Guide Redesign and launch
How often should a website be redesigned? Signs it is time, and signs it is not
By Studio Swing
There is no fixed interval. None of the sources we read gives one, and a calendar is a poor trigger: a site can run well for years, and another can need a rebuild in two. What matters is whether the site has a structural problem. Your website needs a redesign when it cannot hold what the business now offers, the people who run it cannot edit it, it fails on phones, it has been patched until nobody can find anything, or your own numbers show a site-wide problem that small fixes have not touched. It usually does not need one when it merely looks old, when traffic dropped for reasons outside the design, or when it brings few enquiries because a page says too little. Each of those has a cheaper fix, and a good studio will tell you which one you have.
On this page
- 01Why do the signs get mixed up
- 02Which signs point to a real redesign
- 03Which signs look like a redesign but are not
- 04What does a redesign cost you even when it is the right call
- 05What are the options between "do nothing" and "start over"
- 06How often should a website be redesigned
- 07What this guide will not tell you
- 08Checklist: is it the site
- 09Where to go from here
- 10Sources
Why do the signs get mixed up?
This guide lists the signs on both sides, with one test you can run for each before you spend anything.
Say you run a regional accounting firm. The site is seven years old. One partner says it looks dated. Your office manager says phone enquiries have slipped since spring. A new hire says the contact form sometimes fails to send.
That is three different problems. "Looks dated" is about appearance. Slipping enquiries could be demand, a weak page, or a search change. A failing form is a bug, and a bug is not a reason to replace a site. If you treat them as one problem, the answer is always "redesign," and you pay for a new site that may fix one of the three.
The Nielsen Norman Group, a usability research firm, puts the principle plainly: "Before you throw out the old and bring in the new, make sure you have solid evidence that doing so is necessary to achieve user-centered goals."
The signs below are a way of collecting that evidence.
Which signs point to a real redesign?
1. The structure cannot hold what you do now.
The business added services, a second location or a shop, and the site has nowhere sensible to put them. Pages were bolted on, and the menu has grown a "More" dropdown. Test: write a list of everything you sell or do today, and check each item has its own page and a findable place in the menu. If half of them are crammed into one page, the structure is the problem.
NN/g describes a site that, after several years of patchwork, has become "incohesive", and lists it among the few situations where a major overhaul is justified.
2. The people who run the site cannot change it.
A price changes and it takes a week and an invoice to update. Nobody on your team can add a page or fix a typo. Test: ask the person who is meant to maintain the site to change a phone number and publish a new page, without a developer.
If they cannot, the platform is working against you. Sometimes a different editing setup on the same site is enough, and sometimes it means a new platform, which is the case where a redesign makes sense.
3. The site works badly on phones.
Google uses the mobile version of a page for indexing and ranking, and its guidance is to "make sure that your mobile site contains the same content as your desktop site." Test: open your ten most important pages on a real phone. Can you read the text without zooming, tap each button without hitting its neighbor, and find the same information a desktop visitor sees? Then check real-visitor speed data instead of a single lab test: Google's Core Web Vitals targets are the main content appearing within 2.5 seconds, a response to a tap within 200 milliseconds, and layout shift of 0.1 or less, measured at the 75th percentile of visits on mobile and desktop separately. A page that misses them because of one oversized image is a fix. In our view, a template that is slow everywhere is a rebuild.
4. People cannot use it with a keyboard or a screen reader.
The Web Content Accessibility Guidelines (WCAG) 2.2 is the current W3C recommendation, with levels A, AA and AAA, and Level AA is the level we work to. Test: put the mouse aside and try to reach and use your menu, a form and a checkout with the Tab key alone. If you get trapped, or cannot see where you are, the template has a deep problem. Surface problems such as low color contrast or missing image descriptions can often be repaired in place.
5. Years of patches have left the site incoherent.
Different pages look like they belong to different companies, buttons behave differently from page to page, and each new request breaks something else. Test: ask your developer what it would take to change the menu on every page. If the answer is "edit every page by hand," the cost of each future change keeps rising.
6. Your own numbers show a site-wide problem, and small fixes have not helped.
NN/g lists "diminishing returns" among its conditions for a radical redesign: you have tried gradual fixes for years and there is not much more to improve. It also lists a conversion problem too tangled for incremental fixes, and benchmarking that shows customers going to competitors with better sites. Test: have you actually made targeted changes to the pages that matter, and measured them? If not, you have not yet earned this sign.
Which signs look like a redesign but are not?
| What you notice | Why it misleads | What to check first |
|---|---|---|
| "It looks old" | NN/g lists "we haven't updated it in many years" and "it looks old" among common but flawed reasons. An old look can still hurt credibility, but it is a styling decision, not proof the structure is broken | Whether a new look on the same structure would solve it |
| Search traffic dropped | Google lists algorithm updates, seasonality and shifts in what people search for among the usual causes of a drop, none of which a redesign fixes | Search Console, over the longest range available (16 months), against the same period last year |
| Few enquiries | Often a page that does not answer the visitor's question or offers no easy next step | Our guide to why a website is not bringing enquiries |
| A competitor launched a new site | Their site tells you what they did, not whether it works | Whether customers are actually choosing them over you, and why |
| Something is broken (a form, a link, a slow page) | A bug is a bug | Fix the bug, then see whether the sign is still there |
One entry deserves a second look. Search traffic that drops right after a redesign needs diagnosis, not another redesign. Our guide to what to check when traffic drops after a redesign gives the order to work in.
What does a redesign cost you even when it is the right call?
Three costs surprise people.
Your visitors relearn the site. NN/g's point is that "people don't like change": returning customers have learned where things are, and they balk at change even when the new design is clearly better. That is an argument for keeping what already works.
Search risk. If any address changes, you need a map, redirects and launch checks, covered in our guide to redesigning without losing search traffic.
Content work. A new design needs words and photos for every page. Planning that work early is the subject of our guide to planning a website redesign.
What are the options between "do nothing" and "start over"?
A redesign is one point on a range, and the middle of the range is often the right place to stop.
Fix what is broken where it sits: the form, the speed problem on one template, the missing page.
Restyle the existing structure with a new look, keeping addresses and content. This is the lowest-risk visual change.
Rework one section, such as the shop, the services pages or the contact path, and leave the rest.
Replace the platform when editing or structure is the real limit.
Rebuild structure, content and design together, when several of the signs above are present at once.
Each step down that list costs more and carries more risk to your search presence. Start at the top and move down only when the evidence says so.
How often should a website be redesigned?
None of the sources we read gives a fixed interval, and a calendar is a poor trigger. NN/g treats "we haven't updated it in many years" as an insufficient reason on its own. A better habit is a yearly review using the six tests above: the structure, the editing, phones and speed, accessibility, coherence, and your own numbers. A site that passes can run for years, and in our view one that fails two of them may not be worth another year of patches.
What this guide will not tell you
It cannot tell you whether your particular site needs a redesign, because that depends on evidence from your own analytics and your own visitors. A website also cannot fix an offer customers do not want. And if a simple site-builder template does the job for a few pages you can edit yourself, a redesign by a studio may be more than you need. If that is your situation, we would rather tell you so.
Checklist: is it the site?
Checklist
- Write down each symptom, with an example, before you think about a solution.
- List everything the business sells or does today, and check each item has a page and a place in the menu.
- Ask the person who maintains the site to change a phone number and publish a new page without help.
- Open your ten most important pages on a real phone.
- Check real-visitor speed data against 2.5 seconds, 200 milliseconds and 0.1.
- Try the menu, one form and checkout with the keyboard only.
- Check search traffic over 16 months against the same period last year before blaming the design.
- Try the cheaper options first: a fix, a restyle, or a single reworked section.
- If you land on a rebuild, record a baseline (search clicks per page, enquiries per month) before anything changes.
Where to go from here
If the evidence says a redesign is right, the next step is deciding what the new site must do before anyone opens a design tool: our guide to planning a website redesign covers it. If you are choosing between a redesign and fixing the page that brings no enquiries, start with why a website is not bringing enquiries.
If a redesign looks likely and you want to know what it involves financially, read how much a website redesign costs next.
When a redesign is on the table, our website redesign and migration service is where to start, and if a website builder is enough for you, we will say so. You can discuss your project with us without a technical brief.
Sources
Nielsen Norman Group, a usability researcher, "Radical Redesign or Incremental Change?" (2015): https:/
Google Search Central, "Mobile-first indexing best practices": https:/
web.dev, "Web Vitals": https:/
W3C Web Accessibility Initiative, "WCAG 2 Overview": https:/
Google Search Central, "Debugging drops in Google Search traffic": https:/