While I’ve been at Google now for about 2 years, I’ve had hundreds of conversations with teams about API design. One thing I’ve seen challenging for organizations is understanding the business impact of APIs, specifically, the correlation between leveraging APIs and generating revenue. The biggest challenge is to migrate thinking away from APIs as a technology integration, not even APIs as products, but APIs as a consumer language. And I’m not referring to API consumers either.

Organizations need to focus on their primary customer, their people interfaces, with a new story-driven language of intent and semantics. Moving from RESTful, to INTENTful.

For over four decades, CRUD (Create, Read, Update, Delete) has been the lingua franca of data operations. Born in the 1980s when databases were simple record-keepers and applications were straightforward interfaces, these four verbs have served us remarkably well. But as we hurtle deeper into the age of artificial intelligence, it’s worth asking a heretical question: is CRUD fundamentally mismatched for how AI systems actually work?

The answer is yes. And the implications are profound.

The CRUD Paradigm Was Built for Humans

CRUD emerged from a world where humans made explicit decisions about data. A user clicks “save,” and a record is created. They edit a form, and a record is updated. They press delete and, well, you get the idea. These operations are transactional, deterministic, and immediate. They assume perfect knowledge at the moment of action.

AI systems operate nothing like this. They don’t make single, definitive updates. They infer, predict, and learn continuously. They handle uncertainty as a first-class concern. They understand context, relationships, and temporal dynamics in ways that CRUD was never designed to accommodate. When an AI model processes information, it doesn’t just “read” data; it transforms it through layers of understanding. When it “updates” its knowledge, it’s not replacing a field in a database; it’s adjusting the weights across millions of parameters using probabilistic reasoning.

We’ve been forcing AI into a CRUD-shaped box, and it’s time to redesign the box.

From Database Operations to Intentful APIs

What if instead of forcing every interaction through CRUD’s database-centric lens, we designed API methods that speak the language of what users and AI systems actually want to accomplish? Methods that express intent, not just data manipulation?

Consider these examples:

BOOK /flight — This isn’t just creating a reservation record. It encapsulates business logic: checking availability, holding inventory, calculating pricing, processing payment, sending confirmations, and updating multiple systems. When a consumer or AI workflow wants to book a flight, they shouldn’t have to orchestrate a dozen CRUD operations. The intent is atomic: book this flight.

SCHEDULE /meeting — More than creating a calendar entry. This involves checking participant availability, finding optimal time slots, sending invitations, handling conflicts, setting up video conferencing, and managing reminder workflows. The business intent is “make this meeting happen,” not “POST to /calendar_events.”

QUERY /products — Not just reading from a database, but understanding search intent, applying relevance ranking, personalizing results, checking inventory across locations, calculating dynamic pricing, and returning semantically meaningful results. The user wants to find products, not execute SELECT statements.

SUMMARIZE /document — A perfect example of AI-native intent. This is about comprehending, analyzing, extracting key points, and generating a coherent summary. The operation is fundamentally about transformation and intelligence, not data retrieval.

RECOMMEND /content — Understanding user preferences, analyzing behavior patterns, computing similarity, applying business rules (promoting certain content, respecting geographic restrictions), and returning personalized suggestions. The intent is discovery, not database queries.

VERIFY /identity — More than reading user records. This encompasses multi-factor authentication, risk assessment, fraud detection, compliance checking, and session management. The business need is “confirm this is who they say they are,” which may involve dozens of underlying operations.

These aren’t just syntactic sugar on top of CRUD. They represent a fundamental shift in how we think about API design. Intentful APIs move the focus from database operations to business capabilities, from technical primitives to user intent.

Why This Matters for Business Value

The gap between CRUD and intent-based methods lies in the traceable business value and revenue generation. This is where the correlation between APIs and business outcomes becomes crystal clear.

When an API relies on basic CRUD operations, connecting it to business impact can be challenging. Tracking revenue across technical calls such as “POST /reservations,” “PUT /payments,” and “POST /confirmations” requires extra effort. The actual transaction, a booked flight, is distributed across several operations, making analytics complex and time-consuming.

APIs designed around business intent simplify this process. A BOOK /flight call directly represents a sales opportunity. Conversion rates, abandonment points, and pricing insights can be measured at the intent level. Each call connects naturally to revenue and business performance, turning the API into a source of meaningful metrics.

As AI tools become more common API consumers, this alignment grows even more valuable. A travel assistant powered by AI interprets a task as “book this flight,” not “create a reservation record.” Intentful APIs that reflect purpose and business meaning support smooth collaboration between humans, systems, and AI. Instead of technical friction, organizations achieve greater clarity, alignment, and impact.

Expressing business capabilities through APIs also helps product teams communicate effectively. Product managers can prioritize RECOMMEND /products over QUERY /products when recommendation drives higher conversion. They can focus on the SCHEDULE /meeting when coordination supports the company’s value proposition. Intent-based design builds shared understanding between business and technical teams.

The same approach enhances monetization. Value-based pricing becomes intuitive when APIs describe what customers achieve rather than the resources they use. Pricing a SUMMARIZE /document call is simpler than calculating database reads or bandwidth costs. The intent reflects the value delivered.

Implementation Realities for Intentful APIs

CRUD will always have a place within data persistence. The opportunity lies in layering intent on top of it. Behind a BOOK /flight call, the system still performs create, read, update, and delete operations across multiple services. The complexity remains internal, while the external API presents a clear business capability. Consumers can focus on outcomes rather than architecture.

Not every operation needs a new verb. The goal is to identify which intents are both common and meaningful enough to justify specialized methods. A small retailer may succeed with basic CRUD. A global platform like Amazon benefits from RECOMMEND, SUBSCRIBE, and REVIEW, since these represent distinct business capabilities that drive value.

The design pattern is straightforward: shape the API surface around business actions and orchestrate underlying CRUD operations to support them. This approach expresses the organization’s business model rather than its data model.

For AI systems, intent-based APIs accelerate understanding. An AI that calls SCHEDULE /meeting learns about coordination. One that uses SUMMARIZE /document learns about comprehension. The API becomes self-descriptive, teaching systems how the business operates.

The Road Ahead

As AI adoption increases, intentful APIs will define the next generation of intelligent systems. These APIs focus on achieving real outcomes rather than interpreting schemas. Intentful API design supports this evolution by uniting technical execution with business intent in a shared, human-centered language.

Industry movement already points in this direction. GraphQL improved flexibility in data retrieval. gRPC and Protocol Buffers introduced service-level clarity. Intentful APIs extend this momentum, helping teams define interactions that express true business value.

Organizations adopting this model gain measurable advantages: stronger analytics linking API usage to outcomes, easier onboarding for developers and AI systems, simpler pricing structures, and clearer collaboration between teams.

Across industry discussions, a recurring question arises: “How can we measure API ROI?” The answer often lies in the API’s structure. When APIs express business intent through methods such as BOOK, SCHEDULE, QUERY, SUMMARIZE, and RECOMMEND, performance maps directly to business results.

The tools and frameworks for intentful APIs exist today. What is needed is a shift in mindset, from APIs as technical endpoints to APIs as expressions of business purpose. CRUD continues to manage data, while intent elevates every interaction into measurable alignment.

Organizations that embrace this layered approach will see their APIs evolve into engines of intelligence, where each call represents meaningful action, shared understanding, and business growth.