Feature benchmark
Read the admin-only LMS feature benchmark — how CLIP Learn is positioned against the LMS market, how to read the matrix, and the compact/detail toggle.
The Feature benchmark is an internal positioning tool that lives in the admin console. It lays out, capability by capability, how CLIP Learn compares to the mainstream LMS market — Moodle, Canvas, Docebo, TalentLMS, 360Learning, LearnWorlds and Sana Learn — so staff can answer "how do we stack up?" and "what are we still missing?" from one screen. It is a sales-and-strategy reference, not a learner-facing feature: nothing here changes what candidates or courses do.
Admin only
The page is gated to the Admin role (requireRole("ADMIN")). Faculty, Candidate and Corporate accounts cannot open it. The Corporate sponsor view is a separate surface and does not include the market matrix. There is no "share" or "export" control — it is a read-only internal reference.
Where to find it
Open /admin/benchmark in the app (the browser tab reads "Benchmark"). The page title on-screen is LMS feature benchmark, with the sub-line "How CLIP Learn compares to the LMS market — each row explains the capability and shows every platform's depth, with a link into the live feature."
What the page is for
CLIP Learn is a vertically-built legal exam-prep trainer for the CGI / CGQP programme, not a general-purpose LMS. This page exists to make an honest internal case: where CLIP genuinely leads the field, where it is at parity, and where it still trails the incumbents. The competitor notes come from 2025/26 product research; CLIP's own column reflects the shipped codebase. The dataset is kept in sync with the internal docs/LMS-BENCHMARK.md reference.
How the page is laid out
Reading top to bottom, the page has six sections:
The four summary tiles
The tiles are computed live from the matrix data, so they stay accurate as the underlying dataset changes.
| Tile | What it counts | Sub-line |
|---|---|---|
| Capability areas | Total number of capability rows in the matrix | "benchmarked vs 7 platforms" |
| CLIP leads the field | Rows where CLIP's level is strictly higher than every competitor's best on that row | "ahead of every comparator" |
| Shipped native | Rows where CLIP is rated full | "+ N partial" (the count of partial rows) |
| Closed this build | Rows that carry a "shipped" tag, i.e. gaps recently turned green | "gaps turned green" |
How 'leads the field' is decided
Levels are ranked full > partial > none > na. A row counts toward "CLIP leads the field" only when CLIP's rank is strictly greater than the highest-ranked competitor on that same row — parity does not count as a lead.
How to read the matrix
The matrix is one row per capability and one column per platform. CLIP Learn is the first column after the capability name (highlighted with a soft brand background and a coloured left/right border); the seven competitors follow in a fixed order: Moodle, Canvas, Docebo, TalentLMS, 360Learning, LearnWorlds, Sana Learn.
The Capability column and the header row are sticky — the capability names stay pinned on the left and the platform names stay pinned at the top as you scroll. The grid is wide (minimum 1040px), so on narrow screens it scrolls horizontally inside its own container.
Every cell shows a status dot + word (Full / Partial / None / N/A) and, in detail mode, a short note explaining that platform's depth on that capability.
The status levels
There are four levels, shown as a coloured dot plus a short label. The legend at the foot of the matrix spells out the full meaning:
| Dot | Short | Full label | Meaning |
|---|---|---|---|
| Solid green | Full | Full — shipped, competitive | The capability is shipped and competitive |
| Half-filled amber | Partial | Partial — exists but limited | It exists but is limited in scope |
| Hollow grey | None | None / metadata only | Not supported, or metadata only |
| Dashed outline | N/A | Not applicable | Not applicable to that product |
CLIP's cell extras
CLIP's column is the only one that can carry two extra elements below the status:
| Element | What it is | When it shows |
|---|---|---|
| A brand-coloured "shipped" chip with a sparkle icon | Names the specific thing that was closed or advanced in the recent build (e.g. "SCORM 1.2 import + runtime", "Gradebook + CSV export") | Only on rows where CLIP's entry has a shipped tag |
| An Open in app link with an up-right arrow | Deep-links straight into that live feature so staff can click through and see it working | Only on rows that have an href set |
The Open in app deep-links point at the real features, for example: Course Authoring and Video open the faculty catalog (/faculty/catalog), Content Interop opens /faculty/scorm, Public API opens /faculty/api, Assessment opens /faculty/questions, AI Tutoring opens the trainer (/app/trainer), Roles/RBAC opens /admin/users, Cohorts opens /admin/cohorts, Analytics/Gradebook opens /faculty/gradebook, and Communication opens /admin/conversations. Some rows — Proctoring, Mobile, Gamification, Certificates, White-Label, Accessibility — have no deep-link because there is no single dedicated screen for them.
The compact / detail toggle
Above the matrix on the right is a single toggle button that switches the whole grid between two densities. It defaults to detail view.
| Button reads | Current mode | What you see |
|---|---|---|
| "Compact (dots only)" | You are in detail mode | Every cell shows its status word plus its explanatory note; each capability row also shows a "why it matters" line under the capability name |
| "Show detail" | You are in compact mode | Cells collapse to just the status dot and word; the per-row notes and the "why it matters" lines are hidden |
Reading tip
Use compact to scan the shape of the field at a glance — where the green clusters and where CLIP trails. Switch to detail when you need the actual reasoning behind a rating, or the "Open in app" links and shipped chips in CLIP's column (those still appear in compact mode; only the prose notes and why-it-matters lines are hidden).
Where CLIP is ahead
The left panel of the edge/gaps split lists CLIP's genuine differentiators — "capabilities most of the market only markets, and we ship." The recurring themes are:
- A grounded, cite-or-refuse-and-escalate AI legal trainer (MOEgent) with a real anti-hallucination contract — it refuses out-of-materials questions and files a human escalation rather than inventing law.
- One continuous conversation across web chat, hands-free voice and WhatsApp, sharing identical grounding and a single memory thread, now fully readable by staff from the admin console.
- An agentic staff copilot that performs actions (creates clients, drafts modules behind an SME publish gate, ingests URLs/PDFs, schedules sessions) rather than a chat widget.
- Hardened multi-tenant isolation — session-derived, fail-closed tenant resolution with IDOR-guarded writes, verified across two tenants.
- Vertical fit and velocity for CGI/CGQP exam-prep in one platform, with no per-seat SaaS tax.
Roadmap and open gaps
The right panel lists what remains after the recent "turn them all green" build. Each gap carries a priority tag and, in the expanded Gap analysis and recommendation section further down, the market leaders on that capability plus the recommended next step.
| Priority tag | Colour | Meaning |
|---|---|---|
high | Red | Most urgent gap to close |
med | Amber | Mid-priority |
low | Grey | Low priority / deliberate skip |
The current gaps span interoperability (xAPI / cmi5 / SCORM 2004 / LTI — SCORM 1.2 already ships), adaptive-bitrate video with captions and CDN, varied assessment item types and calibrated IRT, embeddings-based hybrid RAG retrieval, a store-native mobile app, a skills/competency ontology and HRIS sync, verifiable credentials plus formal WCAG 2.2 / VPAT / SOC 2 conformance, and productised email/billing (email is code-complete and activates the moment SMTP credentials are set).
A deliberate 'None' is not a failing grade
Two rows are rated none on purpose. Proctoring and secure exams is a considered skip — CLIP's mocks are practice, and the real CGQP exams are proctored by CGI on Rogo. Accessibility and compliance is rated none because strong security hygiene is not the same as stated WCAG 2.2 / VPAT / SOC 2 conformance. The notes in detail mode explain the reasoning in each case, so don't read a hollow dot as an oversight.
Keeping it accurate
The whole page is data-driven — the verdict, tiles, edge/gap lists and every matrix cell come from a single internal dataset, and the tiles recompute themselves from that data. When a capability genuinely ships or advances, updating the dataset (in sync with the internal LMS-benchmark document) is what moves a dot to green, adds a "shipped" chip, and re-counts the summary tiles. Nothing on this page is hand-edited per view.