What are the different types of APIs?
● System APIs
These APIs provide a means of accessing core systems
of record and expose data from these systems, often in
a canonical format. They abstract the complexity of
underlying systems (such as databases, ERP, CRM) and
provide a consistent interface for data access.
● Process APIs
These APIs are responsible for orchestrating and
shaping data across multiple systems. They combine
and process data from various sources (often via
System APIs) to fulfill specific business processes or
logic, enabling reuse and agility.
● Experience APIs
These APIs are tailored to the needs of specific
consumers, such as mobile apps, web applications, or
external partners. They present data in the most
convenient format for the end user or channel, often
aggregating and transforming data from Process APIs.
Comparison
Parameter System APIs Process APIs Experience APIs
Complexity Low to moderate; focused on single systems Moderate to high; involves orchestration and business logic Low to moderate; focused on data presentation
Maintainability High; isolated from consumer changes Moderate; business logic may evolve High; changes are mostly UI/consumer-driven
Reusability High; generic and reusable across processes High; reusable business logic for multiple experiences Low to moderate; tailored for specific channels
Coarseness Fine-grained; exposes specific data entities Medium; aggregates and processes data Coarse-grained; aggregates and formats for end users
Stability High; changes infrequent, tied to core systems Moderate; changes with business processes Moderate; changes with consumer needs
Security Very high; often exposes sensitive data High; may enforce business-level security Moderate; focuses on channel-specific security
Flexibility Low; tightly coupled to backend systems High; adapts to business logic changes Very high; adapts quickly to UI/channel changes
Performance High; direct access to systems Moderate; may involve multiple data sources Variable; depends on data aggregation and formatting
Change Frequency Low; backend changes are rare Moderate; business logic changes periodically High; frequent updates for new experiences
Consumer Type Internal IT, integration teams Internal business processes, middleware End users, mobile/web apps, partners
Typical Use Case Expose database, ERP, CRM data Order processing, customer onboarding Mobile app dashboard, partner portal
Let’s take use case by use case?
[Link] have only one producer (e.g. only one payment provider)?
2. We have multiple producer for same domain (e.g. two or more payment provider).
3. We have a producer and a consumer which have connectors already built for some industry
standard APIs but the producer does not conforms to it.
3. We have multiple producer for same domain and multiple consumers.
4. Anything else!!!!!!
Pros & Cons - All Passthrough API
API Layering : Pros & Cons
Pragmatic API Layering Framework
Decision Matrix
Factor Passthrough Approach Layered Approach
Reuse potential Low High
Integration complexity Simple Complex
Time sensitivity Urgent Strategic
Business criticality Low High
Change frequency Rare Frequent
Implementation Guidelines
Classify by Business Impact
● Critical business capabilities → Full API layering
● Low-impact, one-off integrations → Passthrough acceptable
Start Simple, Layer Later
● Begin with minimal viable API design
● Refactor into layers when reuse patterns emerge
● Document technical debt for future refactoring
Hybrid Patterns
● "System-Experience" pattern for simple domains (skip Process layer)
● "Process-Experience" pattern when systems are stable
● Full layering for complex, high-change domains
Governance Checkpoints
● Review APIs quarterly for refactoring candidates
● Identify passthrough APIs with increasing usage
● Upgrade high-value passthrough APIs to layered approach
Practical Metrics
● API reuse rate (calls per month)
● Change request frequency
● Integration delivery time
● Maintenance effort
Closing Thoughts
Closing Thoughts
Thank You