Restore, modernise, repurpose or start fresh? Deciding what an old domain's website should become

A domain whose website has gone leaves you with a decision that ordinary redesign advice does not cover. Perhaps you let the name lapse and caught it again, perhaps you bought it from someone else, perhaps you registered it after it dropped. Public web archives such as the Internet Archive's Wayback Machine may hold part of what used to be there. The question is what the site should become now.

There are four honest answers: restore the original, restore and modernise, repurpose, or start fresh. They differ in the evidence they need, the rights they depend on and what they do with the old addresses that other sites still link to. Choosing badly costs more than wasted work. A faithful restoration of content you have no right to publish, or a revival that makes a closed business look as if it still trades, can mislead visitors and expose you legally.

Restore original rebuilds the previous website as faithfully as the surviving evidence and your rights allow: its pages, its wording, its images and its addresses. You fix only what is broken and reinterpret nothing.

Restore and modernise keeps what was useful and improves everything around it. The useful part is the page structure, the content that is still true and the addresses that people and other sites rely on. The improvements cover design, mobile layout, accessibility, speed, security and the routes a visitor takes to enquire or buy.

Repurpose gives the domain a new, legitimate purpose. The old site becomes context rather than content. You learn what it covered and who visited, and you keep only what still serves the people who arrive.

Start fresh ignores the old structure and builds something new. It still has to deal with the old addresses, because the links and bookmarks that point at them do not disappear when the content does.

These are not quality tiers. A fresh start is not a lesser outcome than a faithful restoration, and for many acquired domains it is the only defensible one.

Is the old content yours to publish?

Owning a domain name is not the same as owning the website that once ran on it. Copyright in the UK needs no registration, and GOV.UK's list of protected works expressly includes web content. Under section 11 of the Copyright, Designs and Patents Act 1988, copyright belongs first to whoever created the work, unless an employee made it in the course of their job, in which case it belongs to the employer. Commissioned work runs the other way: the Intellectual Property Office says a self-employed person usually keeps the intellectual property unless the contract hands the rights to the client. The freelance designer, copywriter or photographer behind an old site may still own their part of it.

The archive does not settle the question. The Internet Archive "does not make guarantees as to the copyright status of items on archive.org", and its terms require users to "ensure that such use is non-infringing". An archived page proves that something existed. It does not give you permission to republish it.

Does enough of the old site survive?

No web archive is complete, and the Internet Archive's own help pages list the usual reasons. Robots.txt exclusions can stop a crawl. JavaScript that needs the original server "will fail when archived". Orphan pages that nothing linked to were never found, and missing images usually mean "we did not archive them". Where an archived site is incomplete, the Wayback Machine "will grab the closest available date" for the missing links, so one page you browse can be assembled from captures years apart. How much survives varies enormously. In our study of 7,613 recently dropped .uk domains, a sample weighted towards well-linked names, the archive held no working homepage capture at all for 303, a single homepage version for 206 and more than 200 distinct versions for 148. What can actually be recovered from an archived website? has the full results.

Will the site serve the same purpose as before?

If the purpose is the same, meaning the same business, subject or audience, continuity is an asset. Keeping the old addresses working is part of serving returning visitors. If the purpose has changed, continuity becomes a liability. A page that looks like the old business's page, at the old business's address, tells visitors something that is no longer true.

Which old addresses still matter?

A few old URLs keep attracting links, bookmarks and the occasional search visit. Most do not. The value sits in the specific addresses that other sites still point at and that you can serve with genuinely equivalent content. Your redirect and retirement decisions are about that set of addresses, not about the domain as a whole.

When it fits. You are the business, or you act for it, and the site went down by accident. The domain lapsed and you caught it back, the host closed, or a migration failed. It also fits when you have permission from whoever owns the content, and the site's value lies in being what it was, such as a reference resource, a community record or a portfolio.

What you need. Rights to the words and the images. Enough archived evidence to rebuild the pages that mattered, which means more than a home page and a few fragments. Current facts, too: a faithful copy of a 2019 price list is still a 2019 price list.

Risks. Gaps in the archive get filled with guesses, and pages get stitched together from different capture dates. Old third-party scripts and trackers come back with the pages. Contact details, prices and testimonials that were true then may not be true now. Old pages can also carry personal information: the ICO's guidance is that information about identifiable sole traders, employees, partners and company directors "may constitute personal data".

SEO. Restoring the original makes the strongest case for keeping every URL exactly as it was, because each link and bookmark then lands on the page it expects. For pages you cannot restore, Google's site-move guidance asks for a genuine 404 or 410 response; a placeholder page that pretends the old one still exists risks being treated as a soft 404.

Effort. Design effort is low and evidence effort is high. Most of the work is establishing what survived, what you may publish and which facts need replacing. The mechanics are in How to restore a website from the Wayback Machine.

When it fits. The business or subject is the same, and the old site earned links and loyal visitors, but it was built for desktop screens years ago. It also fits when the archive evidence is too patchy for a faithful copy but good enough to show the structure.

What you need. Rights to whatever you reuse. New writing can replace the rest. You also need enough evidence to recover the structure and the pages worth keeping, and an up-to-date set of facts about the business.

Risks. The main risk is changing everything at once. Google's site-move guidance recommends planning changes "one after the other, not everything at the same time". Its Change of Address help puts it more bluntly: "If you combine a site move with a redesign of the site's content and URL structure in the new location, you will probably see some traffic loss as Google might need to relearn and reassess the individual pages." Both are written for moving a live site, but the logic carries over. The more you change in one step, the harder it is to tell what helped and what hurt.

SEO. Keep the valuable addresses. Where one has to change, use a single permanent redirect to its real replacement. Google advises "redirecting to the final destination directly" rather than through a chain.

Effort. Often the highest of the four, because it combines the evidence work of a restoration with the design and writing work of a new build.

When it fits. You bought the name for what it says, not for what it used to host, and you have a legitimate new purpose for it. A closed shop's name might become an independent guide to the same subject. A rebranding company might buy a name that matches its new brand, once used by an unrelated small business.

What you need. An understanding of the old site: its topics, its audience, its active years and the addresses other sites link to. That knowledge helps you avoid confusion and decide what happens to each URL. You do not need rights to the old content, because you are not republishing it.

Risks. There are two. Visitors may mistake you for the previous owner. And the site may look like the pattern Google calls expired domain abuse, which its March 2024 announcement defined as a domain "purchased and repurposed primarily to manipulate Search rankings by hosting content that provides little to no value to users". The same announcement says plainly that "It's fine to use an old domain name for a new, original site that's designed to serve people first." How to repurpose an old domain safely works through where that line sits.

SEO. Redirect an old address only where a new page genuinely meets the same need. Google warns against redirecting "many old URLs to one irrelevant single URL destination, such as the home page of the new site", which "might be treated as a soft 404 error". Everything else returns a 404 or 410.

Effort. Moderate. Most of the work is new writing, with some research at the start.

When it fits. The old site has no bearing on what you plan to do, or very little of it survives. It also fits when the domain's history is one you would rather not be connected to, such as spam in its final years or a business you have nothing to do with.

What you need. Little from the past, apart from a list of the old addresses that still receive links or visits and a check that the domain is clean. Google's site-move guide tells new owners to "Clean up your recently purchased domain" and to make sure "there aren't any URL removals left over from the previous owner", along with any manual action for previous spam.

Risks. Old addresses can be left broken in ways that confuse people who arrive from an old link. The opposite mistake is retiring an address that has a true equivalent on the new site.

SEO. Choosing between 404 and 410 is about meaning, not speed. Google says that "All 4xx errors, except 429, are treated the same". HTTP's own definition, in RFC 9110, says a 410 tells the recipient "that the resource is intentionally unavailable", which is the more accurate message when you have deliberately retired a page.

Effort. The least research of the four. The build itself is a normal new build.

What follows is general information about UK law, not legal advice. If a decision depends on it, take advice.

Do not impersonate the previous owner. Using the old business's name, logo, staff photographs or contact details as if they were yours misleads every visitor who knew that business. It does not matter whether the domain came to you legitimately.

Do not pass off. A business's name can be protected without any registration. Section 2 of the Trade Marks Act 1994 leaves "the law relating to passing off" untouched, and HMRC's manual sums up that common-law claim as "the classic trinity of reputation (or goodwill), misrepresentation and damage". A revived site that lets the old business's customers believe they are still dealing with it is the misrepresentation.

Check registered trade marks. According to GOV.UK, registration lets the owner "take legal action against anyone who uses your brand without your permission". The IPO's online search lets you look up existing marks by keyword, owner or number before you commit to a name.

Do not republish what you do not own. GOV.UK lists "copying your work", "making an adaptation of your work" and "putting it on the internet" among the things copyright lets an owner prevent. Restoring a site does all three.

Do not imply that a closed business still trades. Look the old business up at Companies House before any restored page names it: GOV.UK lists "insolvency information" and "previous company names" among the company details available there free of charge. If the business has gone, no restored page should suggest otherwise. Old phone numbers, addresses, prices and "established since" claims are the usual culprits.

DropOps Sites, the DropOps website builder for domains you own, offers the same four approaches under these descriptions:

• Restore original: "Bring back the previous website as closely as possible."
• Restore & modernise: "Keep its useful pages and addresses, but upgrade the design and content."
• Repurpose: "Build something different while keeping historical SEO value that is still useful."
• Start fresh: "Build a new website from scratch." Old addresses are still redirected or retired, never left broken.

To inform the choice, it reads the domain's history from public web archives (the old addresses and a sample of old pages) to understand its structure, its subjects and the years it was active. Restore original becomes available only after the owner states that the content is theirs to restore (they own it, act for the owner or have permission). DropOps records that statement and makes no legal determination about it.

Whichever approach you pick, recovered details that date, such as phone numbers, prices, addresses, staff numbers, testimonials, awards and opening hours, wait for you to confirm, replace or remove them. Each important old address gets a decision: kept where it is, given a clean new address with one 301, merged, redirected, retired, or flagged for you to decide. Those decisions rest on evidence from linking sites and the archive, not on AI, and where that evidence could not be collected, the address is referred to a person rather than retired by default. Nothing goes live until you publish. What the AI does and does not do along the way is covered in Can AI rebuild an old website?