If you lead marketing, sales or growth at a B2B company and your website or onboarding flow is not converting, this guide is for you. It shows how to design for user experience with a UX design process that a small team or a single owner can run.
The short answer: define the problem, run user research, turn what you learn into user personas and a user flow, sketch wireframes and a prototype, run usability testing with real users, then launch and iterate. This step by step guide has seven steps, each with what done looks like.
|
UX Design Process The UX design process is the repeatable sequence of research, definition, ideation, prototyping, usability testing and iteration that a design team follows to turn user needs into a product people can use. |
You will finish with a tested design and a repeatable routine
At the end of these steps you will have a written problem statement, a set of user personas based on evidence, a user flow, a clickable prototype, a usability test report with ranked usability issues, and a routine for measuring and improving the design after launch.
UX design is a human-centered approach to creating digital products. It balances user satisfaction with business goals and technical feasibility, and it covers every interaction a person has with your company, from the first search result to the support reply.
If you want the full definition, our explainer on what UX design is covers the field and its disciplines. This guide stays on the work itself, with practical tips for each stage.
Designing for the user experience is also cheaper than redesigning later. A problem found in a sketch costs an afternoon, while the same problem found after launch costs a development sprint and lost conversions.
What to have in place before you start
You do not need a large UX design team to begin. You do need five things ready before step 1.
- A named owner. One person accountable for the design process, the schedule and the decisions. UX work stalls when everyone owns it.
- A business goal. The result the design must serve, such as more demo requests or fewer onboarding questions. Business goals keep design decisions honest.
- Access to real users. A short list of customers, prospects or colleagues who match your audience and will give direct feedback.
- Current data. Analytics, support tickets, sales call notes and survey answers that show where people struggle today.
- Simple tools. Pen and paper, a slide deck or a free design tool is enough for early sketches.
Design for user experience in 7 steps
1. Define the problem and the measure of success
Write down who the end user is, what they are trying to do and where they get stuck. State the problem in one sentence, such as "new customers cannot find the setup guide in their first week." Add the business goal and one measure that tells you the design worked.
Defining product goals first keeps the later design decisions tied to a purpose, and a clear problem statement tells the design phase where to focus.
Gather the stakeholders and the product managers who own the outcome, and agree on the scope before anyone draws a screen. Include product teams and developers early, since their view of technical limits saves rework later. New features and ideas can then be judged against the same problem.
Done looks like: a one-page brief with the problem, the audience, the business goals and the measure of success, signed off by the people who will use the result.
2. Run user research to learn how people behave
User research replaces opinions with evidence, so the process is built from real user needs rather than assumptions. A UX researcher, or whoever plays that role on your team, studies how people behave today and why. Start with what you already own: analytics for quantitative data, support tickets and call notes for pain points.
Then talk to actual users. Short interviews, observation of people doing a real task and a brief survey show where users struggle and what they expect, and they help you gather insights about how users interact with the current design.
They also let you identify pain points early. Ask open ended questions such as "walk me through the last time you tried to do this," and avoid leading questions that suggest an answer.
Record observations in plain notes, including body language and facial expressions when you watch someone use a screen. People often do something different from what they say.
Done looks like: notes from five to eight conversations and a short list of the behavior patterns you saw more than once.
3. Turn findings into user personas and a user flow
Group what you learned into two or three user personas. A persona is a short profile of a typical user that covers goals, context and frustrations. Base each one on research, not guesses, or it will only repeat your assumptions.
Next, map the user journey and the user flow. The journey shows how a person moves from first contact to their goal across channels. The user flow shows the exact screens and decisions along one path, such as signing up or booking a demo.
Mark every point where users hesitate or give up. These friction points are where design effort pays back first.
Done looks like: personas on one page each, one journey map, and a user flow for the most important task with the pain points circled.
4. Organize content with information architecture
Information architecture decides how content is grouped, labeled and linked so people can find what they need without help. Card sorting with real users is a quick way to learn how they group topics. Name menu items in the words customers use, not internal terms.
Keep the structure shallow. Every extra click or unclear label adds friction. Check the plan against your user flow: can a person complete the key task in a few clear moves?
If your site structure is part of the problem, our guide to an inbound-ready website shows how pages should connect to a visitor's next step.
Done looks like: a sitemap or navigation outline in which each label has been tested with a handful of users.
5. Sketch wireframes and build a prototype
Start with ideation. Run brainstorming sessions to produce several possible solutions to the problem, then pick the strongest to draw. Wireframes are low-fidelity layouts that show where content and actions sit, without color or imagery.
They are the skeletal framework of the design, and they cover your user interface at its simplest. Sketch several options for each key screen, then choose one. Cheap drawings are easy to change.
Apply visual hierarchy so the most important element on each screen is also the easiest to see: size, spacing and contrast should tell the eye where to go first. Reuse the same interaction patterns across screens so users learn the design once.
Then build a prototype that people can click through. A prototype does not need real code. It needs enough realism for users to complete realistic tasks and for you to see where they stumble.
Done looks like: wireframes for every screen in the user flow and a clickable prototype of the main task.
6. Run usability testing and share findings
Usability measures how easy a product is to use, and usability testing shows you where real users struggle before you spend money building. Good usability shows up as ease, efficiency, engagement, error prevention and enjoyment. Test early and often, and choose a method that fits your questions.
- Moderated testing: a facilitator watches, asks follow up questions and probes why. It gives rich insight, and moderated usability tests work in person or by video call.
- Unmoderated testing: participants complete tasks on their own devices without a facilitator. It is faster and cheaper, and it scales to more people.
- Remote testing: people take part from different locations, often in their own environment. A controlled environment in a lab gives you more control over conditions, while remote testing reflects real life.
To recruit participants, match your personas and offer a fair incentive. Write specific tasks such as "find the pricing page and request a quote," and avoid hints. Ask participants to think aloud.
Measure task success rates, time on task and error rates, and note quotes. The goal is to validate design decisions and identify friction points, not to defend the design.
Small rounds work well. Test with a handful of people, fix what you find, then test again. Each round tends to surface the same major problems, so repeated small tests beat one large study.
Afterward, share findings with the design team and other stakeholders. Rank the usability problems by severity and by how many users hit them, and recommend a fix for each. Keep the list of key issues short enough to act on.
Done looks like: a short report with ranked usability issues, supporting quotes, task success rates and agreed changes.
7. Launch, measure and iterate
Launch is the start of the next round, not the end. Work with developers on a clean hand-off, so the build matches the prototype. Fix the top usability issues before release, then watch real behavior. Compare the measure of success from step 1 with what happens in practice.
Track a small set of signals: task completion, conversion rate, the bounce rate on key pages and a satisfaction score such as the Net Promoter Score. Tracking post-launch metrics shows how the design performs and where to improve next.
Better usability helps conversion rates, because fewer barriers stand between a visit and an action. Pair these numbers with user feedback from forms and short interviews, because analytics show what happened and feedback explains why.
Continuous improvement keeps the product aligned with real user needs and behavior. Because this is an iterative process, schedule the next cycle now. Our guide to getting started with conversion rate optimization shows how to run those cycles with small experiments, and growth-driven design shows how to run them as a standing practice.
Done looks like: a dashboard with your success measure, a list of the next three improvements and a date for the next test.
Check your work against user experience design principles
Principles turn opinion into a checklist. Review each screen against these design principles before and after testing.
- User focus. Real user needs take priority over assumptions.
- Clarity. One main message and one obvious next action per screen.
- Hierarchy. The most important content is the easiest to see.
- Consistency. The same pattern for the same job, so users learn once. Consistency reduces cognitive load and builds confidence.
- Efficiency. Fewer steps, shorter forms and sensible defaults.
- Feedback. Every action gets a visible response.
- Error prevention. Make mistakes hard to make and easy to undo.
- Accessibility. The design works for people with different abilities, devices and conditions.
Accessibility is where many teams fall short. The WebAIM Million analysis of one million home pages for 2026 found that 95.9% of home pages had detected WCAG 2 failures, with low contrast text the most common.
The W3C's Web Content Accessibility Guidelines ask for a contrast ratio of at least 4.5:1 for normal text and, in version 2.2, a pointer target size of at least 24 by 24 CSS pixels. Build these checks into steps 5 and 6.
Speed is part of the experience too. Google's Web Vitals guidance says the largest content element should load within 2.5 seconds, pages should respond to input within 200 milliseconds, and layout shift should stay at 0.1 or less. Test these on a phone, because that is where most people meet your design.
Avoid the mistakes that waste a design round
- Skipping research. Designing from assumptions produces a polished answer to the wrong question.
- Testing too late. Problems found in code are the most expensive to fix. Test the prototype.
- Testing with colleagues only. People who know your product cannot show you what a stranger finds confusing.
- Treating the launch as the finish. Without measurement and a next cycle, a good design slowly drifts out of date.
- Confusing UX with decoration. Visual design matters, but a beautiful screen that hides the next step still fails. Our explainer covers how UX and UI design differ.
What you gain when this is done properly
Sales gets better conversations
When the site answers the questions buyers have, they arrive better informed and ready to talk about their situation. Sales spends less time correcting misunderstandings and more time on fit, price and timing.
Marketing gets pages that convert
A clear user flow, readable pages and fast load times turn the same traffic into more enquiries. Marketing also learns which messages users understand, which improves every campaign and landing page.
Customer success gets fewer avoidable questions
Clear navigation and onboarding screens reduce how-do-I questions. Customer success can then spend its time on adoption and renewals instead of repeating answers that the design should give.
UX design works best inside a connected go-to-market system
A well-designed page only creates revenue when what happens after the click is connected. Form submissions, page behavior and customer feedback need to land in one CRM record so marketing, sales and customer success act on the same picture.
Propello is a HubSpot partner. It designs and builds connected go-to-market (GTM) systems on HubSpot, as described on the about page. If your design work is producing traffic without clear pipeline, a GTM audit is the place to start.
/the-ultimate-guide-to-search-engine-marketing-share-1200x630-v2.png)