SaaS international expansion playbook

Nearly every top ranking guide to SaaS growth names international expansion as something it does not cover. That gap is why most expansion decisions get made in a board meeting on the strength of a market size chart, and why so many of them produce a translated homepage, a flat quarter and a quiet retreat.

The decision is more tractable than it looks. You already hold the only market signal that has been tested against your actual product: the signups arriving from countries where you have spent nothing. Everything below starts there, then works through scoring, sequencing, the translate versus localise question, payments and pricing, site architecture, and what changes in your CAC once you cross a border.

Score candidate markets on six inputs, weighted toward demand you already have

Six inputs, and the first one carries roughly half the weight. Existing unpaid demand beats every other signal because it is the only one that has survived contact with your product.

Pull twelve months of signups, trials and organic sessions segmented by country, then look at paid conversion rate rather than volume alone. A market sending 800 signups that convert at 0.4 percent is telling you something different from one sending 200 that convert at 2.1 percent. The second is the market to enter.

The other five inputs are qualifying filters rather than drivers. English proficiency, which the EF English Proficiency Index tracks annually and which puts the Netherlands, Sweden, Denmark, Norway and Austria in its highest band while France and Japan sit considerably lower. Competitive density, meaning whether a well funded local incumbent already owns the category words. Payment infrastructure, which decides whether a card form is enough. Compliance burden, covering data residency, VAT registration and sector rules. And support timezone cost, which is a real headcount line the moment a market is nine hours away from your team.

Input	Weight	Where to get the data	Disqualifying answer
Existing unpaid signups and paid conversion	40%	Your own product analytics, segmented by country	Fewer than 30 signups in twelve months
English proficiency of the buying role	15%	EF English Proficiency Index plus your own support tickets	Buyers routinely ask for docs in local language
Competitive density	15%	Local SERP check on your three money keywords	A funded local incumbent owns every result
Payment infrastructure fit	10%	Stripe or Adyen local method coverage docs	Cards are a minority of B2B payment volume
Compliance and tax burden	10%	Local VAT rules, data residency, sector regulation	Data residency required and you cannot provide it
Support timezone cost	10%	Your own coverage hours versus the market's working day	Requires a new shift before there is revenue to fund it

The opinion here is blunt: expand where your product already has unpaid signups, and treat any other trigger as a vanity decision. It is possible to enter a cold market successfully, but it costs three to five times more and takes two to three times longer, and almost nobody budgets for that honestly. The SaaS marketing budget calculator is a useful reality check before you commit, because expansion tends to get funded out of hope rather than out of a line item.

The sequence most North American SaaS companies should follow

There is a well worn path, and the reason it is well worn is that each step is cheaper than the one after it. Deviating from it is fine, but do it knowingly.

The UK and Ireland come first, almost always. Same language, similar buying behaviour, minimal legal work, and a currency change plus GBP pricing usually covers the localisation requirement. Most North American SaaS companies already have UK signups they have never actively served. First revenue typically arrives inside a quarter.

DACH and the Nordics come second, and they behave differently from each other despite often being bundled. Germany, Austria and Switzerland buy in German at the mid market and enterprise level, care deeply about data processing terms, and will read your DPA properly. The Nordics buy comfortably in English, which means you can enter Sweden, Denmark and Norway with localised payment and pricing while leaving the copy alone.

France comes third and is a genuine localisation project. French buyers evaluate in French, local competitors are strong in several categories, and a translated homepage attached to English documentation converts poorly. Benelux follows, where the Netherlands is English friendly but iDEAL coverage matters enormously for any self serve motion.

Japan and Brazil are the two special cases worth naming. Both have real, underserved demand for B2B software and both punish half measures. Japan tends to require a local partner or reseller relationship and a level of documentation polish that English speaking markets never ask for. Brazil rewards Pix support so heavily that adding it can move self serve conversion more than any campaign you could run.

Translate, localise or transcreate, and what each actually costs

These three words get used interchangeably and they describe three different budgets. Picking the wrong one is the most common way expansion money gets wasted.

Translation converts meaning, sentence by sentence, at roughly 0.10 to 0.30 dollars per word from a professional vendor. It is right for documentation, help centre articles, legal terms and product UI. It is wrong for anything where persuasion matters, because translated marketing copy reads as translated marketing copy to a native speaker inside two sentences.

Localisation adapts the whole experience: currency, date formats, examples, company names in case studies, screenshots showing the local interface, and compliance references that mean something locally. Budget roughly 1.5 to 2 times the translation cost, because somebody has to make decisions rather than just convert words.

Transcreation rewrites for the market, usually by a native practitioner who understands the category, at 0.25 to 0.60 dollars per word or a flat per page fee. Reserve it for your homepage, pricing page, top three product pages and your two highest converting comparison pages. That is normally eight to twelve pages, which is affordable, and it is where the conversion difference shows up.

What nobody tells you is the maintenance cost. Every localised page is a page that goes stale when you ship a feature or change a price. A team of three shipping weekly will drift out of sync within two quarters unless someone owns the sync, which is why SaaS content localization is an operations problem before it is a translation problem. Companies that skip that ownership question end up with a German site describing a product that stopped existing in March.

Pricing, currency and payment methods, market by market

Currency conversion is not pricing. Charging 49 euros because you charge 49 dollars is an unplanned price increase of roughly 10 percent, and charging 49 pounds is a larger one. Set local prices deliberately, land them on numbers that feel natural in that currency, and remember that European list prices normally exclude VAT while the buyer sees it added at checkout.

Payment method coverage is where self serve conversion is actually won or lost. In the Netherlands, iDEAL dominates online payment to a degree that makes a card only checkout feel broken. Belgium has Bancontact. Germany leans heavily on SEPA direct debit and invoicing, and a meaningful share of German B2B buyers still expect to pay against an invoice with payment terms rather than entering a card. Brazil has Pix and Boleto. Japan has Konbini and JCB.

The merchant of record question has a clean answer for most companies. Use one early. Paddle or Lemon Squeezy will handle EU VAT registration, invoicing, collection and remittance for a take rate in the region of 5 percent, which is expensive and which removes an entire compliance workstream at the exact moment you have no one to staff it. Stripe Tax plus your own One Stop Shop registration is materially cheaper and needs a finance owner. Most teams switch somewhere between two and five million dollars of European revenue.

Site architecture and hreflang without breaking what already ranks

Subfolders, not country domains. Use example.com/de/ rather than example.de, because subfolders inherit the domain authority you have spent years building while a new ccTLD starts from nothing and needs its own link acquisition programme. There are exceptions, mostly regulatory, and they are rarer than agencies suggest.

Hreflang is a bidirectional declaration. Every language version of a page must reference every other version including itself, and if the return tag is missing Google ignores the whole set. Add an x-default pointing to your English page so that an unmatched visitor lands somewhere rather than nowhere. Most implementation failures come from partial rollouts where the English page never got updated with the new pointers.

Three practical rules that save months. Do not auto redirect by IP, because it breaks crawling and infuriates travelling users; offer a dismissible banner instead. Do not publish a language version until the core journey is complete in that language, since a half translated funnel converts worse than an English one. And keep URL slugs in the local language where the search terms differ, because /de/preise earns relevance that /de/pricing does not.

The deeper technical work, including how to handle a market where your docs site and marketing site diverge, sits in international SaaS SEO. The short version: decide the architecture before you translate anything, because retrofitting hreflang onto 200 published pages costs several times what building it correctly once would have.

How CAC and sales cycle actually change across borders

Both change, usually in the same direction, and the size of the change is predictable enough to plan around.

Paid acquisition costs less in most non US markets because the auctions are thinner. LinkedIn and Google both tend to clear at lower cost per click in DACH, the Nordics and Benelux than in North America for the same job title, sometimes 25 to 40 percent lower. That advantage is real and it is partly offset by lower volume, so you reach a ceiling faster.

Sales cycles run longer in DACH and Japan, typically 20 to 40 percent longer than North America at comparable contract value. In Germany the extra weeks are legal and data protection review, which is a genuine process rather than a stall, and the deal closes if you survive it. In Japan the extra time is internal consensus building across a wider group than your champion mentioned. The UK is effectively identical to North America. France and the Nordics sit in between.

Support cost is the line that gets forgotten. A European market served from a North American team means a first response time that reads as overnight to every customer, and that shows up in trial conversion before it shows up in churn. Either accept it and say so publicly, or fund coverage. Employer of record services such as Deel make the first European hire administratively straightforward, which removes the entity argument that used to delay this decision by two quarters.

What this costs when it goes wrong

Two failure modes account for most of it. The first is the translated homepage with an English product: a buyer arrives on a polished German landing page, clicks through to a trial, and hits an English interface and English documentation. Conversion from that path typically runs a third to a half of the domestic rate, and the diagnosis is usually months late because the country level numbers look fine at the top of the funnel.

The second is entering a market with a strong funded incumbent because the TAM number looked good. Category keywords are owned, review sites are dominated, and every buyer you reach has already been anchored on someone else’s pricing. That is a two year fight, not a market entry, and it belongs in a different plan.

Update your ICP definition per market rather than assuming it travels. Company sizes, job titles and buying committee shapes shift more than people expect, and a German mid market buyer with the same title as your best North American customer may sit three levels further from the budget.

What to do next

Segment your last twelve months of signups by country this week and sort by paid conversion rate rather than volume. If one non domestic market stands out, you have your answer and the rest is execution. If nothing stands out, you are not ready, and the money is better spent on the domestic motion described in the SaaS marketing plan template.

Where a market does stand out, write a one page entry plan with a budget, a named owner, a kill date and the score from the table above. Founders running this without a marketing team will find the sequencing shortcuts in SaaS marketing for founders useful, and anyone arriving into a role with expansion already committed should work through the first 90 days checklist before agreeing to the timeline they inherited. If the entry is enterprise led rather than self serve, the tradeoffs in ABM versus inbound apply directly, since a new market is the one situation where a named account list beats broad demand capture.