Somewhere in most qualifying UK businesses there is a person who spends a fortnight each year assembling the energy figures for the directors' report. They chase meter readings, reconcile invoices, rebuild last year's spreadsheet, and produce a number under deadline.
Then the file goes in a folder and the whole thing happens again in twelve months, from scratch.
What SECR actually asks for
Streamlined Energy and Carbon Reporting requires qualifying large UK companies and LLPs to disclose energy and carbon information in their annual reports. In outline, that means:
Energy consumption — typically electricity, gas and transport fuel, in kWh.
Associated greenhouse gas emissions, converted using the published conversion factors for the relevant year.
At least one intensity ratio — emissions against a business metric you choose, such as output or floor area, so the figure means something year on year.
The methodology used to arrive at all of it.
Energy efficiency action taken during the period.
The prior year's figures for comparison, after the first year.
Check the current thresholds and the precise requirements against the guidance for your reporting year and company type — they are specific, and they are not what this piece is for. What matters here is the shape: it is an evidenced, methodology-stated, year-on-year comparable disclosure. Which is to say, an audit-trail problem wearing an energy costume.
Why it gets done by hand almost everywhere
Because nobody ever decided to do it by hand. It accumulated.
The data lives in several places that were never connected: half-hourly meter data with the electricity supplier, gas in invoices, fleet fuel in an expenses system, sub-meters in a building management system nobody has queried in years, and one site whose readings genuinely are written on a clipboard.
Pulling those together is a person's job because connecting them was never anyone's project. And it stays that way because it is only painful for two weeks a year — which is exactly long enough to be miserable and exactly short enough never to reach a budget.
What the manual version costs beyond the fortnight
Errors you cannot see. Transcribed readings, a conversion factor from the wrong year, a site double-counted. The output looks like a number either way, and it is going into a published document.
Methodology that lives in someone's head. When they leave, the next person reconstructs it, differently — and your year-on-year comparison quietly stops comparing.
No visibility for eleven months. The reporting only tells you about a year you can no longer influence. Anything you might have acted on has already happened.
An answer you cannot evidence. If someone asks how a figure was produced — an auditor, a customer's ESG questionnaire, an investor — the honest answer is a spreadsheet and a memory.
What automated looks like
The chain is not complicated; it is just never anyone's job. Consumption data is collected from where it already lives — supplier data feeds, the building management system, sub-meters, fuel cards — on a schedule rather than in a panic. Conversion factors are versioned, so a figure calculated in March and re-run in November gives the same answer, and you can show why.
The intensity ratio calculates itself from the business metric you already track. The disclosure assembles from the same data every time, so the methodology is the code rather than a habit. And the record behind every number is retained, so "where did this come from?" has an answer that takes a minute rather than a week.
The by-product is the part clients end up valuing most: once the data is flowing for the report, it is flowing all year. You stop finding out about last year and start seeing this week.
The honest boundary
This is measurement, record-keeping and reporting. It will not reduce your consumption — that takes an intervention, and anyone offering you a savings percentage before establishing a baseline is quoting a number about someone else's building.
What it will do is make the annual panic stop, make the figures defensible, and give you eleven months of visibility you currently do not have. That is the layer we build — to the same standard we build for financial regulators. If you want to know what your existing data could support, a Reality Check is a week and a fixed fee.
This is the part we do — the crossing from a demo to a system that survives production.