Skip to content
Mediaberg
Use caseOctober 13, 20267 min read

We migrate websites without downtime. And when a country blocks yours, we put a local IP in front of it.

Most migration horror stories end the same way: downtime, lost rankings, broken mail. Ours end with the site running better than before — and when an entire market can't reach a healthy site because of where it's hosted, we answer with infrastructure, not excuses.

A globe of network lines with a website window traveling along a dotted route — Mediaberg site migrations

Share this article

01Moving house, without closing the shop

Ask around and you'll hear the same migration horror stories: the site was down for three days, the rankings evaporated, the contact form quietly died and nobody noticed for weeks. That's why business owners postpone a move for years — even when they're sitting on an old, slow, vulnerable server. A website feels like one thing, but it's a stack: domain, DNS, hosting, mail, certificates, CDN — five systems pretending to be one.

We've been moving websites since 2006 — WordPress installs, shops, multilingual practice sites, legacy stacks nobody else wanted to touch. We register domains in our own registrar panel, connect Cloudflare via nameservers, and run managed hosting on servers maintained by our in-house IT specialists. That's not a marketing line; it's the same setup we deployed for a medical practice in Leipzig that needed three languages and zero surprises.

02The checklist that makes migrations boring

Every migration starts with a full audit: CMS and plugin versions, PHP version, database size, mail routing, DNS records and their TTLs, certificates, cron jobs, and the hidden dependencies — the third-party scripts, the old subdomains, the forgotten staging copy still swallowing crawl budget. Ten years of a "simple" website accumulate more archaeology than you'd believe.

Then comes staging: a complete copy on our infrastructure, tested with real content, real forms, real checkout. Only when the copy behaves identically do we touch DNS — with TTLs shortened days in advance, so the actual switch takes minutes, not days. Boring is the goal. Exciting migrations are how businesses lose money.

03Zero downtime is a plan, not luck

The cutover itself is choreography: final content sync, database import, cache warm-up, and a redirect map that sends every old URL to its new address — not to a 404. Visitors never meet a maintenance page because there isn't one. Search engines see the same titles, the same meta, the same schema, the same hreflang structure they trusted yesterday.

After the switch we keep watching: Search Console re-verified, Core Web Vitals compared, rankings tracked for weeks. A migration done right is invisible — to Google and to humans. If your agency can't promise that, they're rehearsing on your traffic.

04When geography becomes the problem

Sometimes the enemy isn't technical debt — it's political geography. Networks inside some countries restrict or throttle entire IP ranges, and a perfectly healthy site hosted "in the wrong place" becomes unreachable for a whole market. No broken code, no expired certificate — just a border drawn through the internet.

The case we hear most often: a site suddenly unreachable from Russia. The hosting provider shrugs — "everything works from here." Local visitors see endless timeouts. The business loses a market overnight without changing a single line of code, and the support tickets from that region just... stop arriving.

05The Russian IP move

The fix is infrastructure, not pleading with a blocklist: we place a server with a Russian IP in front of the site. A local entry point receives the visitors and forwards them to the real origin — a reverse proxy or mirror node on local ground. The domain's DNS points at that local IP; visitors inside the country reach the site instantly, everyone else notices nothing, and the origin stays exactly where it is.

Operationally it means TLS end-to-end, caching at the edge, and monitoring from inside the country — so we know the site is reachable before your customers can tell us it isn't. We won't pretend every jurisdiction is simple: sanctions, local law and payment rails all matter, and we say so up front. But "unreachable" is an engineering problem with a known solution, and we've deployed it.

06What it takes to stop worrying

A migration on our AI-accelerated track is priced by scope — a typical small-business move is measured in working days, not weeks, and includes the audit, the staging copy and the redirect map. Legacy stacks that need genuine handcraft get a plan of their own. The written proposal tells you which one you need; the audit that decides it is free.

And if your site is fine exactly where it is, we'll say so and wish you a good day. But if it's one DNS record away from a better life — or one local IP away from an entire market — we'll show you precisely what we'd do, step by step, in your language: English, German, Russian, French, Spanish and more.

Reading is cheaper than guessing.

Every article above is a pattern we've lived through with real clients. If one of them sounds like your situation — the audit is free and the numbers are real.

Get my free audit

Keep reading