Researching OpenAI at Gain: From Fragmented Sources to a Usable Investment Profile
Owned a full company profile refresh on OpenAI, triangulating conflicting sources into a structured research output for investment clients.
Swastik N.
Private Equity Intelligence - Junior Analyst at Gain
From their time as
Private Equity Intelligence - Junior Analyst
Gain • 2025
Overview
Swastik was tasked with refreshing Gain's company profile on OpenAI, a high-profile private company where information was abundant but often conflicting, outdated, or sourced from anonymous leaks.
The Story
Swastik was tasked with refreshing Gain's company profile on OpenAI, a high-profile private company where information was abundant but often conflicting, outdated, or sourced from anonymous leaks.
His first step was not to start reading. It was to identify what had actually changed since the previous update and which changes would materially affect the investment narrative. He scoped the work around four research questions: what are the major revenue streams, how is enterprise adoption evolving, what differentiates OpenAI from Anthropic, DeepMind, and xAI, and what strategic risks exist.
Triangulating conflicting sources was the core challenge. Different publications quoted different revenue figures: some cited annualized numbers, others actual, and several relied on anonymous sources. Rather than picking the first credible figure, Swastik applied a source hierarchy. Official company disclosures and blogs came first. Strategic partner disclosures, such as Microsoft's earnings call commentary on commercial traction, came second. Highly credible publications like the FT, Reuters, and Bloomberg came third. Broader media coverage was used for context only.
He applied the same discipline to customer numbers. When one source cited two million enterprise users and another cited three million paying business accounts, he did not average them. He first checked whether they were measuring the same thing: enterprise seats versus paying business accounts are not the same metric. He resolved the conflict only after confirming what each source was actually counting.
For product launches, he used official OpenAI announcements for factual descriptions and media coverage only for market reaction context.
Once the research was complete, he structured the profile so that someone unfamiliar with OpenAI could follow it logically: business overview, products, business model, customers, competitive positioning, funding history, and recent developments. Before submission, he checked for factual consistency, removed duplicates, verified dates, and ensured the narrative flowed.
The final output was a comprehensive company profile used by clients for commercial due diligence, market mapping, and investment research.