Doximity: 3 Key Takeaways for Long-Term Investors
Compute cost discipline, not just user growth, is what lets Doximity scale AI without eroding margins as usage climbs.
Doctoxity's fiscal 2027 second-quarter earnings call on November 5, 2026, highlighted three key takeaways for long-term investors. The company's revenue reached $175 million, driven by the growing adoption of artificial intelligence (AI) across its platform. Clinical professionals rely on Doximity's digital tools for their daily medical workflows, providing a captive audience that attracts pharmaceutical advertising spend.
Management emphasized the company's efficient compute management and its ability to maintain profit margins as AI usage scales.
One significant development was the adoption of Doximity's new AI search capabilities by pharma clients. This feature allows pharmaceutical companies to engage physicians at the point of care using keyword-driven search, tapping into a $14 billion annual market for digital pharma budgets. Currently, 16 of the top 20 pharmaceutical companies have purchased Doximity's AI search commercial offering, with three more in the contracting phase.
Broad physician engagement forms the foundation of Doximity's platform value. As more doctors utilize digital scheduling, telehealth, and AI clinical tools, the platform does not need to spend heavily on acquiring new users. Quarterly active workflow prescribers, which include AI users, grew 34% year-on-year to record levels, accelerating from the previous quarter.
Efficient management of complex AI models is crucial to avoid eroding software margins. Doximity pre-bought compute capacity and optimized its technical infrastructure, resulting in revenue cost growth outpacing prompt volume growth. This efficiency was further enhanced by pre-buying capacity at favorable economics, wider multimodal usage, and continued performance gains in their proprietary harness.
Written by urgent.news from Motley Fool's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.