This assessment is published to satisfy the duty under section 9 of the Online Safety Act 2023(illegal content risk assessment) for user-to-user services. It explains the illegal-content risks posed by School Atlas's user-to-user feature set and the mitigations in place.
1. The Service
School Atlas (schoolatlas.co.uk) is a UK school search and information service operated by Stephen Spence (sole trader). It provides parent-facing information about UK schools and a paid portal for schools to manage their profile.
For the purposes of the Online Safety Act, School Atlas is a user-to-user service because registered users may submit reviews of UK schools that are published to other users after moderation. It is not a search service for OSA purposes.
2. User base and likely audience
- Minimum account age: 16, self-declared and enforced by email verification at signup.
- Primary audience: UK parents, prospective parents, and school staff researching schools for their children or themselves.
- Reviews are a small fraction of total engagement. The overwhelming majority of traffic is to read-only school profiles, guides, and tools.
- No advertising, no user discovery, no messaging, no group or forum features.
- See also our Children's Access Assessment.
3. Functionalities that could increase risk
The user-to-user surface consists of text reviews (rating, title, body up to 1,000 characters, category scores, relationship to the school) and institution replies published beneath reviews by verified school accounts. There are no other user-generated surfaces.
Risk-elevating features that are absent:
- No private or direct messaging between users.
- No image, audio, or video upload by end users.
- No livestreaming.
- No user discovery, friending, following, or contact lists.
- No group chat, forum, or threaded discussion.
- No ephemeral or disappearing content.
- No end-to-end encryption.
- No anonymous posting — all reviews tied to a verified account.
4. Priority illegal content: per-category risk
We have assessed each of the seventeen priority illegal-content categories listed in Schedule 5, 6, and 7 of the Online Safety Act 2023 against our specific service. Ratings reflect the residual risk after mitigations (see section 5).
| Priority offence | Residual risk | Reasoning |
|---|---|---|
| Terrorism | Negligible | Reviews are the only user-generated surface. Content scope is restricted to first-hand experience of a named UK school. Pre-publication human moderation rejects any material outside that scope. |
| Child Sexual Exploitation and Abuse (CSEA) | Low | No image or video upload by end users. No direct messaging. No user-to-user discovery of other users. Reviews are text-only, 1,000-character maximum, moderated before publication. Account minimum age 16 with email verification. |
| Grooming | Negligible | No private messaging, no friend/contact graph, no user discovery, no child-facing accounts. Reviewer identities are not displayed. Minors have no route to contact adults through the service. |
| Encouraging or assisting suicide / serious self-harm | Low | Moderation policy rejects content naming individuals, describing safeguarding disclosures, or containing personal distress narratives. Human moderator escalation to safeguarding contact if self-harm indicators detected. |
| Hate offences | Low | Moderation policy explicitly rejects hate speech, slurs, and content targeting protected characteristics. AI pre-flag plus human review. Reported content taken down within 48 hours. |
| Harassment, stalking, threats, and abuse | Medium | This is the category most closely aligned with the service’s user-to-user surface: a review about a specific school could in theory target a named individual. Pre-publication human moderation specifically rejects content that names staff, pupils, or parents, or that amounts to a threat. Reported content is removed on receipt of a complying notice. The residual risk is rated Medium (not Low) to reflect the realistic possibility that a moderator misses a borderline case; our mitigation is the notice-and-takedown path at moderation@schoolatlas.co.uk and the right-of-reply for schools. |
| Controlling or coercive behaviour | Negligible | No interpersonal relationship features, messaging, or account linking. |
| Drugs and psychoactive substances | Negligible | No commerce, no anonymous posting, no group chats. Reviews describing school drug policies are in scope; content facilitating drug supply is not and would be rejected at moderation. |
| Firearms, knives and other weapons | Negligible | As above. |
| Unlawful immigration and human trafficking | Negligible | As above. |
| Sexual exploitation of adults | Negligible | No image/video upload, no direct contact features. |
| Extreme pornography | Negligible | Text-only reviews. Sexual content is off-scope and rejected at moderation. |
| Intimate image abuse | Negligible | No image upload by end users. |
| Proceeds of crime | Negligible | No marketplace or payment-to-user features. All inbound payments are Stripe subscriptions for the platform itself. |
| Fraud and financial services offences | Low | Limited surface: reviews could in theory describe fraudulent schemes, but the moderation policy rejects content outside first-hand experience of a school, and commercial content is explicitly barred. Institution portal access is gated by verified email and ownership checks. |
| Foreign interference | Negligible | Platform scope is UK schools; no political or electoral content surface. |
| Animal welfare offences | Negligible | Out of scope. |
5. Mitigations in place
- Narrow content scope. Reviews are the only user-to-user surface. No forums, no comments, no direct messaging, no image upload, no user-discovery features.
- Pre-publication human moderation. Every review is read by a human moderator against the Moderation Policy before publication. Reviews naming individuals, containing safeguarding disclosures, or breaching the Moderation Policy are rejected.
- AI-assisted triage. Submitted reviews are screened by an AI classifier for hate speech, threats, self-harm indicators, and defamation risk prior to human review.
- Account controls. Minimum account age 16 (self-declared at signup, email verification required before posting). One account per verified email.
- Identity minimisation. Reviewer names are not displayed. Only relationship (parent / former parent / alumni / staff) and date appear publicly. This reduces targeting risk for all illegal-content categories involving named victims.
- Notice-and-takedown. Any user, school, or third party can report content to moderation@schoolatlas.co.uk. Acknowledgement within 48 hours; resolution within 7 days. Statutory defamation notice-and-takedown under the Defamation (Operators of Websites) Regulations 2013 is published in the Moderation Policy.
- Right-of-reply for schools. Schools can reply publicly beneath any review of their school at no cost (free Claimed tier). This reduces the harm of borderline content while a takedown request is assessed.
- Safeguarding escalation. Content that suggests risk to a child is escalated to the safeguarding contact (safeguarding@schoolatlas.co.uk). Urgent concerns are referred to 999 and NSPCC (0808 800 5000).
- Record-keeping. Moderation decisions, takedown notices, and safeguarding referrals are logged with timestamp and outcome, retained for 24 months.
6. How to report illegal content
Anyone — not only registered users — can report content they believe is illegal. There are two routes:
- Moderation: moderation@schoolatlas.co.uk for review content. We acknowledge within 48 hours and aim to resolve within 7 days.
- Safeguarding: safeguarding@schoolatlas.co.uk for child-safety concerns. For emergencies, call 999 or the NSPCC on 0808 800 5000.
Full notice-and-takedown procedures are published in the Moderation Policy.
7. Review cycle and record keeping
This assessment is reviewed at least annually (next review 21 April 2027) and whenever we introduce a material new feature on the user-to-user surface. Any change that could increase risk triggers an immediate re-assessment before the feature is launched.
In line with the record-keeping duty in section 23 of the Online Safety Act 2023, we retain each version of this assessment, the date it was performed, and a log of moderation/takedown decisions and safeguarding referrals (with timestamp and outcome) for at least 24 months from the date of the action.
8. Responsibility and regulator
Stephen Spence (sole trader, trading as School Atlas) is responsible for this assessment and for the platform's overall compliance with the Online Safety Act 2023.
The independent regulator for the Online Safety Act is Ofcom. The illegal-content duties in Part 3 of the Act apply to all in-scope user-to-user services regardless of size, although some specific codes of practice and duties (for example the duties in Part 3 that apply only to Category 1, 2A or 2B services) are differentiated by size and functionality. Users who are unhappy with how School Atlas handles an illegal-content complaint may raise the matter with Ofcom via ofcom.org.uk/online-safety.
9. Contact
Questions about this assessment: legal@schoolatlas.co.uk.
This document does not create any legal obligation beyond those already imposed by the Online Safety Act 2023 and does not constitute legal advice.