"We just use the cut-out rating as max demand — it always goes through."
That is a UK solar installer, quoted anonymously in Renbee's November 2025 response to Ofgem's ED3 consultation. Renbee sell DNO application software, so read their evidence with that interest in mind. The quote still lands, because you have heard it said in your own office, possibly by you.
Two fields on the application do this. The estimated maximum demand of the property after your installation, and the rating of the service cut-out. Neither was on the form you learned the trade on. Nobody has taught you how to produce them. Renbee's own reading of the installer comments they gathered is the fairest one available:
"These comments reflect the confusion and procedural fatigue faced by installers, more than malice. They are being asked for new data points, such as estimated maximum demand post-installation or the cut out rating, without sufficient guidance or training to work this out."
The two fields answer different questions
The cut-out rating is a fact about the supply. It is the rating of the fuse sitting in the service head, and it tells you what the network has protected the property to. You read it off. It is an upper bound imposed on the house from outside.
Estimated maximum demand after installation is a forward-looking judgement about the property. It asks what that house is expected to draw at once once your work is finished. It depends on what is already in the building, what you are adding, and what you assume about how much of it runs at the same time.
Answering the second with the first tells the DNO that the house's demand is exactly equal to its fuse rating. For any real house that is either wrong or a coincidence. It survives because it looks plausible on a form, not because it is right. "It always goes through" is a statement about the assessment process, not about the number.
Under-declaring capacity is not a workaround
The other quote in Renbee's evidence is the one that should worry you.
"We'll apply for 3.68kW even though we're installing 5kW — the customer wants maximum export."
Two separate problems, and the second is worse. The first: the declared system is not the installed system — the record on file does not describe the array on the roof. The second is the reason given. The customer's export ambition is a commercial preference. The declared capacity is a statement to the network operator about what will be connected to their infrastructure. When the first is allowed to set the second, you no longer have a design. You have a number chosen to clear a gate.
The Only Panels installer podcast put the stakes plainly in an episode titled "Are DNOs Forcing Installers to Break the Law?", flagging the liability, and the potential criminal offence, of installing grid-tied 12kW systems with export limiters without prior DNO permission. Wherever you land on how fair that is, note where the exposure sits: with the contractor who signed the application, not the customer who wanted the export.
Nobody is calling installers crooks. Renbee explicitly do not, and neither do we. But they also report that "compliance rates may be as low as 50%", with "some installers … completing thousands of installations without notifying the relevant DNO; others apply creative 'hacks' to make the process workable." Set that beside their own figure of over 120,000 MCS-certified solar PV systems completed in the first half of 2025, up 36% on 2024. Whatever the true rate, the number of systems unregistered or creatively registered is large, and growing.
Nobody has published the method
This is the point in the article where you would normally be given a formula. You are not getting one, because we cannot source one.
As far as we can find, neither the ENA nor the DNOs have published installer-facing guidance that makes estimated maximum demand for a domestic property calculable at the point of quoting — by which we mean a stated method, with stated assumptions, that an installer could apply during a survey and defend afterwards. That absence is the root cause of the guessing. No amount of installer training fixes a method that has not been written down.
So treat any confident formula with suspicion, including one from a software vendor, including one from us. Ask where it came from, and which DNO has accepted it in writing. If the answer amounts to "it always goes through," you have read this article already.
The process's own numbers suggest the fields are not being answered in a form the system trusts. Renbee report that over 60% of applications submitted via Connect Direct are escalated for manual review, while only 0.6% are automatically approved. Whether the fault lies in the answers or in the assessment, the effect on your week is identical. Your application joins a queue and waits for a human.
What you can actually control
Three things are fully in your hands, and none of them require a standard that has not been published.
Photograph the cut-out properly. On Renbee's account the ENA's automated reader is the weak link: "the tool fails to read cut-out ratings, and fails applications especially when the cut-out is partially obscured, located in tight spaces (garages, under-stairs cupboards), or poorly lit. This leads to installers having to revisit the site for new photos." So take a torch. Move the coats and the shoe rack. Get the rating in focus, square on, legible, and take a second wider shot that shows where the cut-out is in the property. A failed read on a dark photo in an under-stairs cupboard buys you a second site visit and nothing else.
Record the figures you used and the reasoning behind them, on the day you use them. Whatever you enter for estimated maximum demand, write down what it was based on: what you read off the cut-out, what electrical load is already in the property, what you are adding, and what you assumed about how much of it runs simultaneously. With escalation running above 60%, escalation is the normal path and not the exception. An escalated application you can answer costs you an email. One you cannot answer costs you a resubmission, and if the surveyor has since left the firm, another visit.
Let the design set the declared capacity, and never the export target. If the customer wants 5kW, apply for 5kW. If the network says no, that is a conversation about the network, and it is a conversation you can have. Applying for 3.68kW does not move the constraint. It moves the liability onto you.
The handoff nobody built
Both numbers exist at survey. The cut-out rating is read off a fuse by someone standing in front of it holding a phone. The maximum demand estimate depends on what is in the property and what you are adding, which is precisely what the design already knows. Then both are re-entered, from memory or from a camera roll, at a keyboard, weeks later, quite often by someone who was never on site.
That is a data-handoff problem before it is a compliance problem. The survey knows the cut-out rating; the form does not. A design tool ought to carry both figures, and the reasoning behind them, from the roof into the application rather than leaving them to be reconstructed at a desk. Solar Engine does not file DNO applications and does not compute maximum demand — no tool can compute it against a published standard, because there is no such standard. The handoff, though, is solvable today, and it is where design software still owes installers something it does not deliver.
On the next job
Photograph the cut-out with a torch, unobstructed, rating legible, plus a wider context shot. Declare the capacity you are installing. Write down the maximum demand figure you used and what you based it on, in the job record, the same day. Expect the application to be escalated and have the answer ready before it is asked for.
And if the ENA or any DNO publishes a method for estimating maximum demand at the point of quoting, read it the day it lands. It is the missing piece. Everything above is what you do until it exists.