Who Is the Data Science Consultant? An Analysis of the CNSL Directory
What clustering 949 statistical-consultant profiles reveals about specializations in the profession, and which niches AI will and will not touch.
In 2026 the AmStat Statistical Consulting Section announced it would redesign its Consultant Directory. Before any data was lost, all 949 active profiles were scraped, including affiliations, skills, specializations, and languages. A K-medoids analysis identifies archetypal consultants, which gives both a picture of the section's membership and a map of specializations in statistical consulting as a distinct profession. The silhouette criterion suggested an unusually high number of clusters (k=23) relative to the size of the data set. Looking closer, the second-nearest medoid for each profile turned out to take only 10 of the 23 values, revealing a many-to-many relationship between primary and secondary specialties. Key archetypes include a dominant academic biostatistician group (84% PhD) focused on clinical trials, alongside distinct specialties like industrial quality engineers and private-sector survey specialists. The results suggest a real divide between academics and independent practitioners, that a one-size-fits-all directory may fit none, and several niches that look like new opportunities, with AI affecting each unevenly.