CEM Benchmark

Release 2026.3 · Reviewed July 2, 2026

Methodology & features

How we score clinical education platforms: binary rubric, cited primary sources, semantic proof matching, and a public capability specification.

Capabilities

28

Binary scored

Modules

7

Functional areas

Vendors

9

Release 2026.3

Scale

✓—

Cited or not

Scoring rule

Every capability is either cited or not — no partial credit, no inferred features.

✓

Supported

Score = 1

Public vendor documentation with a direct URL and quotable excerpt that substantiates the capability.

—

Not supported

Score = 0

No verifiable primary-source proof, or marketing copy too generic to document the specific feature.

Aggregation

module = supported ÷ total in module

overall = supported ÷ 28 capabilities

Unweighted. No popularity or market-share inputs.

Evidence & evaluation

What counts as proof, and how each capability is verified.

  1. 01

    Fetch primary source

    Product pages, help centers, or API docs — never third-party reviews alone.

  2. 02

    Semantic match

    Does the excerpt substantively satisfy the capability definition, even with different wording?

  3. 03

    Record proof

    Store URL, verbatim quote, interpretive note, and match type (semantic or literal).

Acceptable

  • ✓Vendor product and feature pages
  • ✓Public help center or knowledge base articles
  • ✓Published API or integration documentation

Not acceptable

  • ×Third-party review claims without primary source
  • ×Assumed capabilities based on market position
  • ×LLM-generated summaries without URL verification

Proof matching

Semantic equivalence with verbatim excerpts — not keyword bingo.

Each supported capability requires a public URL and a verbatim excerpt. Proof is evaluated semantically— not by keyword matching alone.

  • ·Quoted text must substantively address the definition, even when terminology differs (e.g. "affiliate network" vs "clinical site database").
  • ·A short note explains the rubric link when wording alone is ambiguous.
  • ·Generic marketing without a specific feature does not count.

Example citation

placement.site-directory

Click to expand — same pattern used in comparison tables.

Safeguards

Design choices that limit bias and over-scoring.

01

Cited evidence required

A capability scores as supported only when a public source URL is attached with a verbatim excerpt. No quotable proof, no score.

02

Semantic proof matching

Evidence is matched on functional equivalence to the capability definition, not exact keywords. The quoted excerpt and a brief note must substantiate the claim.

03

Native platform capability required

A capability scores as supported only when the vendor documents the feature as native to their platform or as a vendor-operated service. Third-party partner products, integrations, or add-ons that require a separate vendor subscription or contract do not count.

04

No platform-level automation bucket

Automated assistance is evaluated within each functional module (not as a separate platform category), preventing disproportionate overall credit for automation-heavy vendors.

05

No popularity weighting

Scores are derived exclusively from the capability rubric. Market share, search frequency, and brand recognition are not inputs.

06

Unknown capabilities score as not supported

Capabilities that cannot be verified from public documentation receive a score of not supported. The benchmark does not infer undocumented features.

07

Versioned benchmark releases

Each release is immutable and dated. Score changes require documented evidence updates and a new release version.

08

Equal evaluation methodology

All vendors are evaluated against the same feature definitions using the same evidence criteria, regardless of vendor size or tenure.

Vendor scope

Release 2026.3 — 9 clinical education management platforms evaluated against the same rubric.

HealthTasksExxatTrajecsysTyphonProject ConcertTracPracCORE Higher EdeMedleyMedatrax

Feature definitions

Canonical capability specification used for evaluation. Internal IDs are omitted from this view.

Multi-program management

Manage multiple programs or tracks within one tenant.

Evidence bar: Public documentation of multi-program or multi-track administration.

Integrations & APIs

Documented LMS, SIS, or API integrations.

Evidence bar: Public integration or API documentation.

Enterprise authentication

SSO, security policies, or enterprise auth controls.

Evidence bar: Public SSO, security, or authentication documentation.

Publisher. The Clinical Education Management (CEM) Benchmark is published and maintained by HealthTasks. All vendors, including HealthTasks, are evaluated using the same published methodology and evidentiary standards based on publicly available primary sources. The benchmark is provided for informational and research purposes only and does not constitute an endorsement, certification, warranty, or guarantee regarding any vendor, product, or service.

Methodology. Scores reflect documented capability coverage based on publicly available primary vendor sources, including product documentation, help centers, knowledge bases, API documentation, and other primary-source materials available at the time of review. Scores are intended to measure documented capability coverage under the published methodology and should not be interpreted as measures of product quality, customer satisfaction, usability, market share, or overall suitability. A capability marked as not supported indicates that HealthTasks did not identify sufficient publicly available evidence to verify support under the published methodology. It does not necessarily mean the capability is absent or unavailable.

Process. Research is produced using AI-assisted analysis together with human review and is refreshed periodically. While reasonable efforts are made to ensure accuracy and consistency, errors or omissions may occur, and vendor documentation may change over time. Users should independently verify information before making purchasing, procurement, or business decisions. Machine-readable benchmark data and methodology are available at /benchmark/dataset.

Updates and Corrections. Vendors and members of the public may submit corrections or additional primary-source documentation for review. Submissions are evaluated using the same published methodology and evidentiary standards applied to all vendors. Accepted corrections may be incorporated into the current release, published in the benchmark changelog, or included in a subsequent benchmark release.

Trademarks. All company names, product names, logos, and trademarks are the property of their respective owners and are used solely for identification, comparative research, and informational purposes. Their use does not imply any affiliation with, sponsorship by, or endorsement from their respective owners.