French Keyword Research: Tools and Techniques | French SEO Blog

Keyword research in English is well documented: plug seed terms into a tool, export volumes, cluster by intent, map to URLs. French keyword research follows the same skeleton but fails if you treat it as translation with a .fr flag. Volumes differ, synonyms diverge, diacritics matter for user experience (if not always for matching), and Search Console data for France reveals query shapes that English tools never surface.

I spend more time correcting bad French keyword lists than building them from scratch for new clients. The errors are predictable: English seed keywords run through automated translation, search volume pulled from the wrong locale, and intent assumptions imported from US SERPs that look nothing like Google.fr. This guide covers the tools and techniques we use in Paris to build keyword strategies that reflect how French people actually search.

Setting up for accurate French data

Before opening any tool, configure locale correctly.

Google Ads Keyword Planner: Set location to France (or specific regions if you target Île-de-France only). Language: French. Searching in English returns misleading or empty data for French terms.

Google Search Console: Filter performance by country France. Export queries quarterly—this is your ground truth for what already earns impressions. GSC shows accent variants and misspellings users actually type.

Google Trends: Compare terms with geo set to France. Essential for seasonal terms (bûche de Noël, maillot de bain) and validating whether a synonym is rising or dying.

Third-party tools (Semrush, Ahrefs, Moz): All support French databases; verify the database refresh date. Ahrefs’ FR index is strong for backlink and keyword data; Semrush excels at PPC overlap and position tracking for .fr domains. No single tool is authoritative—cross-reference volumes above 1,000 monthly searches.

Critical rule: if a tool shows identical volume for English and French versions of a term, be sceptical. Real French markets rarely mirror English this cleanly.

Seed keyword generation: start French, not translated

Effective seeds come from:

Client sales and support transcripts (French calls reveal exact phrasing)
Competitor .fr sites and their GSC-visible rankings via tool estimates
Amazon.fr, Cdiscount, and Fnac autocomplete suggestions
Google.fr autocomplete, People Also Ask, and related searches
French forums (Doctissimo for health, Hardware.fr for tech, JeuxVideo.com for gaming)
Industry glossaries and regulatory vocabulary (especially B2B and medical)

Avoid starting from English seeds unless you are validating a global brand term with established French search demand.

Technique — the bilingual divergence audit: List top English keywords driving traffic in UK/US properties. Research each concept independently in French rather than translating. Document cases where the French winner differs: cell phone → téléphone portable or smartphone; vacation → vacances not voyage for leisure travel in many contexts.

Understanding French query morphology

French compounds search terms differently.

Adjective-noun order affects queries: chaussures running vs running shoes word order
Articles and prepositions appear in long-tail: comment choisir un, meilleur de, avis sur
Elision and contractions in speech-inspired queries: l’assurance auto vs assurance automobile
Anglicisms coexist with French equivalents: marketing and mercatique; email and courriel—often both have volume
Regional variants: péter vs exploser in slang contexts; soixante-dix vs septante in Swiss-influenced border regions

For keyword lists, include accent and non-accent variants where users might type without accents, but use proper accents in on-page copy and titles.

Search intent mapping on Google.fr

SERP analysis must use Google.fr, physically or via reliable proxy. Intent categories align globally—informational, navigational, commercial, transactional—but SERP composition differs.

French informational queries often surface:

Wikipedia.fr (dominant for definitions)
Government sites (service-public.fr, ameli.fr)
Established French media (Le Monde, niche vertical publishers)

Commercial investigation favours:

Comparison sites (LesFurets, Comparateur- verticals)
Forum threads and YouTube reviews
Amazon.fr product pages even for research-phase queries

When you see forum and Reddit results on page one, content must address scepticism and detailed experience—not sales pages.

Intent labelling technique: For each keyword cluster, search incognito on Google.fr, note dominant content type, and tag intent accordingly. If top results are all PDF white papers, a blog post will struggle without equivalent depth and authority.

Tools deep dive: what each does best

Google Keyword Planner

Free with Ads account. Best for rough volume ranges and discovering close variants. Weak for organic difficulty assessment. Use for PPC-informed SEO when clients run Google Ads in France—search term reports reveal converting queries organic strategy missed.

Ahrefs and Semrush

Ahrefs offers strong French keyword and backlink data, Content Gap analysis, and SERP feature tracking. Semrush excels at position tracking for Google.fr with regional targeting, Keyword Magic Tool filters, and Topic Research for French question ideas. Use both for cross-validation on high-priority terms.

AlsoAsked and AnswerThePublic

Visualise People Also Ask trees in French. Excellent for FAQ architecture and voice-adjacent question research. Export question hierarchies directly into content briefs.

Clustering and prioritisation framework

Raw keyword lists overwhelm clients. We cluster using this hierarchy:

Primary cluster — head term + close variants sharing URL target
Supporting long-tail — FAQ and blog targets linking to primary
Local modifiers — city/department appendices (plombier Lyon 3)
Brand vs non-brand — separate reporting lanes

Prioritise by:

Business relevance (margin, strategic focus)
Achievable difficulty (referring domains of top 3 French results)
Current striking distance (GSC positions 4–20)
Seasonality (Trends validation)

Score each cluster 1–5 on relevance and achievability; plot on a simple matrix. High relevance, high achievability clusters launch first.

Competitive keyword gap analysis

Identify 3–5 French competitors—not global English sites ranking incidentally in France. Export their top organic keywords with minimum volume 50, excluding branded terms. Group gaps by whether you lack content, authority, or the correct page type.

Validating with native speakers

Tools miss nuance. Before finalising a target keyword list:

Native French copywriter reviews for unnatural phrasing
Sales team confirms terms match customer language
Paid search test (optional): small Google Ads campaign on candidate terms measures CTR and conversion efficiency faster than waiting months for organic proof

One hour of human review prevents months of optimising for terms nobody uses.

Common failures and fixes

Identical English-to-French keyword maps fail—research each concept independently. Verify tool country settings show France, not global data. Target zero-volume GSC terms with proven impressions. Match SERP content type before chasing meilleur or comparison keywords. Standardise accents in titles while accepting variant queries in reports.

From keywords to content briefs

Each cluster becomes a brief with primary and secondary keywords, observed SERP content type, target URL, FAQ question variants, and tone guidance (vous/tu).

French keyword research is not a one-time spreadsheet—it is a quarterly refresh tied to GSC reality, seasonal shifts, and competitive movement. Tools accelerate discovery; locale discipline and native validation ensure you optimise for France, not a translated fiction of your English strategy. Do that, and the keywords you target will be ones French users actually type, speak, and click.