Skip to content
PROOF48
All answers
marketing

How to actually market an online store that ships to more than one country

Pick one or two countries you can serve well, publish a separate page for each language and mark the versions with hreflang, then measure sales per country before adding another. The channels are the same as for any store: search, content and paid ads. PROOF48 has no case study of a store selling in several countries, and we promise no sales or rankings.

Founder, PROOF48Updated 2026-11-034 min read

01Choose the countries you can serve, not the ones you can list

A country is worth marketing to once you can answer, for a customer there, what shipping costs, how long it takes, what the return process is, which currency they pay in and what tax applies. If any of those answers is vague, the ads will bring people to a checkout that loses them. In our view one or two countries done well beat six done badly, though we have no measured data to show it. Tax rules for selling across borders are a question for your accountant. Our post on selling internationally from a Romanian store is a starting point.

  • Shipping cost and delivery time to that country
  • Returns: who pays and how long it takes
  • Currency and the tax that applies
  • Start with one or two countries, add a third only after you can read results

02One page per language, and tell Google which is which

Google's documentation on localized versions says to tell it about each language or region version of a page, using HTML link tags, HTTP headers or an XML sitemap, and treats the three as equivalent. Each version must list itself and every other version, and the links must be reciprocal: if page A points to page B, page B has to point back. Google also recommends an x-default fallback page. It adds that it works out a page's language with its own algorithms, not from the hreflang tag. We build sitemaps for proof48.com with hreflang for this reason. Write each language natively, not word for word, and note that sites in more than two languages get their own timeline with us.

  • Declare versions through HTML link tags, HTTP headers or an XML sitemap
  • Each version lists itself and every other version
  • Links must be reciprocal, with an x-default fallback page
  • Write each language natively

03The three channels, and what our plans list for each

Search comes first for a store: a product page that answers the question behind the search, one page per product, with the right schema. Our post on product page SEO for a small store goes through it. Content comes next. BROADCAST, at 597 EUR a month, includes FOUND, brings the blog to 4 AI posts a month (FOUND has 2) and adds 12 AI videos. Paid ads come last. GROWTH, at 1,497 EUR a month, adds Meta and Google ad management, 10 new ad creatives a month, audience research and retargeting, and conversion tracking with Pixel, CAPI and GA4. Ad spend is billed separately. The plans do not list ads in several languages or markets, so we would scope that with you first.

  • Search: one page per product, answering the question behind the search
  • Content: BROADCAST includes FOUND, brings the blog to 4 AI posts a month (FOUND has 2) and adds 12 AI videos
  • Paid: GROWTH adds Meta and Google ad management, spend billed separately
  • The plans list no ads in several languages or markets, so scope that first

04Where PROOF48 is the wrong choice

A store is its own build, 2,500 to 4,500 EUR, separate from the 48-hour website and from the monthly plans, and it gets its own timeline. No plan of ours lists marketplace listing management, so if your sales come mainly through a marketplace, we do not cover that part. We are the wrong choice when the offer or price is not decided, when there is no product photography or brand direction, and for industries with strict advertising rules such as medical claims. We also promise no sales, no ranking and no date in any country.

  • Stores: 2,500 to 4,500 EUR, own timeline, not 48 hours
  • No plan lists marketplace listing management
  • Offer or price not decided, or no photography: not a fit
  • No promise of sales, rankings or dates
From a build we shipped

A luxury corsetry store built with its brand: a launch, not a sales story

Noryse is a luxury corsetry label. PROOF48 built the brand and the store in the same pass: the name, positioning, palette and type system, then a production website with six product pages, Stripe checkout and a private portal for the manufacturer, with bilingual editorial copy. It is live at noryse.com and taking orders. The case study does not say which countries it ships to, and it is a launch with no sales history, so we claim neither.

See the build: Noryse

Questions, answered

Do I need a separate website for each country?

Not necessarily. Google's documentation says each language or region version needs its own URL, which can sit on the same or a different domain, and that you tell Google about them with hreflang.

How many countries should I start with?

In our view one or two, then add another once you can read sales per country. We have no measured data on the right number.

Does PROOF48 build online stores?

Yes, at 2,500 to 4,500 EUR with their own timeline. A store is not part of the 48-hour website offer.

Can you promise sales in a new country?

No. We promise no sales, no Google position and no date. Noryse is a launch with no sales history, so we publish no sales results for it.

Where PROOF48 fits

Start with one or two countries you can serve end to end, give each language its own page marked with hreflang, and read sales per country before adding another. A store is a separate build at 2,500 to 4,500 EUR, and no PROOF48 plan promises sales.

· Founder, PROOF48

Cosmin has worked in online marketing and software development since 2018, and has built with AI every day since January 2023. He still writes code from scratch daily, in Python, React, Rust and C#. PROOF48 ships business websites in 48 hours and invoices afterwards, so every opinion on this blog was paid for in real builds, not in theory.

Related articles