International PPC: Running Campaigns in Markets You Don't Speak the Language Of

Take the campaign structure that works at home, translate the keywords, translate the ads, and launch in Germany. The campaigns will spend. They'll even convert a little. Within two quarters someone concludes that "Germany is expensive," and the budget drifts back to the home market. This is the most common way international PPC dies, and it has almost nothing to do with strategy. The mechanics of paid search are the same everywhere. Every input that matters is not: the queries people type, the copy that earns the click, the page that converts it, and the competitors bidding against you are all stubbornly local.

Junto is a French growth agency. We run cross-border campaigns for European brands entering markets nobody on the client side speaks, and we watch international brands make textbook mistakes entering France, our home turf, in the other direction. Both vantage points teach the same lesson: language is not a translation task you bolt onto a campaign at the end. It is the campaign.

Why translated international PPC campaigns underperform

People don't search in translation

A translated keyword list is a list of phrases nobody types. Translation gives you grammatically correct terms; keyword research gives you the terms a real buyer uses, and the two overlap far less than you'd hope.

French finance teams don't look for a literal rendering of "expense management software." They search for "logiciel de notes de frais," built around a French administrative concept with no clean English equivalent. German buyers compress entire product categories into single compound words a translator would never produce unprompted. Dutch and Nordic searchers mix English loanwords into local-language queries, so the highest-intent term is sometimes half English, half not, in a proportion you can only discover in-market.

Match types make this worse. Broad match leans on Google's semantic understanding of the language, and that understanding is thinner outside English. In smaller language markets, broad match wanders further from your intent, and the search term report fills with queries nobody on your team can evaluate. If no one reads the language, junk traffic doesn't get caught. It gets rationalized as "market learning."

Ad copy nuance is the whole job

The headline that wins your home auction carries assumptions that don't travel. Formality is the obvious one: German forces a choice between Sie and du, French between vous and tu, and either choice signals something about your brand before the reader processes a single benefit. Promotional culture is another. Aggressive percent-off framing that reads as normal in one market reads as desperation in a premium segment elsewhere.

Then there's the mechanical problem: character limits that were comfortable in English get crushed by German compounds and expansive Romance-language phrasing. A faithful translation often literally doesn't fit, so someone truncates it, and the nuance you paid a translator for dies in the interface.

The fix isn't better translation, it's transcreation: rewriting the ad from the value proposition up, in the target language, by someone who knows what claims land in that market. The same discipline applies to organic content, which is why we treat SEO translation as a rewriting exercise rather than a conversion of words.

The landing page has to finish the sentence

Message match doesn't survive a language switch. A German ad pointing at an English checkout tells the visitor exactly one thing: this company hasn't really arrived here yet. The visitor then does a quick risk calculation about delivery, returns, and support, and closes the tab.

Localization on the landing side goes beyond copy. Local payment methods matter enormously: a Dutch shop without iDEAL or a Belgian one without Bancontact is leaving conversions on the table regardless of ad quality. So do delivery promises in local terms, prices in local currency with the tax treatment buyers expect, a locally formatted phone number, and the trust signals that market actually recognizes. Your ads inherit the credibility of the page they point to. In a foreign market, that credibility starts at zero.

How to structure an international PPC account

Structure is where most cross-border accounts quietly sabotage themselves, usually by confusing language with market.

• A language is not a market. French-language campaigns can serve France, Belgium and Switzerland, but those are three different markets with different competitors, price sensitivity, purchasing power and buying conventions. Sharing campaigns across them because the ads "work in all three" means sharing budgets, bids and data across audiences that behave differently. Per-market campaigns are the default; share only when you've verified the markets actually behave alike.

• Split for control, consolidate for signal. Smart Bidding needs conversion volume to work with. Slice one budget across twelve country campaigns and every one of them starves. The honest answer is a tension to manage, not a rule: separate the markets that genuinely diverge on offer, pricing or competition, consolidate the ones that don't, and restructure as volume grows. Getting this judgment right is a large part of what a specialist Google Ads agency is actually for.

• Some decisions harden early. A Google Ads account's currency and time zone are set at creation, and currency can't be changed afterwards. Decide early whether markets will live in one account or several, because unwinding that later is genuinely painful.

• Ecommerce multiplies the surface area. Shopping and Performance Max run on your feed, so every market needs feed data in the local language, local currency and local availability. A weak feed that limps along at home fails loudly abroad, because automated campaign types amplify whatever you give them. Feed localization is unglamorous work, and it's where a specialist ecommerce PPC agency earns its keep on international accounts.

Reading competition in a language you can't read

The auction will tell you a lot even before the language does. Auction insights shows who overlaps with you and how hard they push, regardless of what language their ads run in. But the names mean nothing without local context: is that unfamiliar domain a premium incumbent, a discounter, an affiliate comparison site, or a marketplace reseller? Each one calls for a different response, and misreading a discounter as a peer will wreck your bidding logic.

Two habits help. First, actually read the local ads: Google's Ads Transparency Center lets you inspect what competitors run in each market, and a native speaker can tell you in an hour what positioning conventions dominate, what offers are table stakes, and where the gap is. Second, resist interpreting CPC differences as verdicts. Click prices reflect local auction density and local unit economics. An expensive market may simply be a rich one where everyone has done the math.

Local rules shape the auction too. France is a good case study: the Toubon law requires French in commercial advertising, and the twice-yearly soldes run on government-fixed dates that reshape promotional calendars across every vertical. Walk in without knowing this and your Q1 promo plan collides with a national institution. We've covered the specifics in our guide to advertising in France, and every market has its own version of that list.

Measuring international PPC across currencies and markets

Cross-border measurement fails in two directions: reporting that hides differences, and targets that ignore them.

Report in one currency so the business can read the numbers, but judge each market in local context. A single blended ROAS target across markets is a machine for killing new ones: your mature market, with brand recognition and years of optimization behind it, will always look better than a market you entered eight months ago. If the blended number rules, budget flows to the mature market and the expansion strangles itself. Set market-level targets that reflect stage and strategic intent, not just last quarter's efficiency.

Then correct for the economics underneath. Identical revenue in two markets is not identical profit: VAT rates differ, shipping and payment costs differ, and return rates vary enough by market and category to reshape true margin. Consideration cycles differ too, so conversion lag in a new market can make performance look worse than it is if you read the data too early. And when exchange rates move, spot-rate reporting can manufacture a "performance shift" that is really just currency. None of this is exotic analytics. It's a checklist, but someone has to own it.

What this looks like when it's done properly

The pattern behind every international account we've scaled is the same: native-language judgment installed inside the operating loop, not consulted once at launch. Transcreated ads instead of translated ones. Search term reviews run by someone who actually reads the language, on a schedule. A glossary of terms the brand uses and avoids in each market. A feedback loop from local sales or support back into keywords and copy.

That loop is what moves numbers. For Expensya, an expense-management SaaS scaling across France, Spain and Germany, per-market restructuring and localized creative cut CPA by 60% across the three countries while lead generation rose 130%. For Getfluence, a branded-content marketplace, entering Spain, the UK and Italy with market-specific campaigns tripled inbound leads. And for Swissquote, our paid media team has spent over four years coordinating campaigns across 17 countries in four geographic zones, with nine social platforms running simultaneously. Different verticals, same underlying discipline.

None of those clients staffed native speakers for every market. They borrowed the capability, kept decision-making central, and held every market to locally honest numbers.

Ready to run campaigns across borders?

If you're planning an expansion, or running one that's underperforming and you suspect the language gap is why, talk to our team. We'll look at your account structure, your localization and your per-market economics, and tell you plainly what's fixable.

Frequently asked questions

Do I need a separate Google Ads account for each country?

Usually not. Separate campaigns per market inside one account give you budget control and clean data while keeping conversion signal and management overhead consolidated. Separate accounts make sense when billing entities, currencies or local teams genuinely require them. What you should not do is run multiple countries through one shared campaign just because they share a language.

Can I use machine translation to localize my ads?

As a first draft, yes. As the thing you spend money on, no. Machine output misses register, promotional conventions and character-limit realities, and it can't do the part that matters most: keyword research, which has to be conducted in the target language, not translated into it. Budget for native review before anything goes live.

How quickly should I judge a new market?

Later than feels comfortable. New markets carry no brand equity, longer consideration cycles and algorithms still in learning, so early efficiency will look poor next to your home market. Judge trajectory against launch-stage targets rather than snapshots against mature-market numbers, and pre-agree the review horizon before launch so the decision isn't made in a moment of impatience.