The most expensive redesign is the one that changes the colours and leaves the problem in place.
Before you commission anything, spend a few hours finding out what is actually wrong. You do not need a full research project. You need enough evidence to avoid rebuilding a broken thing beautifully.
1. Find out what people do now
If you have analytics, look at the pages people land on, the pages they leave from, and the path between them. If you do not have analytics, ask the people who answer the phone and the inbox. They know which questions repeat.
Three things are usually enough:
- The three pages that get the most traffic.
- The three pages that get the most enquiries.
- The pages people complain about.
If the enquiry comes from a page nobody visits, the problem is not the homepage.
2. Check the search console
If the site has Google Search Console, look at:
- Queries with impressions but low clicks. The page may be ranking for the wrong thing or the title may be weak.
- Pages that lost traffic after a change.
- Coverage errors and pages excluded from the index.
- The mobile usability report.
If Search Console is not set up, that is a finding in itself. You cannot manage what you cannot see.
3. Read the content
Open the site and read it as a customer would. Then write down:
- Pages that say nothing specific.
- Pages with a call to action but no reason to act.
- Old prices, old team members, old addresses.
- Duplicate pages that say the same thing with different titles.
- Pages that exist only because someone once asked for them.
A redesign is a good time to delete. A content inventory is cheaper than a new template.
4. Test the boring things
The failures that cost enquiries are rarely visual:
- The contact form sends to an inbox nobody reads.
- The phone number is wrong or not tappable.
- The booking flow breaks on a phone.
- The site takes more than three seconds on a mobile connection.
- Images are 4MB and the page jumps while they load.
- The cookie banner covers the only button.
- The navigation has no current-page state.
Check these on a real phone, on a slow connection, not only on your laptop.
5. Check what you own
Before a rebuild, confirm:
- The domain is in your name and you have the registrar login.
- The email accounts are in your name and you have admin access.
- The hosting and analytics accounts are yours.
- You have the source files, not only the live site.
If a supplier owns any of these, fix that before the design work. It is cheaper to move an account than to rebuild around one you cannot access.
6. Decide what to keep
A redesign does not have to start from zero. Keep:
- URLs that already rank or get links.
- Content that answers real questions.
- Brand elements people recognise.
- Forms and flows that already work.
Change the structure when the structure is the problem. Change the surface when the surface is the problem. Doing both at once makes it impossible to tell which one worked.
7. Scope the audit, then the redesign
A UX audit should tell you:
- What is broken.
- How you know.
- What to fix first.
- What not to touch.
That is the input to a redesign, not a replacement for one. I quote the audit after a short review, then scope the redesign from the findings. If the site only needs a tune-up, I say that instead of selling a rebuild.
The Product and UX service covers the audit and the redesign. The contact form is where it starts.
A redesign should be a decision, not a fresh coat of paint.