Skip to content

Enterprise

Built for evidence portfolios, not just individual papers.

Trace runs the same measurement across thousands of records, with shared query sets, review states, role-based access, and exports your own reviewers can check. This page describes what exists today and marks what does not.

EfficacySafetyMechanismComparatorPopulationGuidelinesPaper 01Paper 02Paper 03Paper 04Paper 05Paper 06Paper 07Paper 08Paper 09Tint: share of questions in the theme where the paper is retrieved.Dashed: no record in the portfolio is retrieved. The gap is the finding.
Illustrative portfolio coverage from the demo workspace. Rows are papers, columns are question themes, and the dashed cells mark themes where nothing in the portfolio is retrieved.

Three paths

What you are trying to see depends on who you are.

The measurement is the same. The unit of analysis, the comparison set, and the review requirements are not.

Publishers

Journal- and imprint-level views over a whole catalogue: which articles are retrieved, which are structurally under-discovered, and where metadata or availability is the limiting factor.

Portfolio heatmaps, query coverage by subject area, author-facing reports.

Publishers overview

Institutions

Department and centre portfolios with field-normalized comparisons, so a small methods group is not judged against a high-volume clinical unit.

Repository coverage, identifier integrity, exports for research offices.

Institutions overview

Life sciences

Monitor published evidence, stress-test how scientific claims are interpreted, and keep scientific exchange separated from promotional communication by workspace policy.

Compliance modes, review trails, and export watermarking.

Life sciences overview

Portfolios

Measurement that holds together across a catalogue.

A portfolio view is only useful if the numbers inside it were produced the same way. That constraint drives most of the design below.

Portfolio corpora

Group thousands of records into portfolios by journal, department, therapeutic area, or funding programme. Measurement runs on the group, not one paper at a time.

Shared query sets

Query sets are versioned objects owned by the workspace. Two analysts measuring the same portfolio in different months use the same instrument, or the report says which version changed.

Field normalization

Comparisons are normalized by field and publication age. Raw counts across unlike subject areas are shown only alongside the normalized view.

Checkable exports

Every export carries the methodology version, the run window, the provider set, and the sampling parameters, so a figure can be re-derived rather than taken on trust.

Review and access

Controls that match how regulated teams already work.

Roles and separation of duties

Analysts run measurements; reviewers approve outputs; administrators manage providers and retention. Role-based access control is implemented in the application today.

Review states

Reports move through draft, in-review, and approved states. Approved exports record who approved them and against which methodology version.

Compliance modes

A workspace declares its mode — research, publisher, scientific exchange, promotional, or agency. Promotional workspaces route material through review before anything can be exported.

Audit logs

Measurement runs, exports, provider changes, and deletions are recorded with actor and timestamp.

SSO and SAML

Single sign-on and SAML provisioning are on the roadmap and are not available today. Access is managed inside the workspace in the current build.

Directory sync

SCIM user provisioning is on the roadmap. It is listed here so procurement can plan around its absence rather than discover it late.

Roadmap items are marked as roadmap

Single sign-on, SAML, and directory sync are not available in the current build. They are listed so that a procurement review can see the gap rather than infer a capability that is not there.

Deployment

Where the work runs, and what leaves your environment.

Hosted

The default. Tenant-isolated storage, encryption in transit and at rest, and a documented provider registry that says what each data source may be used for.

Regional hosting

Regional data residency is a roadmap item. Ask before assuming a specific region is available for a pilot.

Customer model endpoints

Language-model calls can be pointed at a customer-controlled, OpenAI-compatible endpoint so manuscript text never leaves an environment you already govern.

Data controls

What happens to the content you put in.

No training on customer content by default

Manuscripts and workspace content are not used to train models. Any change to that requires an explicit, revocable opt-in recorded in the workspace.

Rights-aware ingestion

Full text is stored only where reuse rights are confirmed. Where they are not, the analysis runs on metadata and says so on the surface of the report.

Retention and deletion

Retention windows are configurable per workspace. Deleting a manuscript removes the stored document and its derived text, and the deletion is logged.

Provider policy registry

Each upstream data source carries a policy entry covering commercial use, full-text storage, derivative embeddings, and redistribution. A provider with no entry cannot run in production.

No certifications claimedControls documentedSee security for the full control list and its limits.

Next step

Enterprise pilots are open.

Start with a scoped pilot

A defined portfolio, an agreed query set, and a report your team can check against its own knowledge of the literature.

Individual plans are in private beta. Enterprise engagements begin with a pilot on a portfolio you already understand well, because the fastest way to judge a new measurement is to point it at something you can already evaluate by hand.