underfetching

Underfetching occurs when a data request returns insufficient information to render a complete view, forcing additional requests to acquire the missing data. It is a counterpart to overfetching; while overfetching returns more data than needed, underfetching requires extra round trips to assemble the full result. Underfetching is commonly discussed in the context of web APIs, microservices, and mobile applications.

Causes include multi-layered APIs that split data across separate endpoints, service boundaries in a microservice architecture, and conservative query designs that prioritize small payloads over completeness. It can also arise when caches omit fields that are only available from downstream services, or when frontend components fail to express all required fields in a single request. In systems designed to aggregate data, underfetching can occur if data is distributed across services and not composed server-side.

Impacts include increased latency due to multiple requests, a higher risk of partial or delayed rendering, and greater complexity in client code. It can degrade user experience on slow networks and lead to higher server load and battery usage on mobile devices.

Mitigation strategies involve designing APIs for data composition or embedding, using aggregation services to fetch and merge data on the server, and employing query-capable interfaces such as GraphQL to allow clients to specify needed fields. Frontend patterns like progressive rendering, data prefetching, and efficient caching can also reduce visible latency. Instrumentation and monitoring help detect underfetching by analyzing per-view latency and the number of required round trips.