The decision to redesign is usually made right after somebody in management looked at a competitor’s site. That is a poor reason to spend $3,000. Below are seven signs that a site is genuinely losing money, three situations where a redesign treats the wrong illness, and a half-hour diagnostic that tells you which case you are in. All of it based on metrics you can pull today.

Seven signs a redesign is genuinely warranted

Each one is verifiable with a number rather than a feeling. If three or more apply to you, the question stops being whether to rebuild and becomes how much of it to rebuild.

1. 1Mobile conversion is half of desktop at comparable traffic. This nearly always means the site was adapted for phones after the fact instead of designed for them. It is the most expensive sign on the list, because mobile is usually 60–75% of visitors.
2. 2Pages take over 3 seconds on a mobile connection and optimisation does not fix it. If compressing images and pruning scripts bought you less than 20%, the problem is foundational — the theme, the builder, or a dozen plugins that cannot be untangled.
3. 3You cannot add a page yourself. When every edit means emailing a developer and waiting three days, the site stops being a marketing tool. That blocks content, and organic traffic with it.
4. 4The structure no longer fits the offering. Twelve services became forty; three locations became ten. If navigation was patched rather than redesigned, visitors cannot find things and search engines cannot read the hierarchy.
5. 5The technology no longer receives security updates. An unsupported CMS version is not an aesthetic problem, it is a countdown to a breach. Here the redesign simply coincides with a migration you have to do anyway.
6. 6Positioning changed but the site still speaks the old language. You sell a different product to a different audience at a different price, while the homepage carries a promise from five years ago. Copy and structure matter more than the layout here.
7. 7Enquiries go missing. No integration with email or CRM, submissions landing in an inbox nobody reads daily. It is not a design issue at all — yet it is the thing most often uncovered during a pre-redesign audit.

Three situations where a redesign is the wrong move

• There is almost no traffic. A redesign does not bring visitors. At 200 sessions a month, changing the layout is polishing a shop window on an empty side street. Fix the traffic channel first, appearance second.
• The problem lives on one page. If only a service page or an enquiry form underperforms, rebuilding the whole site costs $3,000 instead of $400. Focused work on a single screen gets the same result in a week.
• The site is under a year old and on a current stack. A second redesign within 18 months nearly always means the first problem was never diagnosed. A new wrapper on the same mistake produces the same conversion rate.

The half-hour check before you decide

This is the minimum evidence that separates a justified decision from a wish for fresh visuals. Every item is available in free tools.

If speed is the bottleneck, start with our breakdown of website loading speed — often no redesign is needed at all. If a specific funnel step underperforms, it is far cheaper to run the cycle described in the guide to conversion rate optimisation.

What to do instead of a full rebuild

If you do have to rebuild

1. 1

   Record your baseline before anything starts

   Conversion, speed, traffic by page, positions for 20 core queries. Without this you cannot prove the new site beats the old one.

2. 2

   Keep whatever already works

   Do not radically alter your highest-traffic organic pages. Change the presentation; keep the structure, headings and copy that already rank.

3. 3

   Map redirects in advance

   Every old URL must land on a relevant new one, never on the homepage. This is the main cause of post-launch drops; the full sequence is in our site migration checklist.

4. 4

   Ship in stages if the site is large

   Service templates first, then the blog, then the homepage. A mistake then costs you one section instead of all your traffic.

5. 5

   Watch for eight weeks after launch

   A 10–20% dip in the first fortnight is normal. If metrics have not returned after two months, the cause is technical — not "search engines still catching up".

Frequently asked questions about website redesigns

In short

• A redesign is justified when three or more measurable signs coincide, not when the site "looks dated".
• The costliest signs are weak mobile conversion, load times above three seconds, and being unable to manage your own content.
• With low traffic, a localised problem, or a site under a year old, a redesign treats the wrong illness.
• Targeted work in the $300–900 range resolves most of the symptoms people bring to a redesign conversation.
• If you do rebuild: record baselines first, protect your traffic pages, and map redirects before launch day.