Measuring the activity of DIPs remains one of the most challenging aspects of quantifying digital trade. As highlighted in the Handbook on Measuring Digital Trade (IMF et al., 2023[1]), international guidance on the compilation of statistics relating to DIPs remains at an exploratory stage. Past efforts to identify and classify DIPs – particularly those operating across borders – have often relied on manual data collection and classification methods due to the absence of a standard definition. While much of the early work has focused on peer-to-peer platforms and has not systematically addressed the international trade dimension, it provided valuable methodological insights and laid the groundwork for more structured, internationally comparable approaches.
Recent work by national statistical offices and international organisations is helping to close this gap. The U.S. Bureau of Economic Analysis (BEA), for instance, has produced experimental estimates of digital intermediary services as part of its Digital Economy Satellite Account, covering sectors such as ridesharing, travel, and food delivery (Highfill and Quistorff, 2023[16]). These estimates, which span activities not easily classified under existing industry categories, suggest that digital intermediation in these three sectors alone accounted for at least USD 31 billion in gross output in 2021. UNCTAD (2024[17]) has also contributed to understanding the scale of DIP activity by analysing gross merchandise value and transaction value as reported by platforms in financial statements and filings. However, differences in reporting practices and data availability continue to constrain international comparability.
Building on these, this paper applies and extends the methodology developed by Costa et al. (2021[5]) to construct indicators of DIP uptake and diffusion across OECD Member countries. The analysis focuses on three key consumer-facing sectors: Travel and accommodation booking; ride-hailing and carpooling; and X2C online marketplaces – using monthly website traffic (unique visits across devices and cleaned for known bot and automated traffic (Semrush, 2026[18])) as a harmonised, scalable proxy for platform reach. While not a direct measure of transactions, revenue or market share – and subject to caveats such as seasonal tourism or network centrality resulting from the jurisdiction of major online platform headquarters – this traffic-based indicator effectively captures relative digital presence and enables cross-country, cross-sector, and time-series comparisons.1
The resulting dataset covers around 860 digital intermediary platforms (DIPs) operating across travel and accommodation booking, ride-hailing and carpooling, and X2C online marketplaces (see Annex A for more details). This includes both “pure” DIPs, which solely facilitate transactions between buyers and sellers, and “hybrid” DIPs, which combine intermediation with direct retail activity (IMF et al., 2023[1]). This distinction is particularly relevant for X2C marketplaces, where hybrid platforms tend to invest more in physical infrastructure (e.g. logistics, warehousing) and face greater exposure to local regulations, while pure intermediaries can scale more readily across borders through digital-only operations.
These structural differences shape how digital trade policies affect platforms. Some measures may disproportionately restrict entry or operation for pure DIPs that rely on cross-border digital delivery, while others may encourage hybridisation as firms adapt their market access strategies. Understanding the origin of DIPs is therefore critical to analysing the interaction between digital trade policy and platform activity. Restrictions often affect foreign service providers more than domestic ones, making it essential to distinguish between foreign and domestic DIPs. In this paper, platform origin is identified based on the location of the parent company or global headquarters, using public data from sources such as Crunchbase and Pitchbook. Although this approach does not capture complex ownership structures, it provides a transparent, scalable, and replicable classification suited to large-scale comparative analysis.2
On this basis, three novel indicators are constructed for each country, sector, platform type, and time: total website traffic from foreign-headquartered DIPs (or “foreign traffic”); total traffic from domestic-headquartered DIPs (or “domestic traffic”), and the share of total traffic attributable to foreign headquartered platforms (or “share of foreign traffic”). Together, these indicators provide new insights into the reach of foreign platforms across markets and sectors, helping assess how regulatory environments and market structures shape digital intermediation.