---
title: "How we ran this study"
publisher: "Onsomble Limited"
publication: "NZBrands.ai"
edition: 7
collected_on: "2026-10-31"
html_url: "https://nzbrands.ai/research/nz-election-2026/method/"
---

# How we ran this study

Edition 7 · answers collected 31 October 2026.

[Read on NZBrands.ai](https://nzbrands.ai/research/nz-election-2026/method/)

How we asked the election questions, compared the answers and checked selected statements.

## Study figures

| Measure | Value |
| --- | --- |
| questions | 174 |
| voter situations | 32 |
| AI assistants | 4 |
| answers recorded | 783 |
| collected | 31 Oct 2026 |

## What we set out to learn

What AI tells a New Zealander asking about the election, and what changed when we asked again.

We asked ChatGPT, Gemini, Google AI Mode and AI Overviews about the 2026 New Zealand general election. We counted party names, looked at how answers connected policies to voters’ priorities and checked selected factual statements.

This is Edition 7. We take no view on any party or policy, and no party pays to appear.

[The purpose of a study](https://nzbrands.ai/methodology#purpose)

## Who we asked as

The questions describe 32 voter situations: 23 about household circumstances and nine about life in a particular region.

These are made-up situations designed for the study, rather than a sample of real voters. They cover housing, work, income, caring responsibilities and access to services.

The priorities written into a question belong to that situation. They are not assumptions about everyone with the same background.

The question sets

| Question set | Questions | Purpose |
| --- | --- | --- |
| Household situations | 122 | Practical concerns of the 23 household situations |
| Shared national questions | 21 | Election questions without a particular voter described |
| Regional situations | 31 | Local concerns in the nine regional situations |
| Total | 174 | 20 topics across 32 voter situations |

[Writing the questions](https://nzbrands.ai/methodology#questions)

## How the questions were asked

We put the same questions to four AI assistants, starting a separate conversation for each question.

We asked all 174 questions nationwide, then repeated the 31 regional questions from their relevant locations. That gave 205 planned answers from each assistant and 820 in total.

The nine collection locations were Ashburton, Auckland, Dunedin, Hastings, New Plymouth, Queenstown, Tauranga, Wellington and Whangārei. The regional situations also describe their local circumstances in the question.

[The assistants we cover](https://nzbrands.ai/methodology#assistants)

## How we used Onsomble

Onsomble asked the questions and saved the complete answers.

We used [Onsomble](https://onsomble.ai) to repeat the saved question set through the consumer versions of these four services. We checked that the wording and locations were the same before comparing the dates.

[Working with Onsomble](https://nzbrands.ai/methodology#onsomble)

## What came back

We saved 783 of the 820 planned answers on 31 October 2026.

Three planned ChatGPT answers, one Google AI Mode answer and 33 AI Overviews answers were not saved. The records do not give an individual cause, so we do not treat missing answers as refusals or answers naming no party.

The collection finished on 31 October, with no later replacement answers included.

Planned and recorded answers

| AI assistant | Planned | Recorded |
| --- | --- | --- |
| ChatGPT | 205 | 202 |
| Gemini | 205 | 205 |
| Google AI Mode | 205 | 204 |
| AI Overviews | 205 | 172 |
| Total | 820 | 783 |

## How we count

The figures count party names in answers, rather than recommendations to vote for them.

| Term | Definition |
| --- | --- |
| Answer | The complete saved response to one question, from one assistant and one location. |
| Party named | A party counts once per answer when its name or a recognised abbreviation appears. Repeated names do not add to the count. We exclude web addresses but include visible citation labels and follow-up text. |
| Names with other meanings | We check context when a name can mean something else. National counts when it names the party; a reference to the National Library does not. The same principle applies to other party names. |
| The seven displayed parties | National, Labour, Greens, ACT, NZ First, Te Pāti Māori and The Opportunity Party. The downloads retain all 17 tracked parties. |
| Tone | The coloured bars show whether statements about a party were positive, neutral, negative or mixed. They use Onsomble’s saved automated assessments and do not show voting support or whether a statement is true. A passing comparison with another party does not count towards that other party’s tone. Supporting quotations are shown only when their words appear in the original answer. |
| Sources | Source records are available for all 783 saved answers, including three with no recorded citations. A cited website counts once per answer. Other references stay in the downloads and are excluded from the citation rankings. Visible links in the answer are counted separately. A recorded citation does not establish that it supports every statement. |
| Kinds of website | Websites are grouped as party, government, news, comparison, advocacy, business, social or other. Other includes websites that do not fit a more specific supported classification, including websites we have not yet classified. |

[Counting rules](https://nzbrands.ai/methodology#definitions)

## How we compared the answers

The comparisons use the same questions, assistants and locations in both editions.

We compared 778 answers from each edition, collected on 26 and 31 October. They answer the same questions from the same assistants and locations: 200 ChatGPT, 205 Gemini, 204 Google AI Mode and 169 AI Overviews answers per edition. Five answers from each edition could not be compared this way.

Read Edition 6 for all 783 answers and its original party counts. This edition compares only the answers available for the same question, assistant and location on both dates.

The source comparison covers 778 answers from each edition, with citations recorded in 775 of each edition’s answers. A website counts once per answer, however many links point to it.

To compare the four assistants fairly, we used only questions and locations with an answer from every assistant on both dates. Each assistant’s chart therefore covers 165 answers per edition, or 660 across the four assistants.

The location comparison puts each question’s nationwide answer alongside the answer from its location. The party names differed in 75 of 116 comparisons. A single answer from each setting cannot show whether location caused the difference.

The voting-guidance comparison covers 31 answers from each edition. A preferred shortlist puts one party or a smaller group ahead overall. Comparing parties or suggesting one for a particular priority does not, by itself, count as a preferred shortlist.

Each voter page keeps every current answer available to read. Its before-and-after issue table and source comparison use only answers available from the same questions, assistants and locations on both dates.

We read 31 voting-guidance answers and Gemini’s nationwide answer for each of the 32 voter situations in detail. We compared them with the earlier answers, checking the text and quotations before using our earlier assessments.

[Making comparisons](https://nzbrands.ai/methodology#comparing)

## What we checked

We followed up 21 selected statements from the previous edition. Twelve appeared again in this edition’s answers.

Five of the 12 statements were supported, three were wrong and three needed an important qualification. We could not confirm which party proposed the student-debt terms described in one answer; that check is labelled unclear.

Nine earlier statements were not repeated. The downloads record their absence, but leaving out an earlier error does not show that the assistant corrected it.

We checked the statements against official sources on 1 November 2026. A verdict applies to the selected statement rather than the entire answer, and does not estimate the accuracy of all 783 answers.

[Read the checks](https://nzbrands.ai/research/nz-election-2026/editions/7#checks)

## What this can and cannot tell you

This is a dated record of answers, rather than a prediction of what every voter will hear.

AI assistants can answer the same question differently. One answer from each setting cannot show whether a change is lasting or whether location caused it. This study does not measure influence on votes or establish political bias.

All 783 answers were included in the party-name counts. Detailed readings covered 63 selected answers and were assisted by AI; this was not a factual review of all answers or an independent second review.

The study does not verify every policy statement or establish anyone’s personal entitlement. An earlier error disappearing from a later answer is not, by itself, a correction.

The source rankings identify hosting websites, rather than the authors of every page or post. A Facebook citation can point to material posted there by another organisation, and citation frequency does not establish accuracy, quality or political bias.

[Limitations](https://nzbrands.ai/methodology#limitations)

## The records

Every saved question and answer stays available with the tables behind this edition.

Download the tables behind the findings, or open a question to read every saved answer in full.

Each earlier edition retains its own findings, voter pages and original answers. Read older answers in the context of their collection dates.

[Explore the evidence](https://nzbrands.ai/research/nz-election-2026/editions/7/evidence)

[Evidence guide (Markdown)](https://nzbrands.ai/research/nz-election-2026/evidence.md)

## Publication and corrections

Published and funded by Onsomble Limited. NZBrands.ai takes no view on any party or policy.

[Independence](https://nzbrands.ai/about/independence/) · [Corrections](https://nzbrands.ai/about/corrections/) · [Contact](https://nzbrands.ai/about/contact/)