The platform that turns the entire public drug development record into quantified answers. Built from inside the decision, not from outside the data.
QD [Quantify Decisions] is developed by someone who assessed regulatory dossiers as an FDA reviewer, then spent more than a decade making the development decisions in pharma.
WHAT IT ANSWERS
QD [Quantify Decisions] is organized around key questions, not around data types. Each of these questions return a deliverable you can take into a project team, governance, or regulatory meeting.
What does my target product profile need to look like to win, and against whom?
How should the phase 3 trial be designed, e.g. endpoints, eligibility criteria, number of patients?
What has the agency already said about this endpoint, this dose, this population?
How long will this program take, benchmarked against the ones that have run?
Where is it most likely to fail, and what did the regulator say before the ones that did?
Who else is developing assets in this space, how far along are they, and when do they read out?
HOW IT WORKS
“Find the phase 3 obesity programs and their approvals.” “What did FDA say to sponsors about dose selection in heart failure?” One question, one call, a structured answer. The QD grammar allows anyone to chain queries directly.
Every document is acquired and hashed, then extracted into a typed card by a model, then audited verbatim against its own source — a card that disagrees with the document it came from is flagged, not published.
Source table, source key, extraction tier, source URI. Click a number in a QD deliverable and land on the FDA review PDF, the EPAR or the registry record. Where the corpus does not have something, QD says so rather than filling the gap.
A figure on a board slide with no path back to its source is a claim, not evidence.
WHAT IS DELIVERED
Each one is produced in an afternoon and is reproducible: run it again in six months and you get the current picture, not a stale snapshot.
USE CASES
One platform across the development path — from clinical plans to regulatory approval.
Positioned against every approved label in the indication rather than against memory. Efficacy, safety, dosing, population and label claims, attribute by attribute, with the pivotal trials behind each one.
The whole competitive field, verified, in an afternoon. What the asset has to beat, who else is in the race, what the regulator has said about the mechanism, and where comparable programs came apart.
Every assumption benchmarked to trials that actually ran. Sample size, duration, sites, recruitment rate and dropout drawn from the comparable programs rather than from a planning template.
Your asset ranked against the real class, by phase. Which programs are genuinely comparable, where yours sits on timeline and evidence, and what that means for sequencing and funding.
Prepare for agency questions using competitor meeting minutes. The questions that get asked in your indication, the answers that satisfied, and the ones that did not.
Precedent from every review discipline, source attached. What reviewers raised in comparable submissions, which issues produced complete response letters, and what the sponsors did next.
How to GET STARTED
QD [Quantify Decisions] drug development platform is a paid service. To request access, email contact@pharmology.dk.
QD-MCP connector in your AI agent
QD-MCP is listed in the Claude connector directory: add it once and the QD [Quantify Decisions] platform answers inside your AI agent.
Using another MCP client? Add the server directly at:
https://qd-mcp.pharmology.ai/mcp
QD web application
Point, click, download. Use the Trial Explorer, Competitor Insights, Regulatory Explorer, and Product Plan modules to find the information you are looking for.
WHAT IS UNDERNEATH
The scale matters only because of what it makes possible. It is here for the people who want to know what they are standing on.
Over one million trial records. ClinicalTrials.gov records include per-arm adverse events, outcome measurements by endpoint and arm, site locations, baseline characteristics and dropout with stated reasons.
FDA, EMA, and other major national regulatory agencies. Review documents, approval actions, labels, complete response letters and assessment reports — extracted from the PDFs, not linked to them.
Thousands of sponsor-to-FDA meetings and tens of thousands of question-and-answer pairs, verbatim: the sponsor's question and the agency's response, kept together, pulled out of administrative correspondence that no commercial database ingests.
FDA guidances, EMA guidelines and other major regulatory agency guidelines, extracted into typed, searchable cards rather than left as PDFs. What the agencies have published they expect — on endpoints, trial design, surrogate measures, populations — so you can check your approach against it before building a program around it.
Briefing documents, panel questions and votes from FDA advisory committee meetings. This is where a drug’s weaknesses get argued in the open, by the reviewers themselves — the clearest public signal of what the agency doubts about a mechanism or an endpoint, before it reaches a label.
Over one hundred thousand drug-name aliases, mechanism-of-action mapping, condition synonyms and MeSH. A brand name, an INN and a code name resolve to the same molecule, so a question about one asset returns all of its evidence rather than the fraction that happened to use the name you typed.
Privacy
QD [Quantify Decisions] is built on public sources and is read-only. Your searches are never used to train anything, and never enters the shared corpus. More information about our privacy policy when using QD - see Privacy Policy.
Pharm·ology ApS · contact@pharmology.dk · CVR 45557898

PRODUCT