Test your thinking against the public record
Not a prediction of what the committee will do. A dry-run of your strategy against what they have already published: every Public Summary Document, structured once so you can ask a real research question and get a citable answer.
An independent project built and maintained by one person. The machine is documented in the open so the work is not magic.
Start with a research question
That question can be broad, a full submission strategy: comparators, evidence patterns, economics language, restriction risk, what failed in class. Or specific: one comparator, one pattern, one resubmission path, one messaging claim against the record.
Same public corpus either way. Market access and brand teams often open different doors; the trail underneath is the same.
The public record, structured
Australia already publishes PBAC outcomes and Public Summary Documents. We extract and structure them so the precedent leg is a table and a search, not a week of PDF archaeology.
- Field Matrix: shortlist and compare
- Drug pages: history, fields, links to the source on pbs.gov.au
- Grounded claims: if a quote is not an exact line from the public summary, we do not show it
Open the matrix: /matrix/
How to use it
- Explore. Browse and search the record. Share a URL with your team. No seat required for the public surface.
- Shortlist and compare. The Field Matrix and Studio turn a research question — a comparator, a class, a resubmission path — into a filtered view you can cite, with every cell traceable to its source PSD.
- Ask. Something the dashboard doesn't do — a data cut, a correction, a question about how a field was extracted — send a note and I'll point you at the right slice of the record.
Marketing and HEOR are two entrances to the same record. Whether the question starts from the campaign or the submission, the trail underneath is the same.
How we stay honest
- Public data only. The Australian PBAC and PBS trail; no confidential net prices.
- Cite or silence. Commentary and endpoint phrases we display are exact substrings of the source, or they are blank.
- Systematic QA. Automated checks catch taxonomy errors, for example endpoint labels that do not belong outside oncology; known-bad patterns are suppressed and logged.
- Process literacy. The PBAC recommends; the PBS lists. A recommendation is not a listing until price, restriction and deeds land.
- Not a crystal ball. We rehearse against the published record; we do not sell a probability that replaces your judgment.
- AI organises the public record; humans own the judgment. No confidential submissions in our models; no black-box substitute for your call.
We do not pretend every historical cell was perfect. When the audit routine finds a board-wide error, we fix the rule and record it.
Who this is for
Market access and HEOR teams preparing or defending a filing. Brand and medical affairs teams who need the public story to survive contact with the assessment trail. Affiliates who need AU depth without rebuilding a library from scratch.
An independent project — not a seat platform, not a global HTA mega-suite.
Get in touch
Questions about the data, corrections, or a research question the dashboard can't answer: josh@script.report.
Or open the record first: Field Matrix · About the data · Insights