Quality at Scale
Assess subjective work fairly
and fast at any scale
Human powered Machine Learning
From 30 students to 30,000 candidates
Ministry's and School Groups
National Standards for Complex Learning
Build shared standards for subjective tasks and skills across school groups and education systems.
Explore Education >Awarding organisations
Defensible Standards for High-Stakes Assessment
Turn examiner judgement on complex work into reliable rank orders for boundary setting, moderation, quality assurance and training.
Explore Awarding >High-Volume Recruitment
Stop Screening, Start Ranking
CVs are broken by AI. Use ACJ to rank thousands of candidates based on authentic work samples.
Explore Hiring >Higher Education
Admissions and Peer Learning
A validated way to assess applicant potential at scale, and a tool to empower student peer review.
Explore Higher Ed >Training and Compliance
Validate Internal Assessment
Standardise judgement across sites. Evidence your decisions against audit and external scrutiny.
Explore Training >The RM Compare ecosystem
One platform, three connected modules
RM Compare works as a three-part assessment ecosystem that can be used separately or together. Build trusted standards in Studio, share and govern them in Hub, and use them on demand in Live.
Create and refine standards
Run full comparative judgement sessions, establish high-quality rank orders, and turn expert decisions into robust standards you can reuse over time.
Explore RM Compare | STUDIO >Share, govern and connect
Manage licence centres, control who can access which standards, and support organisation-wide rollout with stronger governance and control.
Explore RM Compare | HUB >Use standards on demand
Put standards to work wherever judgement happens, including mobile-first and on-demand experiences enhanced by the Companion App and RM Compare | NOW.
Explore RM Compare | LIVE >From Efficient Assessment to New Revenue
Turn expert judgement into a reusable asset that saves cost and creates new revenue.
Assess efficiently
Turn ranks into Rulers
Launch new services
Now in BETA
RM Compare | NOW
A lightweight RM Compare experience for quick judgements. Capture or upload an item, compare it to a ready-made standard, and review your score in just a few steps.
- Quick results: Go from item to score in just a few comparisons.
- Built-in standards: Use a pre-built reference scale, so you can focus on judging rather than setup.
Modern Infrastructure
Built for the agile Enterprise
Zero-Friction Scaling
Launch a new assessment cycle in minutes. Our multi-tenant architecture handles demand spikes instantly.
Config over Customisation
Toggle features, adjust workflows, and manage permissions via our admin console. You own your roadmap.
Secure and Compliant
Enterprise-grade security, ISO 27001 certified and GDPR compliant. Millions of judgements processed annually.
The Science of Comparative Judgement
Why Better/Worse beats 7 out of 10
Humans are notoriously bad at absolute judgement. We are highly accurate at Relative Judgement.
1
Capture and Upload
Use the new Companion App to capture physical work instantly, or upload any file type (portfolios, essays, video).
2
Compare Pairs
Judges see two items side-by-side and simply decide which is better.
3
Get Validated Data
Our algorithm generates a highly reliable rank order (0.9+ reliability) and actionable data.
Trusted by world-leading organizations
See the Platform in Action
Assessment in the Age of AI
RM Compare 1 minute explainer
Brilliantly simple. Awesome power.
Turning ranks into rulers
Holistic assessment anytime, anywhere, anyhow
RM Compare is developing fast
Stay up to date with the very latest information
-
Shared Standards: Making Quality Reusable
Shared Standards connects assessment evidence across RM Compare. Studio creates comparative ranks, Hub securely governs mark-derived and ACJ evidence, and Live applies approved rulers to new work. Together, they support consistent standardisation, moderation and on-demand comparison - helping organisations make professional judgement transparent, reusable and reliable.
-
Steady State, Not Static: Maintaining Assessment Standards When Questions Change (3/3)
This post explains how awarding organisations can maintain shared standards even when assessment questions change each year. Rather than relying on a fixed set of historic examples, a living master ruler uses approved anchor items and bridge calibration to test how new task-specific work relates to an established standard. This allows organisations to link, limit or separate new evidence responsibly—maintaining continuity without forcing false equivalence.
-
From Marked Evidence to Reusable Rulers: A Practical Path for Awarding Organisations (Part 2/3)
Awarding organisations already hold valuable standards evidence in marked candidate work, mark schemes, annotations and standardisation materials. This post explains how those assets can become governed reference sets in RM Compare—and, where the assessment construct aligns, be strengthened through ACJ into stable, reusable rulers. The result is a dual-evidence standard that preserves the authority of original marking while supporting more consistent standardisation, moderation and live comparison.