User flow

A user flow follows one goal, such as booking an appointment or buying a gift card, and shows every step: the page someone lands on, what they click, what the system asks, where they might branch off and what happens when something goes wrong. It is narrower than a customer journey, which covers the whole relationship from first hearing about a business to coming back, including offline moments.

A flow is also different from a site map. The site map shows which pages exist; the flow shows the path a person actually walks through them.

How it works

A flow usually starts with an entry point, such as a Google search, an ad or an email link, because the starting page shapes everything after it. Each screen or step is drawn as a box, decisions as branches and the end as a clear outcome: a booking confirmed, a message sent. Good flows include the unhappy paths too, such as a sold-out date, a declined card or a missing field.

Designers sketch flows before drawing wireframes, so the team agrees on the steps before discussing layouts, then test the flows with prototypes.

You can check a flow yourself by walking it from the real entry point. Open the ad, email or search result your customers use, follow it on your phone and count every screen, field and decision until the task is done. Then try the unhappy paths: enter a wrong phone number, go back a step, close the page and return. Write down each point where you had to think, wait or retype something. Those are the points where real customers drop out.

Example

A travel agency maps the flow for requesting a holiday quote. The current flow takes a visitor from an ad to the home page, then to a destinations page, then to a general contact form with eleven questions. The new flow sends the ad straight to a landing page for that destination, asks four questions and confirms by email with the next step. Two screens and seven questions disappear.

The team then draws the unhappy paths. If a visitor picks dates when the hotel is full, the old form accepted the request and someone replied two days later with bad news; the new flow shows that those dates are unavailable and offers the nearest alternatives. If a phone number is in the wrong format, the form explains the expected format next to the field instead of clearing everything. Each branch is agreed on the diagram before anyone builds it.

Why it matters for your business

Every extra step in a flow is a place where a customer can give up, so mapping the flow is the quickest way to see where enquiries and sales leak. It also catches dead ends, such as an error with no way back, before they are built. A flow takes an hour or two for a simple task, which is small compared with rebuilding a form that nobody finishes. Map the flows that earn money first.

Common mistakes

• Mapping only the ideal path and ignoring errors, cancellations and people who arrive halfway through.
• Assuming every visitor starts on the home page when many land on an inner page from search or ads.
• Adding steps to collect data the business would like to have but does not need yet.
• Designing a flow on a desktop screen without ever walking through it on a phone.

Questions people ask

What is the difference between a user flow and a customer journey?

A user flow covers the steps to complete one task inside a website or app. A customer journey covers the wider relationship across channels and time, from discovering a business to buying, getting support and returning. Journeys are useful for strategy; flows are what designers and developers build from.

What tools are used to make user flows?

Anything that draws boxes and arrows will do: a whiteboard, paper, a diagramming tool or a design tool such as Figma. The tool matters far less than including every step, decision and error state, and sharing the flow so that everyone working on the project agrees on it.

How do I know where users drop out of a flow?

Set up a funnel in your analytics for the key steps, for example landing page, form started and form submitted, and see where the numbers fall sharply. Session recordings and usability tests then show why people stop at that point, which the numbers alone rarely explain.