Where the data comes from, what the model can't do, and the institutions we are — and are not — connected to.
Cohortly is independently operated. We are not affiliated with, endorsed by, or partnered with Harvard, Yale, Princeton, Stanford, MIT, Columbia, UPenn, Brown, Dartmouth, Cornell, UChicago, Duke, or any other institution referenced in our materials. School names are used solely to identify the cohorts our reference set was modeled on.
The reference set is a curated panel of 500+ verified accepted profiles, contributed voluntarily and anonymized at intake. Public sources — Common Data Sets, IPEDS releases, and institutional admissions reports — calibrate baselines. We do not ingest scraped admissions data and we do not ingest content from any institution's internal systems.
Where a feature has insufficient cohort data we publish a low-confidence flag rather than imputing. Cohort sizes for school-specific fits are disclosed on the Premium PDF.
Cohortly revenue comes from one-time Premium Reports and advisor bundles. We do not accept payment from institutions, admissions consultants, or essay services to influence model output, school-fit ranking, or recommended next steps.
Statistics on the homepage (profiles benchmarked, features modeled, post-recommendation percentile shift) are based on internal cohort and engagement data as of the most recent cycle. Methodology is described in § I — The Method.