Website Redesign or Small Improvements: How to Decide
For most business websites the right first step is a series of targeted improvements, not a full redesign. A redesign is justified when the evidence shows that the foundation itself is the problem – technology that blocks changes, a structure that no longer fits the business, a poor experience on phones – or when years of small fixes have stopped making a difference. "It looks old" is a reason to take a closer look, not a reason on its own.
Why "let's redesign" is often the wrong first move
Nielsen Norman Group, one of the best-known user experience research firms, advises setting clear goals, making incremental changes towards them, and only resorting to a complete redesign "if the data tells you to do so". Their reasoning is practical:
- Redesigns are often driven by boredom or a feeling that the site looks dated, while customers are task-oriented and "care much more about a usable website than a pretty one".
- Returning visitors have learned where things are. A radical change forces them to learn the site again, and people resist change even when the new design is better.
- "Radical changes have a higher chance of inadvertently breaking something critical for users."
- Big projects are expensive to agree on internally and can stall in long debates.
There is also a search risk. If a redesign changes page addresses, Google has to crawl the old and new URLs again, and its documentation warns that rankings can fluctuate while that happens. Google recommends permanent (301 or 308) redirects from each old URL to its new counterpart, keeping them for at least a year, and avoiding a site move at the same time as other major changes such as a redesign – one change at a time.
Signs that small improvements are enough
- People find the site, but specific pages lose them: an unclear service page, a long form, a phone number that is hard to find.
- Content is out of date – old prices, services you no longer offer, missing answers to questions customers ask by phone.
- Pages are slow because of large images or a few heavy scripts.
- Google Search Console shows fixable issues: pages not indexed, duplicate titles, broken links.
- The site works on phones, even if it is not the newest look.
- The admin panel lets you change what you need.
Each of these can usually be fixed in days or weeks, measured, and followed by the next improvement. Ten small, measured improvements often teach you more than one big launch.
Signs you need a redesign or a rebuild
NN/g lists situations where a larger overhaul may be justified: incremental changes have reached diminishing returns; outdated technology prevents critical changes; the architecture has become a tangled mess after years of patchwork; site-wide conversion is severely low and small fixes cannot reach an acceptable level; or the site is far behind competitors. In everyday terms, watch for:
- A weak mobile version. Google uses the mobile version of a site's content for indexing and ranking, and recommends responsive design as the easiest pattern to implement and maintain. If the mobile version hides content or is a separate, poorer site, that is a structural problem.
- Technology you cannot update. The site depends on an old CMS, abandoned plugins or an unsupported server language version, and every change risks breaking something else.
- A structure that no longer fits. New services, new markets or new languages do not fit the menu and the page types.
- Content you cannot manage. Every small change needs a developer.
- Speed problems built into the template. Core Web Vitals stay poor even after you optimise images, because the layout itself loads too much. Google's thresholds for "good" are a Largest Contentful Paint within 2.5 seconds, Interaction to Next Paint of 200 milliseconds or less and Cumulative Layout Shift of 0.1 or less.
A simple way to decide
- Write down the goal. More enquiries, more online orders, fewer support calls – with a number you can measure.
- Collect the data you already have. Analytics (which pages bring visitors and where they leave), Search Console (search queries, indexing problems), Core Web Vitals reports, and the questions customers ask your staff.
- Watch real users. Ask a few customers to complete key tasks on the current site while you observe. Jakob Nielsen's well-known analysis found that a test with five users uncovers about 85% of usability problems, and recommends several small rounds instead of one large study. NN/g also calls spending a day or two on such a study "invaluable" before committing to a redesign.
- Sort the problems. Mark each one as content, design, technical or structural.
- Estimate the fixes. If most problems are about content and individual pages, improve. If most are structural or technical, plan a redesign.
| Problem | Usually a small fix | Usually needs a redesign |
|---|---|---|
| Visitors leave a key page | Rewrite the page, add a clear call to action | Many pages fail in the same way because of the template |
| Slow pages | Compress images, remove unused scripts | The theme or page builder loads too much on every page |
| Information is hard to find | Rename menu items, add internal links | The structure no longer matches your services |
| Poor on phones | Fix specific elements | The site is not responsive at all |
| Hard to update | Add a field or page type to the admin | No usable admin, or an outdated CMS |
If you redesign, protect what already works
- Keep the URLs of pages that already rank wherever possible, and redirect the rest one-to-one with 301 redirects.
- Keep the content that brings visitors – improve it, do not cut it to fit a design.
- Avoid combining a domain change with a full redesign.
- Consider a phased approach: rebuild one section at a time and measure the effect.
- Watch Search Console and analytics closely for several weeks after launch.
When we split our own site into seoandweb.bg and seoandweb.eu, we listed every old address, set up around 45 permanent (301) redirects and connected the two language versions with hreflang. The checklist we used is in how to redesign without losing Google rankings, and our company website checklist helps you review what the current site is missing.
Not sure which side you are on? We can look at your current site, its data and its code, and tell you which problems are worth fixing first and whether a new website is really justified. Contact us for a review.