← Back to Trial Intelligence Report pitch
Trial Intelligence Report — Procurement FAQ
Six Procurement Objections, Pre-Handled
A buyer paying $25K–$100K for a Trial Intelligence Report pilot sees the same six questions from procurement, legal, and the program team before SOW signature. This FAQ pre-handles them, in order. Jump straight to the objection that needs the most defensible answer; the run-through below is short on purpose.
Request-to-Delivery Window Per Tier
The clock is unambiguous: the kickoff date on the signed SOW is day zero. Requests before SOW signature, scoping reads, and the procurement cycle are not counted against the turnaround window. Every paid pilot ships inside its published per-tier window:
| Tier | Price | Scope | Turnaround (from signed SOW) |
|---|---|---|---|
| Tier 1 | $25K | Single-trial KM decomposition | 4 weeks · calendar weeks, business-day review |
| Tier 2 | $60K | Multi-trial comparative brief (2–3 trials) | 6 weeks · calendar weeks, comparator cross-read |
| Tier 3 | $100K | Engagement-scope pilot, in-person working session | 6–8 weeks · includes the in-person session |
Scoping-read SLA. After you submit the lead form on the pitch page, the request routes into lead_captures and we respond with a scoping read within one business day — which tier fits, which publicly disclosed dataset is intact enough to run on, and what the 4–6-week window looks like from kickoff. Billing starts on the SOW signature date, not on first contact.
License Terms — No Client Data Ingested
HyperCore runs only on publicly disclosed datasets — SEC filings, ClinicalTrials.gov entries, ASCO/ESMO/AACR abstracts, and published KM curves from primary endpoints. No client data is ingested into the pipeline at any tier. Because no client confidential information enters the modeling, no pre-existing IP assignment is created against the client for the methodology itself.
What gets licensed vs. what we retain
Licensed to client (engagement deliverable). The Trial Intelligence Report artifact set — HTML, CSV, and Markdown — produced specifically for the client's trial or trial-set, used internally by the client and its program team. License is internal use; redistribution outside the client's organization requires a separate written agreement.
Retained by DiviScan. Rights to the HyperCore methodology — the 320-classifier ensemble, the decomposition stack, the inflection-labeling approach — and to aggregated, de-identified learnings drawn from across all engagements. The de-identified learnings carry no client-identifying information and inform methodology refinement across pilots.
NDA posture. No NDA is required to scope or run a Tier 1 or Tier 2 pilot — the inputs are public and the deliverable is licensed for internal client use. A mutual NDA is only required for a Tier 3 pilot's in-person working session, where preview materials (full draft narrative, comparative classifier weights, and forward-looking framing variants) are exposed to the program team in advance of delivery.
Data Sources & HyperCore 320-Classifier Stack
Every Trial Intelligence Report decomposes a publicly disclosed phase 2/3 overall-survival (OS) dataset through the same HyperCore ensemble. The audit trail sits on three legs: sources are public, scoring is deterministic, and the CSV companion ships on day one so reviewers can re-derive any chart or read any classifier weight against their own numbers.
HyperCore — the 320-Classifier Ensemble
The 320-classifier ensemble scores each patient against the inflection-labeled boundary the KM decomposition exposes. Classifiers are deterministic at scoring time — re-running the same dataset against the same classifier set yields identical scores, so an internal re-analysis is reproducible without method-transfer drift. Top-12 feature weights are normalized to sum ≈ 1.00 and listed auditable in the CSV companion that ships on day one of the deliverable window.
Auditable companion. The CSV ships on the same day as the HTML and Markdown. Reviewers can re-derive the chart, check the top-12 weights against their own read, or load the row set into R/Python/Excel for an independent re-analysis — without leaving the engagement scope.
Fixed-Bid Pilot vs Time-and-Materials Paid Expansion
A Trial Intelligence Report pilot is a fixed-bid deliverable scoped inside the published $25K–$100K band. Once the SOW is signed, scope, tier, and price don't move: the deliverable shape is the same at every engagement.
How a paid expansion is structured. A pilot is fixed-bid, deliverable-bound, and milestone-billed (see pricing & engagement terms). When the pilot lands and the next question surfaces — validate-on-your-data, extend the comparator set, repeat for a fresh trial — the follow-on work is run on a time-and-materials (T&M) basis with a scoped SOW per phase. The pilot is a decision-opener; the expansion is a continuation.
HTML, CSV, and Markdown — Same-Day Artifact Set
Every Trial Intelligence Report ships in three formats so the reviewer can read, audit, and re-analyze in whatever tool fits the workflow. All three are produced from the same run, so weights and rows match across files.
| Artifact | Format | What it contains | Width by tier |
|---|---|---|---|
| HTML | Self-contained printable | Numbered pages, inline SVG KM chart, classifier table, decision narrative. First-pass Tier 1 review. | 4 / 8 / 12 pages |
| CSV | Auditable raw data | KM curves (24 monthly rows per arm) + classifier-attribution rows. Importable into R, Python, or Excel for re-analysis. | 48 KM rows + classifier weights |
| Markdown | Version-controlled narrative | Page-by-page decision narrative, editorial hooks, diff-able across revisions. A natural input for internal review threads. | Mirrors HTML page count |
Ergonomics, by reviewer. A clinical reviewer reads the HTML first pass (8–12 minutes); an internal biostats reviewer opens the CSV alongside (30–45 minutes); a regulatory or legal review reads the Markdown for the decision narrative and can diff revisions line-by-line, which matters when the engagement includes the Tier 3 revision round.
$25K–$100K Band, Fixed-Bid Pilot, 50/50 Milestone Billing
Published pricing band is $25K–$100K per the published roadmap; tier is chosen during the scoping call. The pilot pricing is fixed-bid; the follow-on T&M structure is scoped per phase after the pilot delivers.
Follow-on engagements (post-pilot). Run on a time-and-materials basis, scoped per phase. Common follow-ons are refresh-on-new-data cycles (T&M, monthly cadence), comparator-set extensions (small T&M extensions), and validate-on-your-data pilots (fixed-bid, quoted per scope). The pilot is always the entry point; the expansion is always separately scoped.
Request a Pilot Engagement
When procurement, legal, and the program team are aligned, submit the lead form on the pitch page. The request lands in lead_captures and we respond within one business day with a scoping read — which tier fits, which publicly disclosed dataset is intact enough to run on, and what the 4–6-week window looks like from kickoff.
Request pilot engagement →What arrives in your inbox. A tier recommendation, a named dataset trial-set, a confirmation of the published turnaround window, and the SOW template scoped to your pilot. Read the pitch page end-to-end before requesting — the FAQ here is the procurement-side pre-handler; the pitch page is the engagement-side opener.