In 1999, the idea of buying books, shoes, or even groceries online still struck many as either wildly futuristic or quietly risky. Yet the undercurrent of commercial innovation was unmistakable. For economists or statisticians hoping to track the expansion of early ecommerce platforms, the search starts with an imperfect tool: ISIC 6201—computer programming activities. In theory, this should catch a fair portion of the digital pioneers, but, as always, the practical story is much more complicated.
The process begins with the assembly of all firms registered under ISIC 6201 in the United States as of 1999. This list is long, populated by everything from legacy software houses to hopeful new startups. Many companies were founded that year with little more than a rented office, a handful of coders, and a business model built around “web commerce.” But the registry doesn’t spell out intentions, and certainly doesn’t differentiate a pure-play ecommerce platform from a generic IT consultancy. Filtering becomes essential.
To isolate true ecommerce platforms, supplementary evidence is required. Archived business descriptions, press releases, and—perhaps most revealing—snapshots of early company websites all play a part. Some firms were explicit: “the leading online marketplace for X,” or “pioneering internet retailer.” Others were cagey, perhaps unsure themselves of exactly how digital sales would pan out. Analysts must read between the lines, sometimes cross-referencing with the earliest reports from trade associations, digital economy roundtables, or even technology conference attendee lists. A mention in the right context—a NASDAQ debut, a dot-com award, a merger—can confirm an ecommerce focus, even if the paperwork remains vague.
Domain registration data is particularly useful here. The late 1990s saw a rush to secure memorable “dot-com” addresses, and, where records remain, one can link domains to company names, founders, and occasionally even investor groups. A surge in registrations around 1998 and 1999 often signals new ventures staking out their virtual territory, with some firms managing dozens of domains as hedges against future expansion or competition.
Connecting these digital footprints to revenue growth is where things become less tidy. Public companies, even in their early years, sometimes broke out ecommerce revenues in quarterly or annual reports, keen to impress investors with rapid adoption curves. For the vast majority of private or pre-IPO companies, figures are harder to find. Clues come from funding rounds, staff headcount growth, or even regional news articles covering warehouse expansions or fulfillment partnerships. An uptick in site traffic, press mentions, or customer testimonials, while not a direct proxy for revenue, at least signals traction.
Some researchers have used web traffic analytics from the period—where available—to estimate potential sales volumes. Others piece together changes in employee numbers, warehouse square footage, or third-party fulfillment contracts as indirect evidence of expansion. The methodology is never seamless. Domain registration and firm-level revenue rarely line up in a one-to-one fashion; some companies sat on domains for years before turning a profit, others pivoted away from ecommerce entirely after lackluster returns.
Another complication comes from the churn: many platforms that made a splash in 1999 were gone by 2002. Mergers, bankruptcies, and business model pivots are part of the terrain. Tracking name changes, successor companies, or acquired domains is a necessary, sometimes exhausting, part of documenting true expansion. Every assumption—about which firms to count, how to interpret missing data, what to do with dormant domains—should be recorded for future review.
What emerges is a sector in rapid motion: firms racing to stake out digital territory, revenues rising unevenly, and the infrastructure of internet commerce taking shape in real time. ISIC 6201, filtered and combined with domain and revenue data, offers a rough but illuminating lens on the forces driving the early ecommerce boom in the United States. The patterns aren’t perfectly symmetrical. There are blind spots and dead ends. Still, for those patient enough to piece the evidence together, a clear sense of expansion—and the wild uncertainty that came with it—begins to surface, year by year, company by company, domain by domain.