The rise of the gig economy has challenged economists and policymakers alike. Its energy is unmistakable, but its contours remain stubbornly hard to define. Ride-hailing apps and digital freelance platforms now shape the livelihoods of millions, but how many? In which sectors? And how are incomes distributed across these new arrangements? These are not idle questions. As gig work becomes a fixture of labor markets, clarity matters for tax policy, social protection, and economic forecasting.
The International Standard Industrial Classification (ISIC) system, while not designed with gig work in mind, is proving adaptable. Two codes are particularly relevant. Ride-hailing platforms, now ubiquitous in urban centers worldwide, are generally classified under ISIC 4922—“Other passenger land transport.” Digital freelancing services, meanwhile, map best to ISIC 6209—“Other information technology and computer service activities.” Together, these codes offer a structured lens through which to measure at least two of the gig economy’s largest segments.
Classification is the first challenge. Many platforms do not fit tidily into legacy industry codes. Some offer hybrid models—ride-hailing, food delivery, courier services—under a single app. Others broker digital services ranging from graphic design to translation. Assigning an ISIC code requires understanding the dominant business activity, not simply what is most visible to users. For national statistical offices, this often means direct engagement with platforms, requesting firm-level data on service volumes, revenue streams, and—ideally—breakdowns by service type.
Once classification is in place, the next step is data extraction. Some countries now maintain registries of digital platforms, often for regulatory or tax purposes. These databases can yield valuable information: number of registered workers, volume of transactions, or geographic spread. Where public data is limited, researchers may turn to platform-provided statistics or, increasingly, to scraping publicly available figures from apps, websites, or company reports. Each approach has strengths and limitations. Platform self-reporting can be selective or incomplete; scraping may miss nuance or detail.
Surveying participants themselves—drivers, coders, designers, and others—remains crucial for understanding the lived reality of gig work. National labor force surveys are adapting, often including new questions about platform work in annual rounds. Key variables to capture include primary versus supplemental income, hours worked, number of platforms used, and subjective measures of job security or satisfaction. Incomes can be highly variable, and averages risk masking sharp disparities. It is not uncommon for a small percentage of high-earning freelancers to skew the picture, while the majority report more modest gains.
Within each ISIC category, it is instructive to analyze not only participant numbers, but also income distribution. For ride-hailing (4922), earnings may depend on city, time worked, and surge pricing algorithms—factors that are sometimes visible in aggregate data but usually require direct survey input to interpret properly. For digital services (6209), skill level, client base, and even language proficiency play a role. Cross-tabulating income brackets with demographic data (age, education, location) deepens the analysis and points toward policy needs.
Guidelines for researchers might begin with clear sampling: identify the universe of relevant platforms, both global and local, and ensure that both large and niche providers are represented. Collaborate where possible with platform operators—some are willing to share anonymized data, especially if it improves public understanding and shapes fair regulation. Supplement this with targeted surveys, focusing on representative cross-sections of participants within each ISIC code. Be transparent about limitations: undercounting is likely, especially where gig work is informal or cross-border.
Comparisons across jurisdictions are only as good as the underlying classification. Definitions of “platform work” still differ between countries, and not all national registries align their categories perfectly with ISIC standards. Adjusting for these inconsistencies—by documenting coding choices, clarifying inclusion criteria, and noting divergences—is essential for credible cross-country research.
Finally, policymakers using ISIC-coded gig economy data must resist the urge to focus solely on growth. Expansion, while headline-grabbing, is only part of the story. Earnings quality, access to benefits, and the stability of gig work all merit careful attention. Trends in registration or transaction volume may signal economic transformation, but they do not substitute for understanding the conditions under which people work.
As the gig economy matures, the importance of structured, reliable measurement will only grow. ISIC codes—though imperfect—offer a common language for analysts and decision-makers seeking to move beyond anecdote. The goal is not just to count, but to comprehend: to see both the scale and the substance of change in the new world of work.