A redesign, a new domain, or several brands merging into one: your budget only covers the important pages first, and the blog posts, news, and documents come later. That's the moment most migrations go wrong, because the easy shortcut is to point everything somewhere "for now" and fix it later.
This checklist walks through sizing the move, building a redirect map, redirecting each old URL exactly once, and then verifying every redirect with Minifetch, including a bulk check of the long tail. It follows Google's published site move guidance (linked inline).
Google's site move guide↗ gives different advice by size: move small and medium sites all at once; large sites may move one section at a time to monitor and fix problems faster, ideally after test-moving one piece first.
The guide doesn't define "large". The only page counts Google publishes are in its crawl budget guide↗ (1 million+ unique pages is large; 10,000+ rapidly changing pages is medium or larger). Our rule of thumb from those lines:
Merging sites? Count the combined URLs arriving on the new domain, not each old site's size. Google treats each hostname as one site, so five 3,000-page sites become one 15,000-page site. Expect a merge to take longer to settle than a simple domain move.
List every URL on every old site: XML sitemaps, a crawl, Search Console's Pages and Links reports, analytics landing pages, and server logs. Then mark each one:
One row per old URL, pointing at its final destination on the new site. This spreadsheet is the source of truth for the whole migration, and it's also the input for the bulk check later:
old_url,intended_url,equity https://old-brand-a.com/services/audit,https://newbrand.com/services/audit,high https://old-brand-a.com/news/2019-launch,https://newbrand.com/news/2019-launch,low https://old-brand-b.com/old-promo,410,low
Use 404 or 410 as the destination for pages you're dropping on purpose.
No. Redirect each old URL to its equivalent page: same topic, same intent.
Google's redirect docs↗
warn against sending many old URLs to one irrelevant destination; it may be treated as a
soft 404, and the old page's signals don't carry over. If there's no equivalent, let the
old URL return 404 or 410. That's an honest answer, and it
beats a misleading redirect.
Yes, but phase the design, not the URLs. Every old URL should be redirected once, straight to its final destination. Pointing pages at a temporary stand-in and re-pointing them later creates redirect chains, risks soft 404s in between, and makes Google process each move twice.
A common case: five brand sites merge into one new site. Phase 1 launches services and sectors pages; phase 2 brings the news, blog posts, and documents. Two ways to handle the phase 2 content:
Don't let Search Console's Change of Address tool drive this decision. It's built for moving one whole site to a new domain, and Google advises against combining several moves into one destination. The 301 redirects do the real work.
Aim for one hop: old URL straight to the new page. Google's crawlers stop after 10 hops, and Google's site move guide recommends keeping any unavoidable chain ideally no more than 3 hops and fewer than 5.
Merger chains usually come from rules stacking up on each old domain: http to https, then non-www to www, then old domain to new domain, then old path to new path. Collapse them so every variant of an old URL jumps straight to its final destination.
Minifetch grades every redirect chain by the same rules:
Use 301 (or 308) redirects. Temporary 302 and 307 redirects can keep the old URL in search results.
Credit card & API key — Sign up on the dashboard and get credits worth 25 free audits automatically. Then create your API key. No credit card required to begin. Top up for as little as $2 with your credit card. Recommended for most builders.
USDC on Base or Solana — Just load your wallet with USDC on Base or Solana and you're ready. No "gas token" (ETH or SOL) required. No Minifetch account setup needed. Recommended for agents and agent builders.
Corporate sites often block bots in robots.txt, and an old domain that blocks
Minifetch blocks every check on it. Run one free preflight per old hostname, plus the new domain:
curl "https://minifetch.com/api/v1/free/preflight/url-check?url=https://old-brand-a.com/"
If you get allowed: false on a site you own, see
How to Allow Minifetch to Fetch Your Pages.
Run the Minifetch technical SEO page audit on each old high-equity URL. Minifetch follows the redirect and grades both the chain and the page it lands on. You can use the form on the homepage to do it by hand with a nice UI, or programmatically you can use our API client on NPM (Quick Start ↗), or just use curl from your command line:
curl "https://minifetch.com/api/v1/run/seo-page-audit?url=https://old-brand-a.com/services/audit" \ -H "Authorization: Bearer [your_api_key]"
Check these first, in the order the audit returns them:
data.url is the final URL. It should match your redirect map.data.redirects lists every hop with its status code.data.performance.ttfb: time to first byte, including the redirect hops. Each hop's own time is in data.redirects.chain[].ttfbMs, so you can see which server is slow.data.responseHeaders["x-robots-tag"]: no leftover noindex header from staging.data.metadata.canonicalMatchesSelf: a canonical still pointing at the old domain is the classic migration miss.data.metadata.canonicalTagCount: more than one canonical tag is a fail. Templates merged from several sites often stack them.data.metadata.robots: no leftover noindex meta tag from staging.data.social.openGraphUrlMatchesSelf: same miss as the canonical, for og:url.
For low-equity rows, the cheaper metadata extraction API endpoint returns everything the redirect check needs.
Narrow it with the fields option to keep responses small (also available when you use minifetch-api on NPM):
curl "https://minifetch.com/api/v1/extract/url-metadata?url=https://old-brand-a.com/news/2019-launch&fields=redirects,url,responseStatusCode,responseHeaders,performance,canonical,canonicalUrls,robots,og:url" \ -H "Authorization: Bearer [your_api_key]"
For each row, confirm, in the order the endpoint returns the fields:
data.redirects.count is 1 (0 means it was never redirected; see the hop limits above).data.redirects.chain is a 301 or 308.data.url matches the map, and no deep page lands on the homepage.data.responseStatusCode is 200.noindex in data.responseHeaders["x-robots-tag"] or data.robots (the robots meta tag). A noindex left over from staging deindexes the new page.data.performance.ttfbMs is 800ms or less. Up to 1,800ms is a warning; over that fails. These are Google's
published TTFB thresholds, and the total includes every redirect hop (data.performance.redirectTimeMs
is the redirect share). Check data.redirects.chain[].ttfbMs to find the slow hop.data.canonicalUrls has one entry. Two or more is a conflict: Google may ignore them rather than guess.data.canonical and data["og:url"] don't point at an old domain.
Minifetch checks HTML pages only, for now: a PDF or other non-HTML URL returns invalid url, and you're not charged. Spot-check those redirects by hand.
Minifetch waits between requests to each domain (1 second by default) to prevent us from slowing your server down for your real users,
and it waits on every redirect hop. Every row ends on the new domain, so the new domain sets the pace: a 2,000-row map
takes about 35 minutes. Moving a single domain? The old domain gets every row too. Since you control both ends of the
migration, speed it up with a sub-second Crawl-delay (ex: 0.25) for the minifetch user agent in the
robots.txt of every domain in the chain, old and new.
At least a year, and longer if you can. That's Google's guidance, because it takes that long to recrawl old URLs and reassign the links pointing at them. In a merger, that means keeping every old domain registered and renewed too.
Re-run the bulk check right after launch, then weekly for the first month. The metadata endpoint is cheap enough to use for routine monitoring; re-audit high-equity pages whenever they change.