Conway’s law is an observation that the design of any system is significantly affected by the communications structure of the organization that develops it. The law is commonly associated with software development but is considered applicable to systems and organizations of all types.
Melvin Conway, a computer scientist and programmer, developed his theory in 1967 as the basis for a paper, "How do committees invent?" that he was submitting to the Harvard Business Review. Here's the original formation: "Any organization that designs a system (defined broadly) will produce a design whose structure is a copy of the organization's communication structure."
HBR rejected the paper on the grounds that Conway had not proved his thesis; the paper was published in April 1968 in Datamation, the leading IT magazine of the time. Fred Brooks cited the observation in his article, "The Mythical Man-Month" and referred to it as Conway's law.
The Harvard Business School subsequently conducted a study, "Exploring the Duality between Product and Organizational Architectures," to attempt to prove Conway's thesis. Researchers compared the codebases of multiple applications of the same type that had been created by loosely-coupled open source development teams and tightly-coupled teams. They found that the tightly-coupled teams tended to develop tightly-coupled, monolithic codebases while the loosely-coupled open source teams tended to create more modular, decomposed codebases. It has been observed, similarly, that if multiple teams are working on program modules and inter-team communication is poor, the interfaces of the program will reflect that fact.
See Dan Slimmon's presentation, Conway's Law: The Skeleton of DevOps: