Skip to content

Pricing

Priced by the portfolio, not by the seat.

Because the value is in how much of a corpus is being watched — and because everyone who needs to see a result should be able to.

Enterprise plans

Department

A single lab group, department, journal, or therapeutic area.

$24,000

per year, indicative

Scope
Up to 500 papers under continuous measurement
Cadence
Monthly re-measurement
Panels
4 stakeholder panels per year
Users
Up to 10 named users
  • Versioned query sets, reviewed with your team before the first run
  • Retrieval, citation, evidence use, and interpretation fidelity reported separately
  • Competitor and substitute analysis across your query set
  • Pre-publication clarity and discoverability recommendations with fidelity checks
  • Launch checklists with an owner per row
Request a scoping call

Institution

Most scoped

A university research office, a publisher imprint, or a medical affairs team covering several products.

$78,000

per year, indicative

Scope
Up to 5,000 papers under continuous measurement
Cadence
Fortnightly re-measurement
Panels
16 stakeholder panels per year
Users
Up to 40 named users
  • Everything in Department
  • Portfolio views by department, imprint, subject, or therapeutic area
  • Field-normalised benchmarking against a comparison set you declare
  • Role-based access with review states and an audit log
  • Compliance modes separating scientific exchange from promotional workflows
Request a scoping call

Global

A full publisher catalogue, a multi-institution consortium, or a company-wide evidence portfolio.

Custom

scoped to catalogue size

Scope
Unlimited papers, catalogue-scale corpora
Cadence
Weekly re-measurement, or on your schedule
Panels
Unscoped stakeholder panels
Users
Unlimited named users
  • Everything in Institution
  • Dedicated measurement capacity and a private query-set namespace
  • Custom instruments and personas developed with your scientific team
  • Integration with your repository, CRIS, or publishing platform
  • Joint methodology validation work, published if you want it published
Talk to us

Indicative annual list for a standard agreement. Scope varies with catalogue size, the number of surfaces measured, repetition counts, and how much of the query-set work your team wants to do with us. Nothing is quoted before a scoping call.

Start here

An eight-week pilot on evidence you already understand.

Not a trial account. A scope of work with a fixed fee, a defined portfolio, and a report your own reviewers can check against what they already know about the literature.

$12,000fixed fee, credited in full against a first-year agreement
Request a scoping call
  1. Week 1

    Scope

    You pick a portfolio your team already understands — a journal, a department, a therapeutic area. The fastest way to judge a new measurement is to point it at literature you can evaluate by hand.

  2. Weeks 2–3

    Agree the instrument

    We build the query set from the extracted study structure and review every question with your scientific team. Nothing is measured against an instrument you have not seen.

  3. Weeks 4–6

    Measure

    Baseline runs across the agreed surfaces, with repetitions and a recorded seed. Competitor analysis and, if useful, a stakeholder panel on one publication.

  4. Weeks 7–8

    Read it critically

    A working session on the results with the people who know this literature, plus a written report and the underlying observations. Where you disagree with a number, we show you the transcripts behind it.

What you are buying

The same loop, three different reasons to run it.

Life sciences and medical affairs

Published evidence is being summarised for clinicians and payers by systems your medical review process never sees, and the first sign of a misreading is usually a question from the field.

  • Find out how a pivotal publication is actually being represented, before a field team hears it back
  • Stress-test interpretation ahead of congress and launch, and record what was tested
  • Keep scientific exchange and promotional workflows separated, with a review trail on both

Universities and research offices

Research output is spread across repositories with uneven metadata, and reporting on its reach still depends on citation counts that lag by years.

  • See which research is visible, under-discovered, or misinterpreted, by department
  • Turn repository and identifier gaps into a defect list somebody can actually work through
  • Report on discoverability with a stated methodology instead of a vendor's black box

Publishers and journals

Your catalogue is the asset, and its discoverability now depends on machine readability and availability characteristics that vary silently across imprints.

  • Measure how the catalogue surfaces by journal, imprint, and subject
  • Find the availability and metadata failures holding back whole segments at once
  • Show authors what happens to their work after publication, with evidence

In every agreement

What the price actually covers.

Measurement

  • A versioned query set per paper or portfolio, generated from the extracted study structure and reviewable by hand
  • Repeated runs across every configured surface, with the sampling seed recorded so any result can be reproduced
  • Retrieval, citation, evidence use, and interpretation fidelity determined independently

Work you can act on

  • Defect lists for metadata, availability, machine readability, and terminology, each with the party who can fix it
  • Pre-publication clarity suggestions, every one checked against the extracted claims and discarded if it would change what the paper asserts
  • Synthetic stakeholder panels that surface likely misreadings before they reach real readers

Governance and control

  • Customer manuscript content is never used to train models, on any tier — not purchasable, not tier-gated, not waivable by a terms update
  • Configurable retention, including deletion on request
  • Customer-controlled model endpoints, so unpublished manuscripts need not leave a provider you have not approved

How we work with you

  • A scoping call before any commercial conversation, to establish whether this measurement is useful to you at all
  • Query sets agreed with your scientific team rather than generated and handed over
  • A named contact from Institution upwards, and a quarterly review of the methodology against your results

Stated plainly

What you are not buying.

Procurement will ask, so it is here rather than in an appendix. A vendor that cannot tell you where its product stops is not a vendor you can check.

  • No SOC 2 report, ISO 27001 certification, or third-party security attestation exists today.

  • Single sign-on and SAML federation are planned and not available in the current build.

  • No guarantee, projection, or target for improved ranking in any answer engine. Trace observes surfaces; it does not model how they rank.

  • No claim that a recommendation caused a change in visibility. Changes are reported as having occurred after an event, never as caused by it.

Questions

Before you ask.

Why price by portfolio rather than by seat?
Because seats are the wrong meter. The value is in how much of a corpus is under measurement, and charging per seat pushes teams to share logins — which breaks the attribution the launch and approval workflows depend on. Everyone who needs to see a result should be able to.
Are these prices firm?
They are indicative annual list for a standard agreement. Real scope varies with catalogue size, how many surfaces you want measured, repetition counts, and how much of the query-set work your team wants to do with us. We quote after a scoping call, not before one.
What happens if we disagree with the numbers?
That is the useful part of the conversation, and the product is built to support it. Every figure decomposes to the observations behind it, and every observation keeps the response text it was scored from. If a result does not survive contact with people who know the literature, we would rather find that out during a pilot.

Point it at a portfolio you already understand.

Request a scoping callWalk the product first

Trace will tell you what it cannot measure before it tells you what it can. Email hello@trace.science.