When is GraphQL particularly advantageous compared to REST?

When is GraphQL particularly advantageous compared to REST?

• A. When clients need different data shapes and over-fetching must be avoided.
• B. When clients always require the full dataset.
• C. When there is no need for a single endpoint.
• D. When you want binary serialization.

Answer: (A)

The key idea is that GraphQL excels when clients need different data shapes and you want to avoid fetching more data than necessary. With GraphQL, a client specifies exactly which fields it wants, including nested relationships, so the response includes only what that client needs. This is especially powerful when you have multiple clients (mobile, web, partners) that each require different subsets of the same underlying data, because you get precisely what you need in a single request rather than making multiple REST calls or pulling a fixed payload. A single flexible endpoint is another practical benefit: instead of calling various REST endpoints for related resources, you can retrieve all the necessary data in one query. This reduces round trips and can simplify client logic. However, note that JSON is the common transport in GraphQL, so it isn’t inherently tied to binary serialization. If a system truly always serves the full dataset, GraphQL still works, but the advantage isn’t as pronounced since you’re not tailoring responses. Also, REST can use multiple endpoints when that fits a given design, and if you really don’t need a single entry point, the single-endpoint benefit of GraphQL is less compelling.

The key idea is that GraphQL excels when clients need different data shapes and you want to avoid fetching more data than necessary. With GraphQL, a client specifies exactly which fields it wants, including nested relationships, so the response includes only what that client needs. This is especially powerful when you have multiple clients (mobile, web, partners) that each require different subsets of the same underlying data, because you get precisely what you need in a single request rather than making multiple REST calls or pulling a fixed payload.

A single flexible endpoint is another practical benefit: instead of calling various REST endpoints for related resources, you can retrieve all the necessary data in one query. This reduces round trips and can simplify client logic. However, note that JSON is the common transport in GraphQL, so it isn’t inherently tied to binary serialization.

If a system truly always serves the full dataset, GraphQL still works, but the advantage isn’t as pronounced since you’re not tailoring responses. Also, REST can use multiple endpoints when that fits a given design, and if you really don’t need a single entry point, the single-endpoint benefit of GraphQL is less compelling.