Twenty-one electricity retailers, assessed against the same checklist in the same window, July 2026. A grade here is not one company reviewed in isolation. It is where a company stands among the people it competes with, which is the only thing that makes a grade mean anything.

Every graded claim quotes the company's own published wording, carries the capture it was taken from, and names the standard it was tested against. Every company was contacted before publication.

Open the public register → Download the full report (PDF, 4.4 MB)

What we tested, and how much of it held up

DimensionClaimsSubstantiatedShare
Social28619568%
Commercial38723460%
Environmental1468458%

819 public claims tested in total. A claim is substantiated when the company's own published evidence supports it. Where it does not, the finding says what would close the gap.

The grades

Ordered by overall score. Every company in the scan appears here, whatever the result. Open any of them for the findings, the quoted wording and the evidence.

RetailerOverallScoreEnvironmentalSocialCommercial
Octopus EnergyA96.0A- 88.0A 100.0A 100.0
Contact EnergyA-87.1B 70.0A 100.0A 91.4
Meridian EnergyA-81.4B+ 80.0A 90.0B+ 74.3
Toast ElectricB+79.0B- 64.0A 90.0A- 82.9
GlobugB+76.5N/AB 70.0A- 82.9
Genesis EnergyB+74.3B- 60.0A- 80.0A- 82.9
2degreesB+73.7A- 85.0B+ 76.0B- 60.0
Electric KiwiB67.7B 70.0B+ 76.0C+ 57.1
TensorB-66.4C- 40.0A- 85.0B+ 74.3
Nova EnergyB-65.6B- 64.0C 50.0A- 82.9
Frank EnergyB-62.4B 70.0B- 60.0C+ 57.1
Mercury EnergyB-61.9B 67.0B 70.0C 48.6
Just EnergyB-61.8B 70.0B- 64.0C 51.4
Grey Power ElectricityC+58.5B- 60.0B- 64.0C 51.4
Nau Mai RāC+55.0N/AB- 60.0C 50.0
PowershopC+53.7C- 40.0B- 64.0C+ 57.1
SlingshotC52.0N/AB- 60.0C- 44.0
EcotricityC48.1C 50.0B- 60.0D+ 34.3
Wise Prepay EnergyC48.0C- 40.0C 52.0C 52.0
Black Box PowerC47.8C- 40.0B- 60.0C- 43.3
Pulse EnergyC47.6C- 40.0B- 60.0C- 42.9

A grade rules on whether a public claim is substantiated. It is not a finding about environmental performance, truthfulness, or the merit of the business.

What we found

Finding OneOffers and their conditions are published in different places

Across the 21 audits, the most common problem is not an inaccurate claim but a misplaced condition: the offer sits on one page and the limit on it in a footnote, an expandable answer, a terms document, a separate fee schedule, or a page behind a labelled link. The information is published; the person deciding may never see it.

Finding TwoContract claims and the terms that qualify them

Twelve claims across eight retailers tell customers some version of no contract, leave whenever, no penalty.

Finding ThreeWhat retailers do not publish about their own record

Four diagnostic flags test whether a retailer is silent on something material to the claims it does make.

Finding FourWhat arrives at your house is not exactly what your retailer generates

A renewable percentage describes the generation a company owns, not the electricity that reaches your home. No retailer created that gap and no grade moves on it, but it changes what every renewable claim in the sector means. If a household chooses a provider on its renewable percentage, it is reading a figure about the company's portfolio as a figure about the power in its own wall, and it is not seeing the whole truth.

Finding FiveComparative claims made without a stated basis

A comparative claim needs three things: a comparator, a method, and a date. These are the retailers' own words, and almost none of them carries any of the three.

Finding SixComparing prices between retailers

New Zealand has comparison tools. Powerswitch is the best known; Billy is backed by the Electricity Authority. A customer who wants to test a "competitive price" claim goes to one of them.

Finding SevenReaching a retailer before publication

Before publishing, we wrote to every retailer in this scan and then telephoned to confirm a person had received it. It is the most ordinary corporate task there is, and it largely did not work; the reasons were structural, not bad luck.

Each finding is set out in full in the report, with the wording assessed, the evidence behind it and what would close it.

Get the next one

We scan a market end to end, or not at all. Which market comes next is decided by who asks for it. Leave your email and we will tell you what we have audited and what we found.

Occasional email, no more. You can stop at any time, and your address is never shown or sold.

Want a sector scanned? Ask for one.

How this was done

Claims were collected from each retailer's own public surfaces, classified, and graded against the HowLegit E-S-C framework. Read the published methodology.

Every retailer was written to and telephoned before publication. Where a company did not reply, its register entry says so. If you are one of the companies here and something is wrong, the right of reply is open and free, and our grievance process covers anything about how we behaved rather than what we graded.