Start with the symptom, not the site
“The website needs redoing” is a conclusion, and it is usually reached before anyone has identified the problem. Ask instead what is actually going wrong, because the answers point in very different directions:
- Nobody finds it. A visibility problem. A new design does not fix it and may make it worse.
- People find it and leave. Could be speed, could be mobile layout, could be that the page does not say what you do in the first screen. Some of these are cheap fixes.
- People stay and do not enquire. A conversion problem, usually about clarity, proof and how hard the form is.
- I cannot change anything myself. A platform and access problem, and the most legitimate reason to rebuild on this list.
- It looks old. A real issue for credibility, and the one most likely to be solved by less than a rebuild.
Two of those five are design problems. The rest are not, and a redesign that ignores them produces a better-looking site with the same results — which is the most common bad outcome in this whole area.
Six checks before you commit
1. Is it usable on a phone, in the real world?
Not “is it responsive” in the abstract. Open it on your own phone, on mobile data rather than office wifi, and try to do the thing you want customers to do. If you cannot complete the form, if text needs pinching, or if a button sits under a sticky banner, that is a serious problem — and it is often fixable with CSS rather than a rebuild.
2. How slow is it, and why?
Measure before you conclude. Speed problems are usually caused by oversized images, too many plugins, or a cheap host — all fixable without touching the design. If the slowness is in the theme’s own code, that is a stronger argument for rebuilding.
3. Can you edit it without paying someone?
If changing a phone number requires a developer, the site is costing you money continuously and, worse, it stops you making changes at all. Over a couple of years this is often the largest hidden cost, and it is a sound reason to move platforms.
4. Is it actually in Google?
Search site:yourdomain.com. If most of your pages are missing, you have an indexing problem, and rebuilding without diagnosing it means carrying the same problem into an expensive new site. There is a separate walkthrough for diagnosing that.
5. Is it secure and maintained?
An out-of-date platform with abandoned plugins is a genuine risk rather than an aesthetic one. Sometimes the correct reading of “the site needs redoing” is “the site has not been maintained for four years”, and the answer is maintenance, not design.
6. Does the first screen say what you do and who for?
Open the homepage and imagine you have never heard of the business. Can you tell what is sold, to whom, and what to do next, without scrolling? A surprising number of sites fail this, and it is usually a copy problem costing a fraction of a rebuild.
The three outcomes
Fix it
Choose this when the structure is sound and the problems are specific: slow images, a broken mobile form, weak homepage copy, missing tracking. The cheapest and most often correct answer, and the one nobody selling websites will lead with.
Redesign it
Keep the platform, the URLs and the content; change the appearance and layout. Right when the site works but looks dated or converts badly. Lower risk than a rebuild because the URLs — the thing search rankings attach to — do not move.
Rebuild it
Justified when the platform itself is the problem: you cannot edit it, it cannot be made fast, it cannot be secured, or it cannot do something the business now needs. “It looks old” alone is not sufficient reason, because a redesign achieves that for less and risks less.
If you do rebuild, protect what you have
This is where rebuilds go badly wrong, and the damage often appears weeks later when the traffic has already gone.
- Record what currently works before you start. Which pages bring traffic and enquiries, and for which searches. Once the old site is gone, this is unrecoverable.
- Keep the URLs wherever you sensibly can. Every URL you change is a ranking you have to re-earn. Tidier addresses are rarely worth the cost.
- Map every changed URL, old to new, with a 301. One by one, to the closest equivalent page.
- Never mass-redirect to the homepage. It is the standard shortcut and it destroys the value of every page it touches — Google treats an irrelevant redirect as a soft 404. The redirect map builder will flag that, along with chains and loops, before you deploy.
- Do not lose the content. Pages that rank do so partly because of what is on them. A redesign that halves the words on a page that was working is a common way to lose rankings while congratulating yourself on a cleaner look.
- Check the staging site is not indexable, and that the live one is. The classic launch failure is shipping with the “discourage search engines” box still ticked. The website launch SEO checklist covers the rest.
- Re-test tracking after launch. Conversion tags are attached to pages and forms, and both just changed. Submit a test enquiry and confirm it arrives.
The question worth asking last
What will be different about the business if this works? “A modern website” is not an outcome. “Enquiries from people who currently bounce on mobile” is, and it tells you what to fix and how to judge whether it worked.
If you want an outside opinion on which of the three answers applies to your site, send me the address. It is a short conversation, and sometimes the answer saves a lot of money.