Energy Growth JournalThe latest growth strategies in marketing and AI enablement for energy, solar and battery companies
Last updated Nov 13, 2026Written and edited by Ali PeivandizadehNo ads. No sponsors.
AI at Work · Action

Turn a customer's power bills into one usage sheet

By Ali Peivandizadeh, business growth and AI implementation specialist.

Published Nov 13, 2026 · Last checked Nov 13, 2026 · Last changed Nov 13, 2026
The short answer

A bank used an AI agent to put new clients' records into one standard format, and staff review what the agent is unsure about. A seller can use the same pattern on the twelve months of bills behind a business proposal. Start with ten past proposals.

Verdict
ACT
For
Energy brokers, battery and solar sellers. Sells to businesses. 5 to 100 people. Owner, sales manager.
Cost and time
The report gives no cost. The bank built a custom system with Microsoft. A small seller starts with an AI tool it already pays for. About half a day for the first ten.

Why

Northern Trust, a bank, unveiled the agent on Nov 9, 2026. It converts years of client records from many formats into the bank's standard format. Data preparation took two and a half weeks before. The bank reports it now takes three days. That number covers preparation only. The bank tested the agent on data with client names removed, and staff handle the uncertain items. A broker or a battery seller has the same kind of step. A business sends a year of bills in different layouts. Someone types the use, the demand and the charges into a sheet before any proposal can start.

What to do

  1. Time how long it takes to type one customer's twelve bills into your proposal sheet. Do it for your next five proposals.
  2. Write your standard sheet. For each month: kWh, peak demand, the energy charge, the delivery charge and the contract end date.
  3. Take ten past proposals where you still have the bills and the sheet a person made.
  4. Remove the customer name, the account number and the ESI ID from the bills.
  5. Have an AI tool the company has approved read the bills and fill in the sheet. Ask it to mark each number it is unsure about.
  6. Compare its sheet with the one a person made. Count every wrong number.
  7. Use it on new proposals only if the error count is low.
  8. On every proposal, a person checks the marked numbers, the twelve-month total and the peak demand against the bills.

What data may go in

For the test: bills with the name, the account number and the ESI ID removed. For live work: only a tool whose terms say it does not learn from what you upload, and only bills the customer agreed to share.

Who checks the output

The person who builds the proposal checks every marked number, the twelve-month total and the peak demand against the bills.

The rule to check

A savings number in a proposal must come from the customer's real numbers, and the proposal must show them. Bills and meter data need the customer's permission. A wrong number from an AI tool is still your claim.

What can go wrong

The bank reports a first-pass quality score above 96%, so staff still find and fix the rest. One wrong demand number can change the whole payback. Check the peak months by hand.

Sources

  • CIO, news report, Nov 9, 2026.
Plain text version