BlogGuides
Your website has been sitting still for years: rebuild it or fix it?
One click, and Google will show my articles more often in your results.
The decision comes down to four checkable signals, not a feeling that the site looks old: whether the content still describes today's business, whether the site passes the speed thresholds Google measures, whether the software running it still receives security updates, and whether a targeted fix costs less than a rebuild. When at least two of these fail together, rebuilding is almost always worth it; with a single isolated signal, fixing it usually is.
The content still describes the business you run today
This is the first criterion, and the easiest to check: open the site and ask whether a customer reading it today would recognize your actual business. Services you no longer offer, photos of a location you left years ago, hours or contact details that changed long ago: these aren't design flaws, they are wrong information a customer takes as true. This gets fixed by rewriting the text, not rebuilding the structure, and it's the easiest signal to isolate from the other three: on its own, it almost never justifies a rebuild.
Speed is measured, not guessed
From here on, the criteria have a public threshold, not an opinion. A site counts as good if its main content appears within 2.5 seconds, if it responds to a tap within 200 milliseconds, and if it doesn't jump around abruptly while loading, according to Google's official documentation on Core Web Vitals, consulted on November 5, 2026. Google doesn't call speed a direct ranking factor: it says these signals "align with what our core ranking systems seek to reward," a distinction I cover with the full data in website speed: how much it matters. What matters here is something else: on a recently built site, a failing score usually gets fixed without touching the structure. On an old base, the same fix can turn out to be impossible, because the underlying system no longer supports the techniques it would need.
Security: does the software still get updates?
This is the most technical criterion, but the question that checks it is simple: does the version of the language the site runs on still receive security patches from its maintainers? PHP, the language behind most Italian business websites, follows a public schedule: version 7.4 stopped receiving patches on November 28, 2022, and version 8.0 on November 26, 2023, according to PHP's official end-of-life page, consulted on November 5, 2026. As of November 2026, only versions 8.2 and later still have active security patches, according to PHP's official page on currently supported versions, consulted the same day. WordPress, the most common system, states this openly: it still works with older PHP versions, but warns that they "have reached their official End Of Life and may expose your site to security vulnerabilities," according to WordPress's official requirements page, consulted on November 5, 2026. The same applies to the CMS itself: only the latest major release gets guaranteed support, and earlier ones receive patches "only as a courtesy, with no guarantee and no fixed timeframe," according to WordPress's official page on supported versions, consulted on November 5, 2026. A site stuck on a base that old isn't just less modern: it's a site nobody is protecting anymore.
Mobile isn't a cosmetic detail
Since October 31, 2023, Google has completed the move to mobile-first indexing for nearly every site: it uses the mobile version to decide how to index and rank a site, not the desktop one, according to Google's official announcement on completing mobile-first indexing. A site a phone displays poorly doesn't just lose visitors to a bad impression: it starts at a disadvantage in search results before anyone even judges it. What being responsive really means, and how much phone traffic actually weighs in Italy today, I cover in responsive website: what it means.
What a fix costs, compared to rebuilding from scratch
The math changes case by case here, but the logic stays the same. A targeted fix costs less than a rebuild when it solves one isolated problem without touching the rest. On a base that has piled up several issues at once, the opposite happens: each correction risks breaking another, because the structure wasn't built for that kind of change, and the time spent making old and new pieces coexist ends up costing more than starting over. As a reference point, a new website in Italy runs from €800 to €8,000 depending on what it needs to do, market brackets I cover in detail in how much does a website cost: that's the number to compare a major fix quote against, not an isolated correction.
When to fix, when to rebuild
| Signal | Usually fix it | Usually worth rebuilding |
|---|---|---|
| Content | Text or photos need updating | The business described no longer exists |
| Speed | An isolated failing score, to correct | Still fails thresholds after simple fixes |
| Security | The software still gets patches | The software hasn't been patched in years |
| Technical changes | Each fix stays isolated | Each fix breaks another one |
A single signal in the right-hand column almost always gets handled on its own, without rebuilding anything. When at least two show up together, the time and risk of continuing to patch the old base start to outweigh a new project.
If you're not sure which column your site falls into, it's the same check I run as part of the digital check-up: otherwise get in touch and we'll look at it together.