If there’s a throughline in the story of ecommerce, it is how rapidly the landscape outpaced the statistical categories meant to capture it. In 2002, “online marketplace” was a concept familiar only to a small segment of the public, but to those tracking economic activity, the signs were there—hidden, perhaps, but real. ISIC 6311, “Data processing, hosting, and related activities,” served as the principal statistical anchor, but it is, at best, a wide net. Not all ecommerce businesses saw themselves as data hosts, and not all data hosts ventured into online commerce. Still, for the analyst, it is a starting point.
The first practical step is to assemble a comprehensive list of firms under ISIC 6311 for the target year. This population, in most countries, tends to be a mixed bag. Some entities offered website hosting or server management. Others were software houses dabbling in web services. Scattered among them were the first true online marketplaces—platforms bringing buyers and sellers together, handling transactions, and, crucially, holding a piece of the digital value chain.
Isolating the marketplaces from this wider set requires layering in external evidence. Company descriptions, early press coverage, and even archived versions of business websites can be surprisingly useful. In 2002, the language is rarely standardized: “internet shopping,” “web auction,” “digital storefront,” and “marketplace” are terms that appear, but inconsistently. Analysts have to develop a kind of ear for the period’s promotional style, recognizing both what is said and what is left implied.
Domain registrations become a valuable data point. National and international registries usually provide access—sometimes public, sometimes via subscription—to lists of active domains, often cross-referenced with registrant details. By linking domain ownership back to ISIC 6311 firms, it is possible to see which businesses staked an early claim in the online space. Not every domain translates into an operating marketplace, but clusters of registrations around particular companies—especially those that launched multiple storefronts or regional platforms—point to genuine engagement.
For larger marketplaces, press releases or news archives sometimes document launch dates, funding rounds, or traffic milestones, each providing a temporal marker that can be matched to registry and domain data. Yet it is revenue that really matters to most policymakers and economists. The connection between domain registration and firm-level revenue is, unsurprisingly, a leap. Many companies in this period operated at a loss for years, prioritizing market share over immediate profit, and very few broke out their ecommerce segment revenues in financial filings.
Some signals can be gleaned from investor updates, tax filings, or sector surveys. Where available, these sources reveal how much of a firm’s income derived from transaction fees, advertising, or affiliate marketing. For marketplaces embedded within larger corporate structures, indirect evidence—like staffing changes, hiring for ecommerce-specific roles, or investment in payment infrastructure—can bolster assumptions about revenue trajectories. Publicly traded companies sometimes referenced “internet operations” as a separate business line, with revenue trends reported quarterly or annually.
Aligning these multiple data streams is painstaking. A single marketplace may have registered domains in several jurisdictions, operated through holding companies, or rebranded multiple times in its early years. Mergers and acquisitions further complicate the genealogy. Careful matching—documenting assumptions and cross-checking company histories—remains crucial for accuracy. It is easy to miss a major player that changed names, or to overcount by including dormant domains.
The analyst must accept that the data is partial. Not every online marketplace registered under ISIC 6311 will have left a digital footprint substantial enough to track. Some operated only briefly; others morphed into different business models as the decade wore on. Nonetheless, by layering domain registration data with revenue disclosures and public records, it is possible to trace the growth trajectory of early ecommerce.
What emerges is not a perfectly symmetrical picture, but a sense of movement—an expansion of digital commerce as both an economic sector and a social force. The boundaries between “data host” and “marketplace” were never as firm as later industry reporting might suggest. Still, within the ambiguity, there are enough signals to reveal the shape of change as it unfolded, year by year, firm by firm, even as the formal classification system played catch-up.