Claim Intelligence is available as a scoped program. The other two tiers still have to earn their place.
Claim Intelligence is the one available to contract now. Claim-trajectory reconstruction against a scope agreed in writing, delivered with the revision history behind every finding. The other two tiers are labelled for where they actually stand.
What you bring
- One claim, drug, or class where knowing what moved would change your priorities. Starting with one keeps the first question bounded and tells you whether a broader review is worth pursuing.
- The decision it feeds: evidence strategy, market access plan, or claim review.
- A willingness to act on the answer, including "not yet consequential."
Available now
Claim IntelligenceHow claims about your drugs, your competitors, or a whole area emerged, changed, and spread, read out of the dated revisions of FDA and EMA label text, guideline entries from bodies such as NCCN, published coverage policies, and the trial literature behind them.
- Trajectory reconstruction. First appearance through wording changes, evidence, persistence, spread. Every step carries the revision date and the source document it came from.
- Historical provenance. When it emerged, whether contested, how long it held.
- Source-by-source comparison. The same claim diffed across the drug page, the condition page and the class review, so you can see where the wording already diverges.
- Cohort analysis. Which claims are shared, contested, or unique to one drug.
- Named-scope reconstruction. Claims you identify, with a dated record of the public changes captured.
Rebuilt from dated public revisions, each one carrying a SHA-256 hash you can recompute in your browser against the same sources. Nothing in a reconstruction is unsourced.
In validation
Trajectory PrioritizationPrioritization, not just reconstruction. The review workflow is being evaluated before it is offered as a separate product.
- Trajectory prioritization. Organizes evidence movement into a review queue.
- Evidence context. Shows the public record and its limits alongside each item.
- Prioritization workflow. Evaluated before it is made contractable.
Not contractable yet. When it ships it is scoped as a paid extension of a Claim Intelligence program.
Research program
ForecastingResearches whether public evidence states support forecasts of defined rung 5–7 transitions: a guideline body changing a recommendation, a regulator changing a label, a payer changing coverage. Tracked via the Verification Ledger.
- Registered authority forecasts. Prospectively registered, immutable, independently adjudicated.
- Evaluation record. Published after a sufficient number of records resolve.
- Timing analysis. Reported with the quality and alert burden needed to interpret it.
Available as a research collaboration. The public record will show what the forecasts establish and what they do not.
Clear boundaries between programs
Each program is judged on its own evidence and evaluation standard. Historical reconstruction is reproducible from public sources; forecasting is reported only when its registered records can support an interpretable result.
Questions involving safety or adverse events require a separate review of the applicable evidence and scope.
For the person who has to vouch for it
The two parts that decide whether the output is trustworthy.
Every event is labelled: evidence movement, public assimilation, authority-language importation, editorial maintenance, or noise. Without it, housekeeping and a claim gaining ground look alike.
Byte-for-byte reproducible from the public record; the hash proves it has not been altered.
Operational leverage
From scattered revisions to a living system of record.
Manual review means tracking disparate documents and hoping not to miss a shifted restriction or emerging subgroup signal. A program compiles that history into a single time-indexed, source-linked asset that can be re-reviewed on demand.
Reviewers move between version histories, supporting documents, and authority pages.
Builds a source-linked trajectory from the public revision history available for the program.
Subtle qualifier changes are easy to lose when versions are reviewed separately.
Shows added, removed, or restored qualifiers with source-anchored timestamps.
Evidence, wording, and authority context often sit in separate documents.
Keeps captured evidence, wording changes, and source spread in the same dated record.
A changed source can force another pass through the same history.
The same captured history can be rerun and compared against the earlier record.
A summary is difficult to check when its source versions and dates are not kept with it.
Each published reconstruction carries source references and a reproducibility marker.
Boundary
What we produce, and what you decide
We reconstruct the trajectories, classify the movement, and hand over the evidence. Whether one is material, and what to do about it, is yours.
How a program starts
Nothing is signed until both scope and figure are in writing.
- First conversation
Send the drug, class, or area and the decision it feeds. Start with public sources and high-level context.
- Fit decision, including no
Whether the public record there is thick enough to reconstruct from. If it is not, you get that answer in writing instead of a proposal.
- Fixed-scope statement of work
Which drugs, how far back, how many claims monitored, what you receive and when, with the figure attached.
- Contracting and delivery
Any confidentiality, data-processing, retention, and delivery terms are settled in writing before work starts.
Validation status by forecast class
Forecasting is offered as structured judgment; no class has a resolved calibration record yet.