Keyword research fails the moment you cross a border. CSA Research surveyed 8,709 shoppers across 29 countries and found 40% will never buy from a website in another language. That is the market most businesses hand to their competitors.
The usual fix does not work. You translate your list, publish the pages, and wait. Months later they rank for nothing. The words were correct, but nobody there types them.
This guide shows you how to research one market at a time. You will learn how to pick your first country, build a local seed list, and read the numbers properly. It also covers what to do when your tool shows zero volume.
Key Points
- Translation is not keyword research. A correct translation can still be a phrase nobody searches, so check every term against country-level data.
- Target a country and a language together, never just a language. Spain and Mexico both speak Spanish but search with different words.
- Start with 2 or 3 markets, not 10. Score each country on demand, competition, buying power, conversion fit, and whether you can actually serve it.
- Zero search volume usually means missing data, not missing demand. Tools cover small markets like Norway and Denmark poorly, so check Search Console and local competitors instead.
- Give every page one primary keyword per market. Reusing a keyword across two country pages makes them compete, and Google picks one.
- Track each country separately and allow 6 to 12 months. A global average hides which market is growing and which is flat.
What Is International Keyword Research?
International keyword research is the process of finding the exact words people in each country type into search engines. It looks at one market at a time. Instead of one global keyword list, you build a separate list for every country you target.
The difference matters because search habits change across borders. Two markets can share a language and still search in completely different ways. So a keyword that works in one country can bring you almost nothing in another.
Why Translating Your Keywords Doesn’t Work
Translating your keywords gives you the right words but the wrong search terms. The translation is correct. It just isn’t what people type.You optimise a page for a phrase with no demand. The page ranks, but nobody searches it.
A German study shows the problem. “Make your lips look bigger” gets 20 searches a month. “How to make your lips bigger” gets nearly 1,200.Same meaning, same language, sixty times the traffic. Translation tools miss this because both versions are grammatically correct.
It happens with single words too. German has “projektor” and “beamer” for the same device. Germany uses “beamer.” A translator hands you the other one.So start with translation, then check every word against real data. Never publish on a translated list alone.
Language and Country Are Two Different Things
A language is not a market. Spain and Mexico both speak Spanish, but they search differently. Your keyword research has to target a country and a language together.This is also how hreflang works. You tag a page as es-ES or es-MX, never just “Spanish.”
One Language, Several Countries
The same word changes across borders. Spain searches “ordenador” for a computer, Mexico searches “computadora.” The UK says “trousers,” the US says “pants.”Traffic without conversions is the usual warning sign.
When People Search in English Anyway
Some markets search in English even when it isn’t the local language. This is common in the Nordics and for technical terms everywhere.One agency targeting Norway found Norwegian keywords had lower volume but converted better, because the local language builds trust.
Step 1: Choose Which Countries to Target First
- Pick 2 or 3 countries, not 10. Spreading your budget thin gets you weak content everywhere and rankings nowhere.
- Choose the market first. Open the keyword tool second.
- Score each country on five things: demand, competition, buying power, conversion fit, and whether you can actually serve it. Rate each 1 to 5 and compare totals.
- That last point stops most bad decisions. High demand means nothing if you cannot ship there or handle local rules.
- Then do rough maths. Multiply monthly searches by 30% clicks, your conversion rate, and order value. A Norwegian keyword with 200 searches and a 900 NOK order is worth around 19,000 NOK a year.
- One shortcut before you start. Open Google Search Console and filter the data by country. You may already rank somewhere you never targeted, and that market is the cheapest place to begin
Step 2: Build a Seed List for Each Market
A seed list is 10 to 20 core terms that describe what you sell in that market’s own words. Build a separate one for every country, and start from how locals talk instead of your English list.
Pull your terms from native speakers, local competitor page titles, Google autocomplete with the country set, related searches, and your own Search Console data. Competitor titles are fastest, since someone already did the keyword research for you.
What If You Don’t Speak the Language?
Use AI to generate candidates, then check them against real data. It suggests words well, but cannot confirm which ones people search.If the tool and your native speaker disagree, trust the tool for volume and the person for meaning.
Step 3: Check the Numbers for That Country
Set the country filter before you read any number. Global search volume tells you nothing about whether a keyword works in Norway or Mexico.
Check three things for each term: search volume, keyword difficulty, and search intent. All three change from country to country, even for the same word.
Ahrefs indexes keywords in 217 locations. Semrush and Google Keyword Planner both let you pick a country and language. Pick both, not only one.
Most people find the difficulty surprising. A term that is brutal in English is often easy in Norwegian or Dutch, because fewer sites compete for it.
What to Do When the Tool Shows Zero Volume
Zero volume usually means missing data, not missing demand. Tools cover small markets like Norway, Denmark, and Finland badly.
Check Search Console impressions, local competitor headings, and Google Ads forecasts instead. One SEO specialist puts it simply: ask your customers, and mine your email threads for the questions they actually send.
Step 4: Look at the Actual Search Results
- Search your keyword in the target country and check what comes back. Numbers tell you demand exists. Only the results page tells you what people mean.
- Use a VPN, or add the country parameter to your Google URL. Then read the top 10 results before you commit to the term.
- The word “chips” shows why this matters. Search it in the US and you get potato crisps. Search it in the UK and you get hot fries.
- Ask three things as you scan the page. Do the results match your product? Are they blog posts or product pages? Are competitors selling, or explaining?
- If the results look nothing like what you expected, drop the keyword. You found a word that means something else there.
- The format tells you what to build. Comparison guides ranking means buyers research first. Product pages ranking means they are ready to buy.
Step 5: Map Your Keywords to Pages
Give every page one primary keyword, and never reuse that keyword on another page. This is what stops your own pages from competing with each other.
Build a simple sheet for each market, with one row per page. Include these columns:
- URL of the page
- Primary keyword (unique to that page)
- Supporting keywords (these can overlap)
- Search intent (informational, commercial, or transactional)
- Monthly volume for that country
The risk grows when two countries share a language. Your UK page and your US page can end up chasing the same term, and Google picks one.
Check seasonality before you finalise the map. Demand peaks at different times in different countries, so publishing dates should follow the local calendar.
Which Tools to Use for What
No single tool covers every market well, so pick one by the job you need done. Most people pay for features they never use.
Here is what each tool is actually good at:
| Tool | Best for | Price |
| Google Keyword Planner | Country-level volume, free | Free with a Google Ads account |
| Semrush | Related keywords and position tracking | From $139.95/month |
| Ahrefs | Competitor keywords, 217 locations indexed | From $129/month |
| Google Search Console | Your own country data | Free |
| Google Trends | Comparing demand between countries | Free |
Start with the free two. Search Console shows where you already rank, and Keyword Planner gives country volume without a subscription.
How Keywords Decide Your Site Structure
Your keyword map tells you how many country versions you need, and that decides your URL structure. Build the structure after the research, not before.
You have three options:
| Structure | Example | Best for |
| Subfolder | site.com/no/ | Most businesses. Easiest to set up and manage. |
| Subdomain | no.site.com | Separate teams or different platforms per market |
| ccTLD | site.no | Big budgets and strong local presence |
Subfolders win for most small businesses. You keep all your authority on one domain instead of splitting it across several.
Add hreflang tags once the structure is live. These tell Google which version to show a Norwegian user and which to show a German one.
One thing hreflang cannot do is fix a bad keyword. It controls which page appears, not whether that page targets a term people search.
Put the local keyword in the slug too. A Norwegian page should read site.com/no/tannlege, not site.com/no/dentist.
How to Tell If It’s Working
Check every market separately, never as one global number. Filter Search Console by country, track rankings with the tool set to that market, and segment your analytics the same way.
Watch conversions, not only traffic. Traffic without conversions usually means you picked the wrong market, and give any new country 6 to 12 months before you judge it.
Common Mistakes and How to Spot Them
Most international SEO problems show up as traffic without sales. The symptom is easy to see once you know what it points to.
Here are the five that cause the most harm:
- Using one keyword list everywhere. Symptom: pages rank but get almost no clicks in some countries.
- Chasing volume over intent. Symptom: plenty of visitors, almost no enquiries.
- Two country pages targeting the same term. Symptom: your UK page shows up for US searches, or neither ranks well.
- Skipping small markets. Symptom: you never checked Norway or Denmark because the tool showed low numbers.
- Doing the research once. Symptom: rankings slide slowly over 12 months and nobody knows why.
What It Costs and Who Should Do It
Budget 10 to 20 hours for your first market, plus a native speaker to check your list. The time costs more than the tools do. One subscription and two free tools will cover most of the work.
Do the research yourself, then pay someone local to review the final 30 terms. That review is where the money belongs, because a native speaker catches meaning problems no tool will flag.
Hire a freelance native SEO for one or two markets. Move to an agency only when you run four or more and need several languages at once. Rates vary a lot by market, so get two or three quotes before you set a budget.
Start With One Market
Pick one country and do it properly before you add a second. One market done well beats five done halfway. You also learn the process on a small scale, where mistakes cost little.
The principle behind all of this is simple. Research the market, do not translate your list. Everything else follows from that.
Open Google Search Console this week and filter by country. If you already get impressions somewhere you never targeted, that is your first market.
FAQs
Can I just use Google Translate for my keywords?
No, Google Translate gives you a grammatically correct word, not the term people search. Run every translated keyword through a tool with the country filter set before you use it.
Do I need a native speaker?
Yes, at least for a final review. Tools give you volume, but only a native speaker catches meaning errors like “projektor” versus “beamer” in German. Paying someone to check 30 terms is enough for most small businesses.
How do I check search volume for another country?
https://www.semrush.com/free-tools/keyword-search-volume-checker/Set the country filter inside your keyword tool before reading any number. Google Keyword Planner, Semrush, and Ahrefs all let you pick a country and language. Ahrefs indexes keywords across 217 locations.
Can AI replace a native speaker for keyword research?
No, AI generates useful candidate terms but cannot confirm which ones people actually search. Use it to build a list, then validate every term against real data and a native review.
How much does international keyword research cost?
Budget 10 to 20 hours for your first market plus a native speaker review. The time costs more than the tools. A single subscription and two free tools cover most of the work.
How many countries should I target at once?
Two or three at most. Spreading budget across more markets produces weak content everywhere and rankings nowhere. Add the next market only after the first one performs.
Conclusion:
International keyword research works when you research each market instead of translating your existing list. Pick two or three countries, score them on demand and your ability to serve them, then build a separate seed list from native speakers and local competitors.
Check volume, difficulty, and intent with the country filter set, look at the real search results before committing, and give every page one primary keyword so your own versions never compete. Track each market on its own and allow 6 to 12 months before judging it.
Open Google Search Console this week, filter by country, and start with the market where you already get impressions.
22 October 2026
