Asked by the site, the day after the bill lands
Why is Bloomsbury up forty per cent on last year?
Open the site and the answer is on the same screen as the question. Billing history, meters, contracts, compliance and half-hourly consumption sit in one record, so you are comparing this period against the same period last year rather than rebuilding both from exports.
If the rise is real, the load shape shows you when it happens: overnight, weekends, or a step change on a date you can name. If it is not real, the bill validation has usually already flagged it, because an estimated read and a wrong rate look nothing alike once the invoice is rebuilt.
How consumption is read
Asked by finance, quarterly and then urgently
Are we going to land on budget?
Budgets are built from your own load shape rather than last year's total, and tracked against actual billed spend per month as the invoices arrive. Variance is per month and per site, with the sign and the size, so the conversation is about the three sites that moved rather than the portfolio average.
Months that have not been billed yet stay visibly unbilled. A forecast that quietly fills the gap with an estimate is how a budget conversation goes wrong in the second half of the year.
Budgets and reporting
Asked by you, at the point of approving it
Is this invoice actually right?
Every line is rebuilt from the contract, the meter data and the published industry charge that applied on that date, then compared line by line. 35 charge types are modelled across power, gas and water, including the network and policy charges that are usually passed straight through unchecked.
The distribution charges are priced against your own distributor's published Schedule of Charges, for all 14 GB distribution networks, on the meter's own profile class and agreed capacity. That is the difference between querying a bill and evidencing a query.
What gets rebuilt
Asked by sustainability, ahead of the report
What is our carbon number, and can you show your working?
Emissions come from the same readings that validated the bill, not a separate spreadsheet built once a year from annual totals. Consumption normalised by floor area is on the same screen, so sites compare fairly rather than by size.
The working is the point. When someone asks where a figure came from, the reading, the period and the factor are all still attached to it.
Asked by procurement, or by nobody at all
Which contracts end this year, and what happens if I miss one?
Contracts are effective-dated, so each day is priced by the agreement actually in force on that day rather than by the current one applied backwards. Expiries are visible before they are urgent.
A meter that rolls out of contract does not go quiet. It starts being billed at deemed rates, and the validation notices the rate no longer matches anything you agreed.
Asked by you, before you ask for budget
My data is a mess. Does that rule this out?
No, and it is worth being specific about why. The engine degrades rather than refuses: where half-hourly data covers the period it prices from that, where a register read covers it, it derives from the read, and where neither exists it prices from the bill's own stated total and says so on the invoice.
You are told which of the three happened on every bill, so a result built on weak evidence never reads the same as one built on complete metering. Gaps and stale feeds surface as their own finding rather than as a silently wrong number.
What we need from you
