Old billing system
0
accounts, 38 tariffs, 6 years of history
New billing platform
Go-live
every account must bill the same, or better explained
Today A team checks a sample of 200 accounts by hand and hopes the rest behave.Behind the scenesBilling errors cluster in the unusual accounts: split-rate meters, estimated reads, mid-period tariff changes. A random sample of a few hundred rarely hits them.What it achievesChecking every account finds the small groups that cause the complaints after go-live.
Bills recalculated in the new system
0
twelve months for every account, compared line by lineBehind the scenesThe last twelve months of meter reads and tariffs are loaded into the new platform in a test copy, every bill is produced again, and each line is compared with what the old system charged.What is automatedProducing and comparing all the bills. People only look at the differences.In this demoInvented supplier, tariffs and customer.
Old system, Economy 7 customer, one month
| Day units | 212 kWh × 31.2p | £66.14 |
| Night units | 164 kWh × 14.6p | £23.94 |
| Standing charge | 31 days × 53.4p | £16.55 |
| VAT at 5% | £5.33 | |
| Total | £111.96 |
New system, same month
| All units | 376 kWh × 31.2p | £117.31 |
| Standing charge | 31 days × 53.4p | £16.55 |
| VAT at 5% | £6.69 | |
| Total | £140.55 |
£28.59 more
Every difference, grouped by its causeBehind the scenesDifferences are grouped by what caused them: which line moved, on which tariffs, for which kind of account. Each group is traced to a rule or a mapping in the new system, not left as noise.What it achievesA few causes explain almost all the differences, so the team fixes rules instead of chasing accounts one at a time.
Before fixes, 83.2% of accounts bill the same.
The one that matters
2,140 Economy 7 customers
The new system charges their night units at the day rate. Each would pay about £343.08 a year too much, £734,191 across them, until the complaints arrived.Behind the scenesSplit-rate meters are a small share of accounts, so a sample often misses them. Comparing every bill finds all of them in one group, with the tariff mapping that caused it.AssumptionThe rates and usage are invented; the kind of error is a real risk whenever tariffs are mapped between systems.Could go furtherKeep the mapping checks running after go-live, so a tariff added later cannot repeat it.
Same customer, after the fix
| Day units | 212 kWh × 31.2p | £66.14 |
| Night units | 164 kWh × 14.6p | £23.94 |
| Standing charge | 31 days × 53.4p | £16.55 |
| VAT at 5% | £5.33 | |
| Total | £111.96 |
Matches the old system
Accounts billing the same
83.2%99.97%
Go or no-goBehind the scenesThe evidence for every account sits behind each line: which bills were compared, what differed and why. The programme board signs off on that, not on a sample.What it achievesThe switchover date is set by evidence, and the regulator and customers can be shown it.
0
bills in the first month
0%
checked against the old rules before sending
0
held for a person to look at
0%
fewer billing complaints than the last migration (modelled)
The same comparison keeps running after go-live, so every first bill in the new system is checked before it reaches a customer.AssumptionInvented supplier and volumes. Modelled: bills held, the fall in complaints, and 6 days of sample testing replaced by 3 hours of review each run.What it achievesThe real figures come from running your own last twelve months through both systems.
Switchover readiness
99.97%
of 124,800 accounts billing the same
Interactive demo
Watch it run in 89 seconds, or click through at your own pace. The i buttons show what happens behind the scenes.
That was one day
£734,191 a year of overcharging caught before go-live
Modelled on an invented supplier. Your own last year of bills would give the real figure.