A product that needs its users to define it has not decided what it is
A product should be clear about its job on first use. Tools that let users invent the job are the exception, and they need a different kind of patience.
If you need the user to tell you what you’re selling, then you don’t know what you’re selling, and it’s probably not going to be a good experience.
Google announced Wave in May 2009 with a long demonstration and great enthusiasm. It was part email, part chat and part shared document, and it could be used for almost anything. That turned out to be the problem. People who got invitations logged in, looked around and could not work out what it was for. Google stopped developing it as a standalone product in August 2010, saying that too few people had adopted it.
Marissa Mayer, who led Google’s search products and user experience for most of the past decade, usually introduces this idea as a theory held at the company. It explains Wave’s trouble well. If a product’s purpose only becomes clear once users decide what to do with it, the first experience will be confusing, and most people will leave before they decide anything.
The same principle shows in Google’s search page, which Mayer oversaw for years. It does one thing, says so with a single box, and needs no instructions. The product knows what it is selling, and the user never has to guess.
The line is less reliable for tools meant to be open-ended. Twitter launched in 2006 asking users what they were doing, and people used it for something quite different: news, conversation and links. Users invented the reply convention and the hashtag before the company built either into the product. In a real sense, Twitter’s users told it what it was selling, and the service did well by listening. Spreadsheets, email and the web itself also grew by letting users find their own purposes.
The difference is that those products were simple enough to try without knowing their purpose. Posting a short message takes seconds. Wave asked for a great deal of learning before offering anything back.
The practical reading is that a product should be clear about at least one job on first use, even if users later find others. Knowing what is being sold does not rule out users discovering more. It means the first visit does not depend on them doing so.