Protocol Integrity Audits for Funds & Boards
Surface milestone fragility—before capital is committed.
We don’t judge outcomes. We formalize decision logic and stress-test assumptions—then deliver board-ready, forwardable artifacts.
Board-ready, forwardable outputs—designed for IC and governance workflows.
Not a CRO. Not a biostats service. Decision infrastructure for governance.
De-identified example • no sensitive data required • designed to be forwarded
The failure mode funds feel too late isn’t always biology. It’s design fragility.
Most funds diligence “mechanism risk” well. But many late-stage losses come from a different category:
protocol choices whose assumptions fail under real-world variance (sites, missingness, adherence, endpoint behavior, operational drift).
When that happens, you rarely get a clean failure—you get an ambiguous readout. And ambiguity is where capital gets damaged.
If you don’t audit this (VP-cost bullets)
- You fund a milestone that can’t produce a decisive signal under real-world drift.
- You inherit “decision debt”: choices that were agreed internally but are indefensible in writing.
- You pay for a protocol that becomes amend-heavy mid-stream (and interpretability suffers).
- You exit diligence with false confidence—until the first operational variance hits.
What you buy: an auditable decision trail
Protocol Risk Register (Logic Risk Register / MRR)
Ranked assumptions + failure modes + “what must go right,” each tied to milestone exposure.
Milestone Fragility Map
Which milestone is assumption-sensitive; what would make the readout ambiguous vs decisive.
Board-ready Decision Note
Forwardable memo: key risks, minimum fix set, questions to ask, and thresholds that change conviction.
Artifacts are inspectable (LaTeX/PDF), versioned, and reviewable—not slide decks or black-box outputs.
What decision are you at right now?
Pick the governance decision gate you’re at today. Same decision library as research delivery—packaged for risk triage.
Governance modules (same library, different packaging)
- Design = before protocol lock / capital sign-off
- Review = before partner/regulator scrutiny
- Mid-Study = during execution under drift/amendment pressure
- Post-Readout = after results, next-step decisions
- Asset = continuity, auditability, reuse
Phase A — Design (Before you run it)
VP question:
What breaks first—and what does that do to the milestone?
Why you need it:
“Sound” protocols often fail on a small number of silent assumptions.
If you don’t do it:
you fund exposure that becomes ambiguous once real-world variance hits.
We deliver:
Risk Register + fragility notes + minimum fix set
We surface:
recruitment feasibility collapse • endpoint mismatch • detectability erosion under drift
Best used:
2–6 weeks before protocol lock / capital sign-off
VP question:
Is the study powered for the world you’ll actually face?
If you don’t do it:
you pay for a “powered” trial that can’t produce a decisive signal under variance.
We deliver:
power-drift scenarios + N/budget trade-offs + “what changes conviction.”
We surface:
dilution • variance inflation • missingness/non-adherence
Best used:
before approving N, budget, timeline
VP question:
Are we funding measurement burden that doesn’t move the decision?
If you don’t do it:
you spend on panels/assays that increase cost and missingness without improving inference.
We deliver:
minimal must-measure set + timing windows + burden/cost flags.
We surface:
low-information markers • timing misses • over-collection
VP question:
Is the regimen robust—or only optimal on paper?
If you don’t do it:
regimen fragility (adherence/exposure drift) can erase signal or inflate toxicity.
We deliver:
regimen scenario matrix + robustness options + defensible dose rationale memo.
We surface:
exposure variability • compliance realism • efficacy–toxicity trade-offs
VP question:
What transfers—and what breaks—across settings, phases, or real-world rollout?
If you don’t do it:
you scale a result that doesn’t survive the next context and learn late.
We deliver:
transferability map + adjustment scenarios + “what must change before scaling.”
We surface:
population shift • exposure shift • workflow/endpoint shift
VP question:
Will the pathway hold at scale—or break at handoffs/capacity constraints?
If you don’t do it:
execution bottlenecks create outcome variance independent of biology.
We deliver:
bottleneck map + drop-off points + intervention what-ifs
We surface:
bottlenecks • drop-offs • capacity constraints
Not sure which block fits your situation?
Stage-0 helps you identify which decision risks actually matter before you invest further.You get a short written verdict — modelable as-is, modelable with changes, or not a fit right now — plus the lowest-effort next step.
Phase B — Review (Before external scrutiny)
VP question:
Can the team defend the design logic under partner/regulator scrutiny?
If you don’t do it:
you hit rewrite cycles, delays, or partner hesitation due to indefensible rationale.
We deliver:
defensibility pack + assumptions log + claim↔endpoint alignment note.
We surface:
unstated assumptions • logic discontinuities • claim–endpoint mismatch
Phase C — Mid-Study (During drift/amendment pressure)
VP question:
What adjustment reduces risk without breaking interpretability?
If you don’t do it:
you spend more and still end with an un-interpretable readout—worst-case capital outcome.
We deliver:
constrained scenarios + lowest-risk amendment options + board-ready rationale note.
Guardrail:
Not a replacement for clinical/regulatory oversight; scenario transparency only
Phase D — Post-Readout (After results)
VP question:
Null because biology failed—or because design drifted?
If you don’t do it:
you kill a viable program—or fund the wrong next study.
We deliver:
divergence map + next-step decision options + what to test next
VP question:
Would this hold if rerun—or is it fragile to plausible ranges?
If you don’t do it:
you build strategy on a result that won’t survive scrutiny.
We deliver:
consistency metrics + sensitivity notes + reproduction-style report.
VP question:
What unmodeled factor most likely shaped the outcome?
If you don’t do it:
the next study repeats the same blind spot at higher cost.
We deliver:
candidate confounders list + testable next-step plan (not speculation).
Phase E — Asset (Reuse / Continuity)
Protocol integrity as an auditable asset
We translate protocol logic into explicit, inspectable structures—so risk becomes measurable, discussable, and fixable.
Ranked Protocol Risk Register (Decision Impact–Linked)
- Prioritized fragility points
- Severity rationale
- Minimum Fix Set (highest-leverage changes)
Built from our protocol fragility & first-break use cases
Decision Thresholds & Gate Criteria
- Go/No-Go logic grounded in explicit assumptions
- Sensitivity to real-world variance
- Clear “what must be true” conditions
Derived from real-world power, robustness, and sensitivity use cases.
Transferable Audit Artifacts
- I/O logic contracts, scenario definitions
- Code & tests (when applicable)
- Built to be shared in governance, diligence, and handoffs
Matches our reviewer-ready logic pack + reusable decision asset use cases.
Independent • Inspectable • Method-focused
Protocol Blind-Spot Scan + Real-World Power Check (lite)
Best for:
early diligence / kill-switch questions
Outputs:
1-page Risk Register + governance question set
Scope:
top assumptions + first-break analysis
Best when you need a fast IC-screen view.
+ Measurement Decision Design + Regimen Robustness + Transferability (as the asset requires)
Best for:
pre-term sheet / pre-go-no-go milestone decisions
Outputs:
Risk Register + Fragility Map + board-ready Decision Note
Includes:
2–3 scenario stress tests
For upcoming capital or milestone decisions.
Reviewer-Ready Logic Pack + Reusable Decision Model Asset (+ code/tests)
Best for:
high-exposure milestones / pivotal trials / platform bets
Outputs:
full Assumptions Map + architecture + scenario pack + audit trail
Includes:
expanded scenario coverage + “what changes conviction” thresholds
For high-stakes or late-stage programs.
Scope discipline line: Stage-0 and audits test decision logic—not outcomes.
Inspect our math
Public, versioned artifacts—reviewable by peers.
De-identified examples on Zenodo and accompanying GitHub repos. Outputs are delivered in research-grade formats (LaTeX/PDF) suitable for citation and review—not marketing decks. Zenodo shows the artifact format. Stage-0 opens a case-specific audit trail for your program.
Before capital is committed, make the protocol defensible.
Governance triage only — not fundraising
If you’re a fund or board evaluating a milestone decision, the Free Feasibility Check is the fastest way to route your case to the right audit tier.
You’ll receive a short confirmation and (when appropriate) a sample risk-register view — and you can also inspect our public Stage-0 → Stage-3 example deliverables (Zenodo) to see the artifact format end-to-end.
Run Stage-0 before your next IC discussion.
Stage-0 is the first gate in our governance audit process. We review protocol logic and assumptions only—no outcome optimization.