture.
Lead time to a supplier is 10 working days from receipt of purchase
order.
My cabin supplier fitted the kit: who do I call?
Us.
Where a supplier fits the units they carry out the electrical installation and testing, and
Gaia attends site to commission the system, setting each unit up on the platform and checking
the installation and circuit labelling are correct. The system is ours and so is the
support.
It also changes something for your supplier. They can't fix a faulty heater they
can't see. Now they can. Faulty heaters and water heaters, wiring issues,
mis-labelled circuits and missing door closers all show up in the data, which makes it a
quality conversation rather than an argument.
My cabins come from four different suppliers. Will your data line up across them?
Yes, because we don’t use their labels.
Every supplier labels a distribution board differently. One board says
“Immersion”, the next says “Water Htr”. One says
“Ring main”, another says “Radial”. Left alone, that is four
suppliers’ worth of naming conventions and no way to add them together.
At install, every circuit is normalised to a standard set of load types.
“Immersion” and “Water Htr” both become Water boiler.
Whatever the board in front of the installer says, what lands in the data is the same short
list of load types on every cabin, from every supplier.
That is why two cabins from different suppliers, with different labelling
conventions on their boards, appear in the same report on the same terms, and
why the numbers add up rather than sitting in four separate piles.
What that gets you at programme level
Once the load types are consistent, the reporting rolls up:
organisation → project → site → building → room → load. You can ask what
every drying room across a whole programme cost you last month, and get an answer,
not four answers in four formats that somebody has to reconcile by hand.
This is usually the real question behind “will it work with my other sites?”
One standard across the programme, without anyone having to police it. The
normalisation happens at install, every time, whoever supplied the cabin.
What happens at handover and decommissioning?
It depends how the box got there.
Factory-fitted a Gaia Enabled cabin
It arrived in the cabin and it leaves in the cabin. Nothing to remove,
nothing to coordinate.
Retrofitted by us typical for a first free DataMate site
We come and remove the unit before the cabin goes back to the hire company.
Go Gaia Enabled from day one and decommissioning becomes a non-event. That's the
main practical argument for getting your cabin supplier involved early.
Does this work on a green-powered site?
Yes. We measure and control kWh at circuit level regardless of how the
electricity is generated, so the energy saving is the same.
On a generator site the saving is realised as diesel rather than kWh. At Morgan
Sindall's Maendy Primary School, generator until 10 September 2024, grid afterwards,
3,792 litres of diesel were avoided across the generator period.
Generator periods save more than grid periods, because generation is less
efficient. Maendy read 71% across its generator sub-period, 7 May 2024 to
10 September 2024, and 54% across its grid sub-period, 10 September 2024 to
31 October 2025, against 57% for the whole period, 7 May 2024 to 31 October
2025. We publish the whole-period
figure and show sub-periods underneath it; a generator result doesn't transfer to a grid
site.
Will this let me get away with a smaller generator or a smaller supply?
Yes, and it is usually worth more than the energy saving.
A site specifies its generator or its grid connection for the worst quarter-hour
it will ever see, not for what it uses on an average afternoon. Cut the peak and
you can specify smaller, for the whole job.
On a generator site that is the entire cost base. At Maendy a
60 kVA generator was replaced with a Gaia-enabled 10 kVA solar
hybrid delivering the same 590 kWh, with fuel cost per kWh falling
from £1.12 to £0.32 at £1.50 a litre, over the month
recorded, 9 May to 8 June 2024.
Being precise about what did that: peak shaving is what made the
smaller plant viable; the battery and the solar are what carried it. Load shedding
on its own does not take a site from 60 kVA to 10 kVA, and we would not claim it does.
On a grid site it is the connection. A temporary construction supply is
priced on capacity, not consumption. If the peak demands a bigger connection
than the site has, that is a five- or six-figure conversation with the network operator and a
lead time measured in months. Not having to have it is the saving.
How do I answer an energy-monitoring clause in a tender?
Take what you need from this site. It's built to be quoted.
The price is published: £25 per cabin or bay per week
The guarantee is published with its condition
Three measured case studies are web pages, each on its own URL, so you
can cite them rather than attach a PDF
The technical documents are published as HTML: wiring scheme, circuit
labelling, technical overview
If something's missing and you're on a deadline, ask. We'd rather get it to you than have
you write around a gap.
How much does welfare cabin energy monitoring cost?
DataMate monitoring is free, with installation included. Monitor-only, no
control, no commitment.
AutoMate, which adds control, is £25 per cabin or bay per week. Discounts
are available for upfront terms, ask.
On projects of six months or more it carries a guarantee: at least a 50% saving on
the energy waste in the areas we control, and if our fees come to more than we've
saved you across the project, we credit the difference. So you cannot end a
project out of pocket.
Can my site manager change the settings themselves?
No, not directly. Your site tells us what it needs, in plain English, and
Gaia makes the change. Nobody on site adjusts a rule at the board.
That boundary is deliberate. Two parties both able to touch the same heater is how kit
ends up switched off at the mains by somebody trying to help. When you flag something,
we make the change, and whatever it does next is on us, which is also
what makes the savings guarantee possible: we cannot guarantee a number if someone else can
alter the rules behind it.