What to learn in the two weeks before a website redesign
Redesigns built on opinions tend to repeat old mistakes in new colors. A focused two-week research plan that grounds decisions in evidence.
Most website redesigns start with a feeling: the site looks dated, a competitor launched something new, or leadership is tired of the current look. Those are valid triggers, but they are poor foundations. Redesigns driven by opinion tend to repeat the old site's problems in a new visual style, because nobody stopped to understand what those problems were. A short, focused research phase changes that. Two weeks is enough to replace most assumptions with evidence, and it usually makes the rest of the project faster, not slower.
Week one: understand the business and the data
Stakeholder interviews
Interview six to ten people inside the organization, one at a time: leadership, marketing, sales, customer service and whoever maintains the site. Ask what the site needs to achieve, how success is measured, what they hear from customers, and what frustrates them about the current site. Individual interviews surface disagreements that group workshops tend to smooth over, and those disagreements are exactly what the redesign needs to resolve.
Analytics and search data
Review twelve months of analytics. Look at where visitors land, which pages lead to conversions, where people leave, what they search for on the site and how behavior differs between mobile and desktop. Search console data shows which queries bring people in and which pages carry organic traffic, which matters enormously for protecting visibility during the redesign.
Content audit
Export every page with its traffic and conversions. On most sites we audit, a small fraction of pages drives the vast majority of meaningful traffic, while hundreds of pages receive almost no visits. Mark each page to keep, improve, merge or remove. This audit shapes the new information architecture and often reduces the scope of content migration significantly.
Week two: understand the users
User interviews
Speak with six to eight real users or customers, recruited from recent leads, customers and, if possible, people who considered buying but did not. Ask about the problem that brought them to you, how they compared options, what information they needed and where they looked for it. Avoid asking what they want from a website. Ask about their decisions and let the implications for the site emerge.
Usability sessions on the current site
Run five short sessions on the existing site with the most important tasks, such as understanding what you offer, finding pricing or making contact. Watching people struggle with the current site is the fastest way to identify what must change, and it gives the team a baseline to compare the redesign against.
Competitive review
Review four or five competitor or peer sites with a clear lens: how they explain the offer, structure information and handle the key tasks. The point is not to copy. It is to understand what your audience is already used to and where you can be clearer.
A redesign built on opinion repeats the old site's problems in new colors. Two weeks of research is usually enough to break that cycle.
Keep the research lightweight and visible. Share short notes after each interview in a common channel, invite stakeholders to observe a session or two, and avoid disappearing for two weeks before revealing conclusions. When people have watched the evidence accumulate, they are far more likely to accept what it implies, even when it contradicts their original assumptions.
Turn findings into a brief
Research only matters if it changes decisions. At the end of the two weeks, synthesize findings into a short document the whole team can use. We keep it to a few pages:
- Goals and measures: what the redesign must achieve, with specific metrics and current baselines.
- Audiences and their key tasks: who the site is for and what they come to do, in priority order.
- Top problems: the issues found in analytics and testing, with evidence for each.
- Content decisions: what stays, what merges, what goes and what is missing.
- Constraints: technical, legal, brand and SEO considerations the design must respect.
Present the brief in a workshop with stakeholders and get explicit agreement. From here, design debates can be settled by referring back to the evidence rather than by seniority.
What this changes in practice
The benefits show up throughout the project. Information architecture is based on how users think, not on the organization chart. Content work is scoped realistically. Design reviews are shorter because the goals are agreed. And launch has a baseline, so the team can show what improved.
In one recent project for a professional services firm, the research phase revealed that most qualified leads came from three service pages that were buried two levels deep, while the homepage, which had consumed most of the internal debate, played a minor role. The redesign brought those pages forward and simplified the path to contact, and qualified inquiries rose substantially in the first quarter after launch.
Research also protects what already works. A redesign that ignores existing search visibility can lose a large share of organic traffic overnight. Identifying high-value pages and URLs early means the new structure can preserve them, and it gives developers a clear redirect plan before launch rather than after traffic has already dropped.
Our UX research service runs this discovery phase as a fixed-scope engagement, often followed by wireframing and prototyping and visual design. When organic search matters, we bring in our SEO-safe migrations team from the start so the new structure protects existing visibility.
Start your redesign with evidence
If a redesign is on your roadmap, a short research phase is the best investment you can make in it. Ask us to scope your discovery and receive a fixed-price quote within 24 hours.
