Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Scan for outdated or missing drivers - takes under a minute3Repair Windows errors before they cause bigger problemsiTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
A website redesign should start with evidence about what visitors are trying to do and where they get stuck—not with a popular color palette, layout, or animation. Treat “the site looks dated” as a hypothesis: find out whether appearance is causing a problem with a real user task, then test any proposed change with the people who use the site.
Why user problems should come before visual trends
A redesign is a costly way to change a website, so its purpose should be more specific than making the site look current. Start by identifying who uses it, what they need to accomplish, and what makes that task difficult. The UK Government’s Government Design Principles put it plainly: “Service design starts with identifying user needs.” The guidance calls for decisions informed by data, testing, iteration, inclusion, and the context in which people use a service.
A visual trend can suggest a design possibility, but it does not show that visitors need that change. A new navigation style, for example, is not justified merely because it is fashionable. It becomes a reasonable option if evidence shows that people cannot find important information or are taking unnecessary steps—and testing shows the proposed navigation helps them.
Free tools Windows power users keep installed
One-click scans. No signup required.
What to find out before redesigning
Build a picture of the task and its context before choosing a solution. Combine what existing data can show with direct conversations and observation; no single source is likely to explain the whole problem.
#1 Best Overall
- HTML CSS Design and Build Web Sites
- Comes with secure packaging
- It can be a gift option
- Who is using the site? Identify actual and likely users, including people who struggle with the current experience and people with relevant access needs.
- What are they trying to do? Describe the task and the outcome they want, rather than starting with a page, feature, or aesthetic.
- Where is the friction? Look for confusion, errors, repeated attempts, unnecessary effort, or a need to ask for help.
- What context shapes the task? Consider the devices people use and any relevant offline channels or service steps that affect their experience.
Audit what the site already tells you
Review analytics, onsite search logs, prior research, and relevant service or support information. These sources can reveal patterns worth investigating—for example, what visitors search for or where a task may be going wrong. They do not necessarily explain why people behave that way, so use them to identify questions rather than to prescribe a layout.
Separate observations from assumptions. The Government Digital Service advises: “Treat any opinions or suggestions that do not come from users as assumptions that have to be proven by doing research.” A stakeholder’s concern may point to a genuine issue, but it should be checked with evidence from the people the site serves.
Interview and observe users doing the task
Ask actual or likely users about their goals, then observe them attempting relevant tasks. Seeing where someone hesitates, misunderstands a label, makes an error, or switches to another channel can reveal problems that an analytics report alone cannot explain. Include people who experience barriers, and account for disabled people and assistive-technology users where relevant.
Rank #2
Do not recruit only the easiest people to reach. The GOV.UK introduction to user research says research helps teams understand the problems to solve, what to build, and whether a service works well for users. Its guidance is written for government services; other organizations should adapt recruitment, legal obligations, measures, and operating constraints to their sector and location.
Write the need without choosing the design
Describe the user’s problem and desired outcome in language they would recognize. A useful prompt is: “I need to [do something] so that [achieve an outcome].” Keep a proposed interface, channel, or feature out of the need statement.
For example, a need might be to receive a reminder in time to complete a task. Deciding immediately that the reminder must be an email—or a particular page element—confuses the need with one possible solution. The GOV.UK guidance on learning about users and their needs uses this distinction to help teams avoid locking into a design before understanding the problem.
Users’ requests are evidence to interpret, not instructions that every proposed feature must be built. Check what outcome the request represents, whether others share the need, and whether a proposed solution actually helps.
How to test a redesign idea
Once the need is clear, compare possible designs by whether people can complete the task and reach the intended outcome—not just by which option they say they like. The GOV.UK guidance on making a service simple to use recommends frequent testing of design ideas and usability, including on devices that reflect how users access the service.
- Choose a task and success outcome. State what a user should be able to accomplish. Avoid making “likes the new design” the main measure.
- Prototype the alternatives. Explore more than one way to address the need, including the existing design if it provides a useful comparison.
- Test with relevant users. Ask actual or potential users to try the task. Include people with different needs and abilities, and use devices that reflect their behavior.
- Observe what happens. Note whether people reach the intended result, where they hesitate or make mistakes, and whether they need assistance.
- Refine based on evidence. Keep, change, or remove elements according to what the sessions and other evidence show. Include accessibility checks and testing with disabled participants and assistive technology during development.
- Continue after launch. Review live behavior and service information, and test again as the site and users’ needs evolve.
The GOV.UK guidance and tools for digital accessibility provides further direction on accessibility. An accessible design should be considered throughout development, not treated as a visual polish step at the end.
Rank #4
How to compare a visual refresh with a problem-led change
There is no universal numeric benchmark in the cited guidance for deciding whether a redesign has succeeded. Set measures that fit the task and compare options using consistent criteria:
- Task outcome: Can users complete the task and reach the result they need?
- Friction: What errors, delays, confusion, or requests for assistance do you observe?
- Inclusion: Does the change work for people with different access needs and abilities?
- Real-world use: Does it remain usable on the devices people actually use?
- Evidence over time: Do research and live service data indicate improvement after the change?
A visual refresh may be part of the right answer if it addresses a demonstrated obstacle. But a preference poll or a trend alone cannot establish that it improves task completion. The sources do not show that user-led redesigns always raise sales, conversion, or satisfaction by a particular amount; judge the result against the user outcomes the project set out to improve.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Make research part of the redesign, not a one-off gate
Testing once before launch is not enough if the site changes, its audience shifts, or live behavior reveals new problems. Keep a regular cycle of research, design, and refinement, and make sure the team can observe users rather than relying only on reports about them.
For government service teams, the GOV.UK user-research introduction recommends that user researchers typically work with a team at least 3 days a week and that teams run research sessions at least every 2 weeks. It also says team members should ideally observe at least 2 hours of user research every 6 weeks. These are operational recommendations on that guidance page, not universal requirements or research findings; teams elsewhere should choose a cadence suited to their work and capacity.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
