Research and methodology
We publish how our data is built and how it holds up against the gold standards. Transparency is not a tagline for us, it is how a data company earns trust.
How TrialForge benchmarks against the gold-standard providers
We compared our clinical trial dataset head to head with TrialTrove, the industry gold standard, across coverage, classification, and recency on ClinicalTrials.gov trials. The headline result, shown here in full, is parity on ClinicalTrials.gov trials, with honest disclosure of where the boundaries are.
Why a general-purpose LLM cannot do this
We also tested what a leading general-purpose model produces when asked to classify trials from the open web. It invented trials, misclassified modalities, and could not cite its sources. That gap is the reason TrialForge exists: rigorous data engineering and human review, not a thin wrapper over a public model.
White papers, landscapes, and method notes
This is where we publish our clinical trial landscape analyses and method notes.
Research note
We added China’s drug trial registry to TrialForge. Here’s what you’d be missing with CT.gov alone
Three of every four CDE trials is not on ClinicalTrials.gov. A 63 percent bigger picture of China, plus thousands of novel CDE only drug assets.
Research note
We tested TrialForge's coverage of cardiorenalmetabolic assets
The audit, transparently: 100 percent of clinical stage assets with a ClinicalTrials.gov trial, held in TrialForge.
Who builds this
TrialForge is built by an ex-BCG team that spent a decade advising biopharma on the pipeline, licensing, and portfolio decisions this data informs. We built the dataset we wished we had.