An empty state is what an interface shows when there is no content to display: a new account with no data, a search returning no results, a filter combination matching nothing, an inbox that has been cleared, or a list awaiting its first entry. Empty states are among the most neglected screens in most products, because they are not the case designers are thinking about and are rarely present in the mock-ups used to review a design.
That neglect is costly because empty states appear at the most consequential moments. Every user of a new product begins in an empty state, and what they see there determines whether they understand the product's purpose and take the first action. A blank screen with a generic message offers nothing; a well-designed first-use state explains what belongs here, why it matters, and provides a single obvious way to begin. Onboarding effectiveness is largely determined by these screens rather than by any tour or tutorial layered on top.
Search results with no matches are the second high-value case. A bare "no results found" is a dead end, and the visitor's only remaining options are to guess again or leave. A useful version of the same screen explains what was searched, offers spelling corrections, relaxes the most restrictive filter, suggests related terms, shows popular alternatives, and provides a route to help. In commerce, zero-result searches represent visitors with clear intent and no path forward, which makes them among the highest-value recoverable failures on a site.
Empty states also serve as error and edge-case surfaces that reveal system problems. A list that is empty because of a failed request should not look identical to a list that is empty because there is genuinely nothing there, since the correct action differs entirely. Distinguishing between empty, loading, and failed states, and communicating each clearly, prevents users from concluding that a working product is broken or that a broken one is simply new.
Designing them well requires enumerating the cases explicitly during design rather than discovering them during implementation. For each list, table, search, and dashboard, the questions are: what does this look like with nothing, with one item, with a typical amount, with an extreme amount, while loading, and after a failure. Answering these systematically is quick, and it prevents the common outcome where the empty variants are improvised by a developer at the end of a sprint.
Empty states are also an opportunity to teach rather than merely to explain, and this is where they most often earn their place. A first-use state that shows an example of what a populated view looks like, using realistic sample content clearly marked as such, communicates the product's purpose faster than any description, and gives the new user a target to work toward. The same approach applies to states following deletion or completion, where showing what the space is for prevents the user from concluding that something has gone wrong. The constraint is honesty: sample content that is indistinguishable from real content causes confusion and, in products handling money or records, can cause serious errors.
Because empty states sit at the boundary between interface design and product comprehension, they are usually addressed as part of the flow rather than as isolated screens. In product design work they are specified alongside the populated states as standard practice, and in a UX audit they are a reliable source of findings, since almost every product has at least one empty state that was never designed and now silently blocks a proportion of users at their first encounter with it.