Primary buyer
Medical AffairsThe team that has to have an answer ready when a claim about your drug, or a competitor's, becomes something an institution acts on. This is the buyer the product is built around.
Flagship venture
NextConsensus
Claims becoming harder to ignore.
You already know when a paper publishes. What no feed shows is the slower thing underneath — a claim surviving its ninth revision, shedding the qualifier that contained it, turning up on three pages that used to disagree. By the time a guideline confirms it, the window to prepare has closed.
For teams working against dates that do not move.
The gap the work lives in
A label change is a snapshot of something that had been moving for years. Reconstructing that movement by hand is days of work per claim, so most teams read the snapshot and infer the rest.
A concern that persists through nine revisions is a different object from one that appears and disappears in a week. Same alert, opposite meaning — only the trajectory separates them, and the trajectory is the part nobody has.
What a program delivers
Available now Claim Intelligence
Claim-trajectory reconstruction against a scope agreed in writing, delivered with the revision history behind every ranking. Rebuilt from dated public revisions, each carrying a hash you can recompute against the same sources.
- Trajectory reconstruction First appearance through wording changes, evidence, persistence, and spread.
- Historical provenance When a claim emerged, whether it was contested, and how long it held.
- Source-by-source comparison How the same claim reads across drug, condition, and class pages.
- Cohort analysis Which claims are shared, which are contested, and which are unique to one drug.
- Ongoing monitoring Claims you name, with alerts when their signals change.
What a program needs from you
- One claim, drug, or class where knowing what moved would change your priorities. Starting with one is cheaper, and tells you whether portfolio-wide tracking is worth it.
- The decision it feeds — a CI brief, a safety review, an evidence strategy, a market access plan.
- A willingness to act on the answer, including when the answer is “not yet consequential.”
How it works
- 01 Extract Refract reconstructs claim trajectories from public revision history. Every wording change, citation, certainty shift, and propagation event is dated and verifiable.
- 02 Classify Each event is labelled: real evidence movement, public assimilation, authority-language importation, editorial maintenance, or noise.
- 03 Prioritize Direction and strength of movement rank what deserves attention now — gaining support, shedding qualifiers, spreading across influential sources, appearing in regulator or guideline language.
- 04 Deliver Prioritized claim intelligence arrives with the trajectory evidence attached, early enough to investigate, brief, or escalate.
The classify step carries the weight. Without it, a burst of editorial housekeeping looks identical to a claim genuinely gaining ground — and a system that cannot tell those apart will confidently rank the wrong things.
What is proven, and what is not
Three tiers, each labelled for where it actually stands rather than where it is going.
- Claim Intelligence Available now
Contractable today. Trajectory reconstruction against an agreed scope, with the revision history behind every ranking and a recomputable hash on every source.
- Discovery In validation
Reads accumulating evidence and surfaces the authority decisions worth asking about. Eighteen entries currently published, each checked against its own source by hand rather than generated and left unread. It reports how far evidence has moved, never a probability.
- Authority-transition forecasting Research program
Registers a probability and a deadline against a defined transition, freezes the evidence state, and scores the outcome against public resolution rules. The machinery is built and the resolution rules are public. It is not yet a proven capability, and the worked examples on the public site are labelled illustrative because no registered forecast has resolved yet.
Copy elsewhere on this site used to lead with the forecasting claim. It now leads with the tier that is finished. Forecasting is the more interesting claim and the one I expect to be judged on — which is exactly why it should not be sold before a registered forecast has resolved.
Who it is for
Where label-language importation is the signal that matters.
Where coverage and formulary movement is the decision downstream.
Three functions, three budgets. The product is designed for the first and useful to the other two — naming all three as one buyer was a way of avoiding the choice.
My role
Founder, and sole engineer to date. I built the registration system, the deterministic claim-scoring engine, the trajectory-reconstruction pipelines, and the public resolution rules — on top of Refract, the open-source change-detection engine underneath it.
Continued The rest of what I'm building Ambit, Refract, and the evaluation work around themThe useful first conversation is about one claim, not a portfolio. Name the drug or the class, and the decision it would change.