New website from scratch or redesign: how to decide (with honest criteria)

“I was told I need a new website” or “I think my website needs a redesign” or “I just want to change the look, not lose what I have” are the three phrases I hear most in the first call with a client. And all three have different answers, depending on 7 specific things. In this article I explain how to decide, with honesty and without selling you anything.

Why this Decisión matters (and why most people get it wrong)

The most expensive mistake I see in my work is not “picking the wrong colour for the site” or “running a month over schedule”. It is deciding wrong between new website and redesign. This happens because:

The client who wants a new website sometimes only needs a redesign and would be saving EUR 2,000-5,000.
The client who wants “just a redesign” sometimes needs a new website because their current platform is so broken that any patch is throwing money away.
The client who “just wants to change the look” sometimes discovers that their site takes 6 seconds to load and that is why they are not getting customers.

The trick is having objective criteria, not following your gut or whatever Google says first.

The 7 criteria that matter (in order of weight)

1. Platform: what is under the hood

This is the first thing I look at. The platform (WordPress, Wix, Squarespace, Joomla, a 2012 custom build) determines 60% of the Decisión.

Signs you need a new website (not patches):

WordPress with 15+ outdated plugins (some surely has a 2019 vulnerability)
Built with a page builder like Elementor, Divi or Visual Composer (the HTML output is heavy: 800 KB - 2 MB of HTML for a page with 5 sections)
Wix, Squarespace, Weebly (you cannot touch the code, you cannot migrate SEO properly, you cannot install full Schema)
Joomla or Drupal not updated in over 2 years
A 2015-2020 custom build with jQuery, no forced HTTPS, no support for PHP 8
Your site is a platform subdomain: yourbusiness.wixsite.com/my-site or similar (kills SEO)

Signs a redesign is enough:

WordPress with a modern theme (Kadence, GeneratePress, Astra, Blocksy) and few plugins, all up to date
Custom built in the last 3-4 years with a modern stack (Astro, Next, Nuxt, SvelteKit)
Webflow or Framer site (recent and maintained)
Your platform is alive (receives updates, has documentation, has people who can help)

2. Speed: how long it takes to load (on mobile, on real 4G)

Not in your office with fibre. On a real mobile phone, with 4G and 60% signal, what your customer sees. Test: open pagespeed.web.dev and paste your URL with the “mobile” option on.

Lighthouse score (mobile)	Diagnosis	Recommended action
90-100	Excellent, leave it	Invest in content, not platform
70-89	Good, room to improve	Tune-up: images, fonts, CSS, 1-2 days of work
50-69	Bad, the user notices the slowness	Redesign on the same platform or migration
30-49	Very bad, Google penalises you	New website or full migration
0-29	Critical, your site feels broken	New website, no question

For a local business with customers in Spain, below 50 on mobile is lost revenue. People do not wait, they go back to Google and click the next result.

3. SEO: where you rank for the searches that matter

This is the most misunderstood criterion, so I will explain it well. The question is NOT “are you on Google” (almost everyone is), but “are you on the searches that bring you customers”.

The searches that matter for an SME are the ones with purchase intent (transactional) or local search (local). Examples:

“employment lawyer in Madrid” (transactional + local)
“best pizza near me” (transactional + local)
“how to apply for unemployment in 2026” (informational, not transactional)
“second-hand car prices” (transactional, no localisation)

For each important search for your business, check in Google incognito: what position do you show up in? If you do not show in the first 3 pages, you have an SEO problem. If you are on the first page but below position 5, you have a content + authority problem. If you are in the top 3, your problem is not SEO, it is conversion or design.

If you do not show up on key searches: before deciding between new website or redesign, do an SEO audit first. Sometimes the problem is not the platform or the design, it is that your site does not have the right content or it is missing structured data. That gets fixed with an afternoon of work, not a new website.

4. Content: what you have and how up to date

I count the pages, look at quality, look at when it was last updated. This tells me if a migration is worth it.

3-5 pages with 200-400 words each: you have minimum content. It works for a presentation card website, but not for SEO. You need more pages, either on the current site or the new one.
10-20 pages with 500-1000 words, not updated in 2+ years: you have a content problem, not a platform one. Before migrating, consider rewriting.
30+ pages, updated content, all coherent: your site has real SEO value. Be careful migrating: if you lose URLs or content changes too much, you eat a 3-6 month traffic drop. The migration has to be flawless.

5. Conversion: how many leads the site brings you per month

This is the most important data point of all, and the one almost nobody looks at. If your site brings you 5 leads per month and each lead is worth EUR 200 on average, that is EUR 1,000/month or EUR 12,000/year. A EUR 3,000 new website pays for itself in 3 months if you get +2 leads/month. A EUR 3,000 new website that keeps the same 5 leads/month is throwing money away.

Quick formula:

How many leads do you get per month? (forms, calls, WhatsApp)
How much is each lead worth on average? (average ticket / close ratio)
How much are you losing to a bad website? (30% less conversion is typical)
How much does a new website cost? (including content migration and SEO)
In how many months does it pay for itself? If under 6 months, go for it. If over 12, do not.

If your current site brings 0 leads (no proper form, no visible phone, etc.), the Decisión is easy: whatever you do, you will get more leads. But you may not need a new website, just fix the basics.

6. Age of the site: how long it has been live

This is counter-intuitive, but a site over 4-5 years old accumulates technical debt that is not visible at first glance:

Outdated plugins and themes
Obsolete technologies (jQuery, Flash, external fonts that no longer load)
Code from whoever made it that is no longer in the business and there is no one to maintain it
Accumulated SEO penalties (Google sometimes “forgets” sites for years)
Design that no longer matches the current brand

From 5 years onwards, consider a full migration. Not on a whim, but because maintaining a 2018 site with its technologies costs more over 3 years than building a new one.

7. Your team: who maintains the site now and who will maintain it after

This is the criterion almost nobody mentions and the one that kills the most sites.

If a cousin or friend who “knows about computers” maintains it: you need a site that anyone can maintain, not one that requires a senior. Migrate to something self-manageable or assume you will pay for external maintenance forever.
If whoever made it maintains it, but they are no longer available: you have a vendor lock-in problem. Migrate.
If you maintain it and you are comfortable with WordPress: do not migrate just to migrate. Invest in good plugins and training.
If you do not want to maintain anything and prefer to pay someone: hire monthly maintenance and forget about it. But choose well (or read my other article on the topic).

The quick test: 2 minutes that tell you what to do

If you do not want to read all the above, do this test:

Open your site on mobile with 4G and time how long it takes to see something useful (text, photos, menu). If it takes more than 4 seconds, you need to migrate or tune up.
Type your business name + city in Google incognito. If you do not show in the top 3, you have an SEO problem that probably needs more than a redesign.
Count the leads you got last month from the site. If 0 or almost 0, the problem is conversion or absence, not platform.
Open your site and view source (right click, “view source”). If it is full of div class="et_pb_section" or data-w-id= from Wix/Webflow, you cannot tune well. Migrate.
Ask whoever made your site if they can give you the code and the credentials. If they say “no, it is on my server” or “I have to ask my partner”, you have vendor lock-in. Migrate.

If 3 of 5 say “be careful with this site”, you have to do something. If 4-5 say “be careful”, migrate. If 5/5 say “be careful”, migrate and do it fast.

When migrating is a mistake

In case you thought I always recommend migrating: no, there are cases where it is a mistake:

Your site has 30+ pages, ranks well, brings leads. Migrating implies risk of SEO traffic loss. Sometimes the best is a technical tune-up (change platform underneath without it being noticeable) and a visual redesign.
Your site is 2 years old, made with a modern stack, loads in 2 seconds. Leave it alone. Invest in content, not platform.
You do not have time or money for a migration right now. Better a EUR 500-1,500 tune-up that fixes the urgent (speed, basic SEO, schema) and save the migration for when you are ready.
Your market is local and your site is “good enough” for what you need. You do not have to be the fastest in the neighbourhood for the customer to find you on Google Maps and call you.

In those cases, I tell you “do not migrate” without charging you for the consultation. My reputation depends on that.

Conclusión

The new website vs redesign Decisión depends on 7 things: platform, speed, SEO, content, conversion, age and team. It is not black and white: every case is unique.

If you want me to look at your specific case, write to me or by WhatsApp and I give you an honest diagnosis in 30 minutes over a call. No commitment, no selling you anything you do not need.

And if you end up deciding to migrate, here are the services and pricing with the full breakdown.