A data flow diagram (DFD) is a graphical representation of how data moves through a system, showing the processes that transform it, the stores that hold it, and the external entities it flows to and from.
Data Flow Diagrams
Data Flow Diagrams (DFDs) are a graphical way of representing how data moves through a system — the processes that act on it, the data stores that hold it, and the external entities that send or receive it. Popularized in the late 1970s by structured-analysis pioneers like Larry Constantine, Tom DeMarco, and Ed Yourdon, a DFD deliberately ignores control flow and timing so it can focus on a single question: where does the data go?
- Four core symbols - Processes, data stores, external entities, and the labeled flows that connect them.
- Leveled decomposition - A high-level context diagram explodes into progressively more detailed levels without changing the modeling vocabulary.
- Control-flow agnostic - It shows what data moves and where, not the order of operations or the logic that decides.
- A shared analysis artifact - Readable by analysts, engineers, and non-technical stakeholders alike, which is why it survives as a design and documentation tool.
In API work, data flow diagrams are how teams reason about what an interface actually exposes before arguing about endpoints. Mapping the flows between a client, an API, its backing data stores, and third-party services surfaces trust boundaries, privacy-sensitive movements of personal data, and the seams where governance controls belong. They pair naturally with data modeling — one describes how information is shaped, the other describes how it travels — and increasingly they are the picture a security or agentic-governance review starts from.
