How many steps in a flow are just enough?
The question is not about how many steps, but whether each step has a reason for existence.
There is no fixed number of steps for every flow. The reasonable number of steps is just enough for the customer to complete the task without having to fill out or click on unnecessary items. Each step needs a clear purpose, and each time information is requested, it should answer the question: is this step really necessary?
- Simple tasks, clients only need a few pieces of information to complete.
- requires gathering a lot of information, so steps need to be divided to avoid overload
- irreversible or risky actions requiring confirmation steps
- counted steps as the only metric while ignoring the quality of each step
- stuffing all fields onto one page to reduce steps
When clients ask Sinh Vũ, "How many steps are enough in this flow?", the real answer does not start with a number. It begins with a reverse question: what does each step in the customer service flow address? Keep the steps that answer that question. The steps that do not answer it are wasting the client's time.
Why is there no golden number?
Baymard Institute measures various e-commerce payment flows in the US and Europe (updated to around 2023-2024) and finds that many flows average about 11 input fields while most only need around 8. A significant percentage of customers abandon their carts due to overly long or complex flows. This is a valuable reference number, but there are two limitations to remember: it measures Western e-commerce checkouts, and it measures input fields, not steps. Directly applying that number to a contact form or service consultation page in Vietnam is comparing the wrong subjects.
The context of the task determines the reasonable length. Scheduling a house viewing requires less information than signing a long-term lease. Signing up for a newsletter requires fewer steps than opening a business account. Simple flows need fewer steps, while critical flows can accept additional careful steps because customers also want to be sure before confirming.
Factors that are truly worth considering
- Purpose of each step: each step should only accomplish one task. Any step without a clear purpose is a candidate for removal or consolidation.
- Number of input fields per screen: Each unnecessary field is an opportunity for customers to drop off. Asking for information that is not needed wastes customer attention and dilutes the purpose of that step.
- Perceived length, not just actual length: A 6-step flow with a clear progress indicator often feels lighter than a 3-step flow where customers are unaware of how much further they have to go. The progress signal does not shorten the flow but makes it more comfortable.
- Linearity: a long flow is much more pleasant when it follows a straight path, without making customers jump back and forth or return to previous steps to correct something that was asked later.
When to combine, when to break down
Breaking down into smaller steps when asking for a lot of information and cramming it into one page will overwhelm customers. Breaking it down helps each screen focus on one decision group, reducing distraction. It is also advisable to add a confirmation step when actions are difficult to undo or have clear risks.
Streamlining is not an end in itself. Only remove unnecessary steps or add elements that lighten the process, aiming to reduce overall interaction costs rather than cutting indiscriminately.
Nielsen Norman Group, 7 Ways to Analyze a Customer-Journey Map
Common errors when designing a flow
- Counting steps as the only metric: "Reducing to 3 steps" sounds concise, but if each step includes 10 items, customers will still drop off as usual. The number of steps and the quality of each step must go hand in hand.
- Stuffing everything onto one page to "reduce steps": A longer page with many steps is not optimal; it just burdens one place.
- Ask for internal information that does not serve the client: If an organization exists because the internal team wants to know, not because the client needs to complete a task, it should be reconsidered. Clients are not obligated to fill out reports for the business.
- Long flows without progress signals: if customers do not see the destination, they will become discouraged and leave, even if the flow is not actually that long. This is a small cost to implement but significantly affects perception.
Sinh Vũ's viewpoint
Sinh Vũ does not begin with a question of how many steps, but starts by reviewing everything you need to fill out and everything you need to click. Any excess is cut, anything cumbersome is broken down, and anything unclear is clarified before touching the interface. The number of steps is the result of that process, not the starting point.
Because the flow is built at the experience stage before the interface stage, the final page should lead to action rather than making clients jump through hoops. You can start by listing each step in the current flow, then for each step ask: if this step is removed, will the client lose anything? The answer will indicate which steps to keep and which are unnecessary burdens.
Decision checklist
Topic: How many steps in a flow are just right. Sinh Vũ guide, sinhvu.com
Select each item you find appropriate, then print or save as PDF to take with you.
If you have marked most of the signs above, this is the time to discuss in more detail. Sinh Vũ can help you review and propose a direction.
References
Baymard Institute, E-Commerce Checkout Usability Research. Baymard Institute, Why Your Checkout Process Should Be Completely Linear. Nielsen Norman Group, 7 Ways to Analyze a Customer-Journey Map.
Frequently asked questions
Is a 3-step flow better than a 7-step flow?
Not necessarily. A three-step flow with too many input fields can be heavier than a well-structured seven-step flow. The right criterion is the total interaction cost, not the number of steps on the progress bar.
Should we add a final confirmation step in the flow?
You should add confirmation when actions are difficult to reverse or have clear risks, such as confirming an order or sending a contract. For simple and easily corrected tasks, an additional confirmation step often just slows down the customer without providing any protective value.
Does a long flow necessarily need to be shortened?
Not necessarily. Shortening is not an end in itself. If the process is lengthy because significant decisions require substantial real information, then keep it as is and focus on making each step lighter, more linear, and with clear progress signals so clients do not get discouraged.