I am making a request under the Official Information Act 1982 for the following information held by New Zealand Police.

Definitions used in this request

For the purposes of this request:

- "Social media post" means any publicly or semi-publicly visible communication made on a third-party online platform, including but not limited to X (formerly Twitter), Facebook, Instagram, TikTok, LinkedIn, YouTube, Reddit, Bluesky, Threads, blogs, and online comment sections.
- "Police contact" means any of the following, where the originating reason recorded by Police relates wholly or substantially to the content of a social media post: (a) a phone call, email, or other communication initiated by Police to the author of the post; (b) a request that the author attend a Police station or meet with Police; (c) a visit by Police to the author's home, workplace, or other location; (d) a formal interview, whether under caution or not; (e) a recorded warning, prevention-first conversation, or file note; (f) any arrest, charge, or prosecution.
- "Government-originated referral" means any complaint, request, referral, or notification received by Police from a Minister, Ministerial office, government department, Crown entity, statutory officer, local authority, or any employee or contractor of the foregoing acting in their official capacity.

Information requested

1. The number of Police contacts (as defined above) with members of the public arising from a social media post, broken down by calendar month, for the period 1 November 2020 to 31 October 2025 (a 5-year window).

2. For each month in the same period, a breakdown of those contacts by category of outcome, including at minimum: (a) no further action; (b) warning or prevention-first conversation; (c) formal interview without charge; (d) charge filed; (e) other (please specify categories used).

3. For each month in the same period, a breakdown of the originating source of the complaint or referral that led to the contact, including at minimum: (a) member of the public; (b) social media platform itself; (c) another Police district or unit; (d) a government-originated referral as defined above; (e) Police-initiated (proactive monitoring); (f) other.

4. For contacts originating from a government-originated referral, the name of the referring Minister, agency, department, or Crown entity, and the number of referrals from each, broken down by year.

5. Any internal Police policy, guidance, standard operating procedure, training material, or instruction document currently in force, or in force at any point during the requested period, that addresses how Police should respond to:
(a) complaints about the content of social media posts;
(b) speech that is lawful but considered offensive, hateful, or harmful;
(c) referrals from government agencies regarding online speech;
(d) the application of section 61 of the Human Rights Act 1993, the Harmful Digital Communications Act 2015, or section 14 of the New Zealand Bill of Rights Act 1990 to social media content.

6. The total number of Police staff (FTE or headcount) assigned at any point during the requested period to roles whose primary or substantial function involves monitoring, triaging, or responding to online or social media content, including any specialist team, unit, or programme dedicated to this purpose. Please identify the team(s) by name.

Format and scope

I am content to receive aggregate numerical data in spreadsheet form (CSV or Excel). I am not seeking the names of any private individuals contacted by Police, nor the content of specific posts, and I accept that personal information of that kind is properly withheld under section 9(2)(a).

If any part of this request is likely to be refused under section 18(f) on grounds of substantial collation or research, I ask that you contact me under section 18A to discuss how the request might be refined to make it manageable. In particular, I would welcome advice on:

- whether NIA, the Case Management System, or any other Police database holds a flag, code, or tag that identifies social-media-related contacts;
- the earliest date from which such data is reasonably retrievable;
- whether the data can be produced for a shorter period (e.g. 3 years) without triggering s 18(f);
- whether any of the requested categories above are recorded in a way that allows extraction without manual file review.

Please acknowledge receipt and confirm the date by which I can expect a substantive response under section 15(1).

If it's all the same to you, I'd prefer to do this in writing rather than by phone, simply so we both have a clear record of what's discussed and agreed.

To refine sensibly, it would help to understand where the difficulty actually lies. Could you let me know:

1. Whether NIA or the Case Management System holds any flag, code, tag or field that identifies a contact as arising from a social media post - i.e. whether items 1–4 are extractable by query, or only by manual file review;
2. Whether the source of a complaint or referral (member of the public, the platform, another Police district/unit, or a government agency) is recorded in a way that can be extracted;
3. The earliest date from which any of this is reasonably retrievable; and
4. Whether a shorter period (e.g. three years) or annual rather than monthly figures would make items 1–4 more achievable.

On scope, a few things that should help narrow this:

– Items 5 (policies, guidance, SOPs and training on responding to social-media complaints, lawful-but-offensive speech, government referrals about online speech, and the application of HRA s61 / HDCA / BORA s14) and 6 (whether any team or unit monitors or triages online content, its name, and headcount) are document and staffing matters that shouldn't depend on any data extraction..

If it's easiest, you could set out in writing what the systems can and can't produce, and I'll refine the request to fit. I understand a reframed request may be treated as a new request for timeframe purposes; if so, I'd appreciate the readily-available parts (items 5 and 6) being dealt with under the current request where possible.

Hi Greg,

Thanks for the detail on NIA's limitations. I'd still prefer to do this in writing so we have a clear record. Having reviewed the NIA Manual released under IR-01-25-40381, I think there's more extractable from structured fields than your email suggests, and I'd like to propose a narrowed scope based on what the system actually records.

The Create Occurrence wizard (Chapter 15) lists "Reporting channel" as a mandatory field. NIA holds specific offence codes for the HDCA offences - 1765, 1766, and 1767 between them covering ss 21, 22, and 22A - plus 7431 for inciting racial disharmony and 6P for bullying of children and young persons. The case query function (Chapter 9) allows filtering by offence/incident code across a date range. The "Hate" contributing factor is mandatory for every incident/offence code on an occurrence, with structured sub-fields for hate type and perceived prejudice. Case outcomes are recorded as clearance types (arrest, summons, field resolution, etc.) in Chapter 10. None of this requires narrative searching.

I accept there's no single "social media post" tag. But a case query filtered to those codes would produce a meaningful dataset from structured fields alone. With that in mind:
Items 1–4 (revised)
1. The number of occurrences per calendar year (Nov 2020 – Oct 2025) bearing any of: codes 1765, 1766, 1767, 7431, or 6P. I'm content with annual figures. If Police considers other offence or incident codes relevant, I'm happy for those to be included.
2. For those occurrences, a breakdown by clearance type.
3. For those occurrences, the "Reporting channel" values recorded at creation, aggregated. If a full breakdown isn't practical, I'd accept confirmation of what the available values are and a count for any that can feasibly be reported on (e.g. the "Reported by 105 Call Centre" filter visible in the occurrence query).
4. For those occurrences, the number where the "Hate" contributing factor was flagged "Yes," broken down by hate type and perceived prejudice sub-category, by year.

If even this narrowed scope triggers s 18(f) concerns, I'd welcome advice on which of items 1–4 are feasible and which aren't.

Items 5 and 6 (unchanged)

These are document and staffing matters that don't depend on data extraction, so I'd ask they be dealt with under the current request.
5. Any internal policy, guidance, SOP, training material, or instruction document in force during the requested period addressing: (a) complaints about social media posts; (b) lawful-but-offensive speech; (c) government referrals about online speech; (d) application of HRA s 61, the HDCA, or BORA s 14 to social media content. The three proactively released chapters you linked are partially responsive - I'm asking whether any further internal guidance exists beyond those, such as the "Online Reporting Guide" referenced in the NIA Manual (managed at Kapiti FMC), any guidance from the Police Cybercrime Unit, or district-level instructions. If nothing further exists, confirmation of that would suffice.
6. Any specialist team or unit whose primary or substantial function involves monitoring, triaging, or responding to online/social media content - names and staffing levels. The NIA Manual references a "Digital Online team" and the HDCA chapter references the Police Cybercrime Unit as Police's Netsafe liaison - I'm asking for confirmation of these and any others.

Hi again Greg,

Just a follow-up to my last reply to correct the date range for Item 1:

1. The number of occurrences per calendar year (Nov 2020 – present) bearing any of: codes 1765, 1766, 1767, 7431, or 6P. I'm content with annual figures and understand the current year will be partial. If Police considers other offence or incident codes relevant, I'm happy for those to be included.