Industries / Energy, water and waste / Electric transmission and distribution
Metering
Metering measures how much electricity each customer uses, reads that figure, and uses the data for billing and for running the network.
7
Milestones achieved, from 1882 to 2018.
1
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
In depth
The same ground is charted in more depth, as an industry of its own.
The road
AchievedWatch forOpen
Achieved
1882
The chemical meter
Edison's first meters measured electricity by letting it plate zinc onto electrodes, which were weighed to show how much had been used. The Henry Ford museum holds one from about 1882 that was part of the Pearl Street system and calls it one of the earliest surviving electricity meters.
Credit
Thomas Edison
1888
A meter for alternating current
In 1888 Westinghouse engineer Oliver Shallenberger found, by accident, the principle for an AC meter: in effect a small motor whose turning counts the watt-hours a customer uses. IEEE's history site says the meter "was known worldwide by the end of the century."
Credit
Oliver B. Shallenberger; Westinghouse
2006
A whole country read remotely
Enel, Italy's largest distributor, replaced mechanical meters with electronic ones read over the power wires. In July 2006 Italy's energy regulator ARERA reported about 24 million meters already read and billed automatically, and in 2010 it reported 30 million installed by mid-2009, about 90% of Italian meters.
Credit
Enel Distribuzione; ARERA
2009
Europe sets an 80% target
The EU's 2009 electricity directive told member states to assess smart meters by 3 September 2012 and, where the assessment was positive, to fit at least 80% of consumers by 2020. In 2014 the European Commission counted 16 member states committed to large rollouts, close to 200 million meters, about 72% of EU electricity consumers.
Credit
European Parliament and Council; European Commission
2012
Customers get their own data
Green Button, launched in January 2012 after a White House call to action, lets customers download their detailed electricity use in one standard file format, published by the energy standards body NAESB with help from NIST, the US standards agency. The Department of Energy reports that more than 50 utilities and suppliers signed on, covering more than 60 million homes and businesses.
Credit
NAESB; NIST; participating utilities
2015
Victoria's costly rollout
By March 2015 Victoria, Australia, had installed 2.79 million smart meters, a year later than first planned. The state Auditor-General estimated consumers had paid A$2.239 billion in metering charges by the end of 2015, 11.4% over the distributors' forecasts, while most benefits had not yet arrived.
Lessona rollout can be paid for long before its benefits show up.
Credit
Victorian Auditor-General's Office (findings)
2018
Britain's meters lose their smarts
In 2018 the National Audit Office found 12.5 million first-generation smart meters installed in Great Britain against 5.4 million planned, about 70% of which lost their smart functions when a customer changed supplier. In 2023 it reported that only 57% of meters were smart, and about 3 million of those were not working in smart mode.
Lessoninstalling meters before the shared communications network was ready left millions that work no better than old meters.
Credit
UK National Audit Office (findings)
Watch for
Meters that spot faults
Sense says its software, running inside new smart meters, detects arcing, downed lines and wearing equipment in real time, and that pilots found developing faults caused by vegetation. It names no pilot utility and gives no figures.
Achieved whena named utility or outside party publishes field results.
Claimed by
Sense, February 2026
Open
01
Testing tariff readers
A public test set of electricity tariffs (the rate schedules that turn a meter reading into a bill) with every charge checked by hand, and a tool that reads a tariff document into structured data, scored against it.
The briefCredit · unclaimed02
Forecast street demand openly
A public benchmark of next-day demand forecasts for individual low-voltage feeders (the cables that run from a local substation along a street), using the aggregated smart meter data British distribution companies now publish.
The briefCredit · unclaimed
Open 01 · the brief
Testing tariff readers
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public test set of electricity tariffs (the rate schedules that turn a meter reading into a bill) with every charge checked by hand, and a tool that reads a tariff document into structured data, scored against it.
Why it is still open
The US Utility Rate Database covers over 3,700 utilities, but its maintainers update rates each year only for about 150 utilities serving 70% of US load, and say they are looking for a "less labor-intensive" way to gather tariffs. The firm Nectar says it is building a free, machine-readable database of US tariffs with each value cited, aimed at 2027.
Built with what exists
Tariff documents from utility websites and state commission filings; vision-language models that read PDF tables; open-weight models for local runs (see the Note "A trillion open parameters"); the database's existing records as a first answer key; two human checkers.
First step
Take 20 residential tariffs updated in the last year, compute a sample bill by hand and from the database record, and count the mismatches.
Done looks like
200 tariffs from at least 50 utilities with every charge hand-checked, and the tool reproducing a year of monthly bills for three sample customers within 1% on at least 90% of tariffs.
Open 02 · the brief
Forecast street demand openly
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public benchmark of next-day demand forecasts for individual low-voltage feeders (the cables that run from a local substation along a street), using the aggregated smart meter data British distribution companies now publish.
Why it is still open
GridLive, a University of Sheffield project, gathers in one place the feeder-level smart meter data published by five British distribution companies, but publishes no forecasts or benchmarks.
Built with what exists
GridLive's collected data, under each company's license; open forecasting models such as Chronos-2; public weather data; Python.
First step
Download one company's data for 100 feeders and score the baseline.
Done looks like
Forecasts for at least 1,000 feeders from at least three companies over six months, scored by mean absolute error against a same-day-last-week baseline, with code and scores public and a second person reproducing them.