Economy

The API economy is the idea I started API Evangelist on, and it’s right there in the very first substantive thing I wrote, on September 25, 2010, titled simply “The API Economy.” My opening line was that APIs are driving the internet and our economy — that developers rely on them for data and functionality, that businesses exchange data and services through them, that the press uses them for real-time analysis. And I made the comparison that has defined my entire career: in 2000, businesses were still working to understand the importance of having a website; in 2010, businesses were faced with the growing importance of having an API. That framing — APIs as the next inevitable layer of business infrastructure, as fundamental as the website became — is the seed from which everything else grew. The API economy was never a marketing term for me. It was a genuine observation that economic value was increasingly being created, exchanged, and captured through APIs, and that this would only accelerate.

The concept matured quickly in those early years, and I spent a lot of energy trying to make it legible to people who weren’t developers. The business of APIs primer I wrote in 2011 was aimed squarely at executives and decision-makers, because I understood early that the API economy would only become real if business people, not just engineers, understood it. I wrote about visions from the API economy, about the four potential levels of an API business ecosystem — internal, partner, public developer, and broadly open — because the economy isn’t one thing; it’s a layered set of markets, each with its own participants and value flows. APIs were becoming the mechanism through which companies exposed their capabilities to partners, to developers, and to the public, and each of those exposures was an economic act. The API economy was the sum of all those acts: every integration, every transaction, every dependency, every partnership mediated through an API.

The transaction insight is the one I keep returning to, and it’s the deepest thing I understand about what the API economy actually is. APIs reduce everything to a transaction — I wrote that explicitly in 2017 and expanded on it in 2024, and I mean it as both a description and a warning. The genius of the API is that it takes some capability — sending a message, processing a payment, checking an address, generating an image — and reduces it to a clean, discrete, repeatable, meterable transaction. That reduction is exactly what makes the API economy possible: once a capability is a transaction, it can be priced, sold, counted, rate-limited, and composed with other transactions. The whole economic machinery of APIs — the pricing plans, the rate limits, the usage metering, the marketplaces — depends on this reduction of messy human and business activity into countable transactions. But I’ve always paired the insight with its shadow: when everything is reduced to a transaction, including human attention and human relationships, you get the commodification and surveillance dynamics that the API economy also enables. The reduction is powerful and it is double-edged, and an honest account of the API economy has to hold both.

Value is the concept underneath the transaction, and I’ve been precise about where value actually lives. An API has no value until it is actually used, I wrote in 2015 — which sounds obvious but cuts against a lot of how people talk about the API economy. The value isn’t in the API definition, or the infrastructure, or the potential; it’s realized only at the moment of consumption, when a real transaction occurs that delivers something a consumer needs. This is why I’ve thought so much about how you measure value in the API economy — about Postman collections as a unit of measurement for transactions, about OpenAPI representing the value possible while the executed call is evidence of value actually delivered. The API economy is not the universe of APIs that exist; it’s the universe of API calls that actually happen, and the value flows along those calls. Measuring the API economy means measuring real consumption, not counting endpoints.

The maturation of the API economy is a story of moving from hype to plumbing, and I welcomed that shift even though it was less exciting. I wrote in 2016 that we should prepare for the more boring and very business age of APIs — and I meant it as a good thing. The early API economy had a lot of breathless hype, a lot of conference-stage enthusiasm, a lot of inflated claims. What I wanted was for APIs to become boring infrastructure — to recede into the plumbing of how business actually gets done, the way electricity and the web receded. APIs are just the next step of the web and not the latest trend, I argued in 2016, and the proof of that is precisely that the API economy stopped being a hype topic and became an unremarkable operational reality. By the time 83 percent of the Fortune 100 have a developer or API portal, the API economy isn’t a vision anymore; it’s just how large enterprises operate. That’s success, even if it’s less thrilling than the early evangelism.

The skeptic’s challenge has always shadowed the term, and I’ve never run from it. People have questioned whether the “API economy” was ever real or just a buzzword invented by vendors to sell API management, and that skepticism deserves an honest answer. My answer is that the term was sometimes overhyped by vendors, yes — but the underlying reality it pointed at was always genuine. The fact that I had to write a post in 2017 literally titled about convincing people the API economy is the future tells you the skepticism was real and persistent. But fifteen years of evidence has settled the question. Stripe processes a meaningful fraction of internet commerce through its API. Twilio rebuilt communications as API transactions. Open banking is regulation mandating that an entire industry expose its capabilities through APIs. The cloud is APIs all the way down. The skeptics who said the API economy was hype were wrong, not because the term was never abused, but because the phenomenon it named — economic value increasingly created and exchanged through APIs — turned out to be one of the defining economic shifts of the era.

The history reminds us the API economy is both newer and older than people think, and I’ve used EDI to make this point. Looking at Electronic Data Interchange, I wrote in 2020, reminds me that the API economy is just getting started — because EDI was businesses exchanging structured data to do commerce since the 1970s, the grandfather of the API economy, proving the fundamental impulse is decades old. And yet the modern, web-based, self-service, developer-driven API economy was still in its early innings even in 2020. The economic impulse to connect systems and exchange value programmatically is ancient; the lightweight, accessible, web-native form it took in the 2000s and 2010s is what made it explode. The API economy is the latest and most accessible expression of a very old idea about machine-to-machine commerce.

The public and political dimensions of the API economy are where I’ve insisted the conversation can’t just be about business. The politics of the API economy, which I wrote about in 2015, is about who holds power in these markets — the asymmetries between platforms and developers, the control platforms exert, the way value flows toward those who own the infrastructure. And the public data dimension is enormous: our economy depends on a steady flow of data from the public sector, and APIs are how we quantify, protect, and channel that public value, as I argued in helping the public data commons drive our economy. API literacy, I wrote in 2013, is as important as financial literacy in the API economy — because if economic value increasingly flows through APIs, then understanding APIs becomes a precondition for full economic participation, and the gap between those who understand and those who don’t becomes an inequality that matters.

When I look back from 2025 at that first 2010 blog post, what’s striking is how much was right and how much has shifted. I wrote a deliberate retrospective comparing the 2010 vision to the 2025 reality, and the honest assessment is mixed. The core thesis held: APIs did become essential business infrastructure, the API economy is real and pervasive. But the balance shifted in ways I didn’t fully anticipate — toward consolidation and platform power, toward enclosure, away from some of the open, accessible, commons-spirited promise of the early days. The community changed, the business models hardened, the simplicity I valued got buried under complexity. And now AI agents are arriving as the newest participants in the API economy — but as I’ve insisted, agents are just APIs, nothing fundamental has changed. An agent transacting through APIs is the API economy doing exactly what it always did, with a new kind of consumer. The API economy I named in 2010 is now the substrate of the digital economy, the connective tissue through which an enormous and growing share of all economic value flows. I called it early because it was already visibly happening. It just hadn’t been named yet.

Guidance

API Learnings

APIs

API Governance

API Solutions

API Discovery

API Building Blocks

API Evangelist LLC

Economy

References

Guidance Details

Explore more

Discover APIs