Skip to content

Bring protocols to life.

Turn your protocol into a study you can inspect, approve, and build.

HumanTrue finds conflicting instructions before they become conflicting builds. Your team resolves what the study should mean. The Study Assistant then structures the approved design in CDISC USDM and helps create EDC and eCOA builds, ICFs, SAPs, translations, training, and other study assets from the same governed knowledge.
Request a Demo
A protocol becomes a structured study model and EDC, eCOA, and document assets on one HumanTrue workbench

  • 1,000+

    protocols from completed Phase 1‑4 studies analyzed by HumanTrue

    See the protocol analysis

  • 20,450

    possible conflicts or ambiguities flagged in those protocols for human review. These were initial flags, not confirmed defects.

  • Best UX

    2026 TransCelerate Protocol Review Challenge award for HumanTrue's review experience

    Read about the award

Independent industry evidence

Sites are showing the industry where protocol ambiguity lands.

A Tufts CSDD and CRIO study of 209 investigative-site professionals found that inconsistent protocols and related study documents create operational ambiguity, rework, protocol deviations, missing data, out-of-window visits, and potential patient-safety and data-quality risk.

HumanTrue addresses the problem before it propagates. The protocol and supporting materials are reviewed together, every finding remains linked to its source, sponsor decisions are recorded, and affected outputs can be regenerated from the same governed study knowledge.

Read the Tufts CSDD and CRIO study

Harper, Ford, Krayem, Nomizu, Andrus, and Getz. Applied Clinical Trials, April 2026.

Two clinical research professionals reviewing study instructions together
Ambiguity reaches people first. Site teams must reconcile it before the study can move.
investigative-site professionals surveyed
209
average wait for the information needed for source preparation, with reports of up to 20 weeks
5.8 weeks
reported more than 21 hours of source-preparation work for high-complexity studies
80%+
“Ensure consistency across all relevant documents.”
Site recommendation reported by Tufts CSDD and CRIO

The order matters

Speed is a multiplier.

Make sure it is multiplying what you meant.

AI will not pause to ask which interpretation was intended. It will act on what was written, at scale, across every connected document and system. Understanding is what makes speed safe.

  • What goes inUnexamined ambiguityTwo reasonable readings remain unresolved.

    What scalesInconsistencyDifferent documents, systems, and teams act on different meanings.

  • What goes inUnderstood intentThe source, interpretation, and decision are visible.

    What scalesCoherent executionEvery connected output starts from the same governed understanding.

Speed multiplies what it is given. Unexamined ambiguity creates inconsistency at scale, while understood intent creates coherent execution at scale.

One governed workflow

From protocol intent to governed execution.

Once the protocol is understood, speed becomes the advantage. HumanTrue carries that governed understanding into every connected document, decision, and system.

  1. 01 · Examine

    Protocol assurance

    Surface inconsistency, ambiguity, missing logic, and the questions that require sponsor decisions.
    Explore assurance

  2. 02 · Know

    Study knowledge base

    Create a deterministic, source-linked foundation from QA'd USDM, sponsor decisions, supporting evidence, and versions.
    See the platform

  3. 03 · Simulate

    Executable study

    Test study logic, evaluate patient paths, and compare operational scenarios without exposing the working system.
    See simulation evidence

  4. 04 · Deploy

    Governed outputs

    Create study documents, plans, translations, analysis metadata, and system inputs from governed study knowledge.
    See outputs
A cross-functional clinical study team resolving a study-design question together

Human judgment stays in control

One source of study knowledge.Different roles.Decisions made with the same evidence.

Your study knowledge base

One study. One governed knowledge base. Every downstream use.

HumanTrue turns the protocol, QA'd USDM, sponsor decisions, supporting documents, and version history into a structured, source-linked knowledge base for the study.

The clean USDM provides deterministic structure. Linked evidence preserves context. Recorded decisions make the sponsor's intended meaning available to every authorized team and system.

  • Know what the study says and how its requirements relate.
  • Reuse the same governed meaning in answers, documents, and specifications.
  • See what changes when the protocol is amended.
See the knowledge architecture
  • Protocol
  • Investigator brochure
  • SAP and plans
  • Lab and operational materials

Governed study knowledge base

QA'd USDM + linked evidence

  • Source
  • Decisions
  • Versions
  • Review status
  • Study Assistant
  • Documents and CRFs
  • Study systems
  • HumanTrue Sim
Study documents and sponsor decisions form a governed study knowledge base that supports assistants, content, systems, and simulation.

HumanTrue Study Assistant

Ask the study. Verify the answer.

The Study Assistant is the role-aware interface to the study knowledge base. It gives clinical, operations, site, data, and statistics teams evidence-linked answers that remain aware of the protocol version under review.

  • Clarify eligibility, visits, procedures, endpoints, and operational requirements.
  • Move from an answer to the exact supporting protocol language.
  • Use role-aware views without exposing the working application publicly.
Learn about Study Assistant
Study Assistant

Study team question

Which participants require central FGFR confirmation before randomization?

Verified answer

Central confirmation is required before randomization for participants enrolled through the tissue-testing pathway.

  • Protocol §5.1
  • Eligibility criterion 4
  • Source verified
Illustration of a Study Assistant question and a verified answer linked to its protocol source. Illustrative, non-interactive view. The Study Assistant remains available only within authorized customer environments.

Executable study simulation

Test the study before you build it.

HumanTrue Sim evaluates the study as a working model, then reveals how design logic, patient characteristics, burden, geography, and operational assumptions interact over time.

  • Separate structural design feasibility from population-conditioned performance.
  • Replay every participant path, transition, guard, and disposition.
  • Compare scenarios with versioned assumptions and reproducible evidence.
View Simulation Evidence
One study. Four evidence views.Model inspection · Participant pathways · Outcomes and cost · Interpretable sensitivity

Protocol integrity

AI will not ask what you meant. It will act on what you wrote.

An inconsistency that once confused one reviewer can now propagate through every generated artifact and configured system. HumanTrue makes it visible before it travels.
inconsistency

Conflicting treatment assignment

The synopsis reassigns additional Group A2 participants to Group A1, while Section 3.1 keeps them in Group A2 for later influenza vaccination.

Synopsis
Reassign to A1 for 13-PCV
Section 3.1
Remain in A2 for influenza vaccine
ambiguity

Which endpoint structure applies?

The summary blends UAS7 with ISS7 and HSS7, while the objectives tables define separate alternative primary endpoint scenarios.

Protocol summary
Blended primary objective
Tables 2-1 and 2-2
Separate endpoint scenarios
See How Assurance Works

Documents, plans, and study materials

Create every study artifact from governed study knowledge.

HumanTrue gives medical writers and functional teams a traceable first draft, with protocol-derived content identified by source and confidence. When the protocol changes, teams can see what must change with it.

  • Participant and study materials

    Drafted from the protocol at the reading level each audience needs.

    • ICFs and assent
    • Screeners
    • Patient brochures
    • Training and study guides

  • Study delivery plans

    • Data Management Plan
    • Clinical Monitoring Plan
    • Safety Monitoring Plan
    • Risk and Quality Plans

  • Analysis and financial planning

    • Statistical Analysis Plans
    • Estimands and ARS metadata
    • Billing and budget plans
    • Coverage analysis

  • Global study readiness

    Certified translation that is 70% faster and less expensive than traditional methods.

    • Country variants
    • Plain-language adaptation
    • Multilingual materials
    • ISO 17100 translation

Built for the teams accountable for delivery

Assurance that creates commercial advantage.

  • For sponsors

    Design with evidence. Start with confidence.

    Give clinical development, study startup, biostatistics, and data management one governed, source-linked representation of the trial.

    HumanTrue for sponsors

  • For CROs

    Win on speed without inheriting hidden risk.

    Differentiate bids, preserve expert capacity, and deliver coherent startup packages across programs.

    HumanTrue for CROs

Trust is part of the product

Every answer. Every dependency. Every version.

  • Source provenance

    Trace conclusions and outputs to protocol evidence.

  • Model isolation

    Customer information is not used to train external models.

  • Human review

    Add expert consensus and certification where it matters.

  • Open interoperability

    Use standards and APIs without surrendering your systems.

Start with one protocol

Find what your current process cannot see.

Request a Working Session