Kadence RH, a 22-person HR technology startup based in Paris's 2nd arrondissement, had developed a workforce scheduling platform aimed at mid-market retail and hospitality companies in France. With 2.3 million euros in seed funding closed in late 2024, the team was executing a deliberate go-to-market strategy: content marketing, founder-led LinkedIn presence, and strategic partnerships with French retail associations. Their problem was not awareness. It was that their awareness was not reaching the buyers who mattered most.
A sales team debrief in February 2026 revealed a pattern: multiple enterprise prospects had arrived at Kadence only after first evaluating a competitor they had discovered through an AI assistant recommendation. In one case, a prospect had asked ChatGPT for HR scheduling software for French retail chains and had been given a shortlist of four tools. Kadence was not on it despite being directly relevant and technically superior to two of the four listed alternatives. The founder recognized this as a structural problem requiring a structural solution.
AISOS was engaged for an 80-day implementation targeting the French market. The challenge was twofold: building AI visibility in a competitive, well-funded category while doing so in a way that respected the specific linguistic and market context of France. Our guide on AI visibility for SaaS companies and our AI visibility overview provided the framework. For Paris-specific context, see our Paris AI visibility page.
The Challenge
The HR technology category in France is crowded and well-funded. Several of Kadence's direct competitors had raised Series A and B rounds and were investing heavily in content production. Two had English-language content that ranked globally and French-language content that dominated locally. Kadence's content library was strong on product education but weak on the comparison and recommendation signals that AI systems use to make category selections.
An AI visibility audit at the start of the engagement showed Kadence appearing in 3 of 36 relevant queries (8.3%) across ChatGPT, Perplexity, Claude, and Mistral (relevant for the French market). The three best-performing competitors appeared in 24 to 31 of 36 queries. The gap was not primarily attributable to brand awareness or review volume. It was attributable to content structure, third-party signal presence, and schema implementation. Kadence had none of the technical AI visibility infrastructure that its competitors had, even partially, assembled.
A specific challenge in the French SaaS market was the importance of Mistral-based AI responses. Mistral's models are increasingly used by French enterprises and government-linked platforms. Their training data emphasizes French-language sources and French institutional credibility signals. An AI visibility strategy for a French startup needed to include Mistral-specific optimization alongside the standard ChatGPT and Perplexity tracks. Understanding AEO applied to multi-model environments shaped the technical approach from the outset.
The AISOS Strategy
The strategy was organized around three competitive advantages Kadence genuinely possessed: strong customer outcomes in retail scheduling, a robust native French onboarding experience, and a partner network with three major French retail associations. Each of these advantages was invisible in AI responses at the start of the engagement. The implementation was designed to make them visible and machine-readable.
Phase one was entity definition and schema deployment. Kadence had no structured company entity data beyond a basic Organization schema. AISOS created a comprehensive SoftwareApplication schema covering use case definitions, supported industries, geographic service scope, and integration ecosystem. A Wikidata entry was created referencing the company, its founders, and its primary use case. French-language structured data was explicitly included to improve Mistral indexing. An llms.txt file was deployed in both English and French.
Phase two was comparison content. The category where Kadence was weakest in AI responses was the "compare X versus Y" query type. When buyers ask AI assistants to compare HR scheduling tools, they receive structured comparison information. Kadence had no comparison content. AISOS built six structured comparison pages targeting the six most common competitor pairs in AI queries, each formatted with explicit criteria tables, use-case differentiation, and verified outcome data. These pages included links to relevant industry pages and to the AI SEO checklist. Phase three placed Kadence in eight French-language SaaS directories and three HR technology editorial publications with confirmed AI sampling rates.
The Results
By day 80, Kadence appeared in 21 of the 36 original audit queries (58%), up from 3 (8.3%). The improvement was most pronounced on Perplexity, where the company went from 1 to 10 appearances in 12 relevant queries. ChatGPT showed 8 of 12, up from 2. Claude showed 6 of 12, up from 0. Mistral showed 5 of 12, up from 1, representing the most strategically important improvement given the French enterprise market context.
Pipeline impact was measurable within 45 days of the first content publication. Inbound demo requests from French enterprise buyers increased by 47% over the prior equivalent period. Average deal size from this new segment was 31% higher than the SME baseline, consistent with enterprise buyers arriving pre-educated through AI research. Four enterprise pilots were initiated within the measurement window, with a combined annual contract value of 212,000 euros if all convert. Two had progressed to contract negotiation by the end of the engagement.
The comparison pages generated an unexpected SEO benefit: three of the six pages reached page-one Google rankings for their target queries within 50 days. The disciplined, factual structure required for AI readability produced content that performed equally well in traditional search. Total organic traffic to the site increased by 23% over the engagement period, with comparison pages accounting for 38% of new traffic despite representing less than 2% of total page count.
Key Success Factors
The decision to build comparison pages was the highest-leverage single action in the engagement. In competitive SaaS categories, AI assistants frequently receive queries in the form "compare Kadence versus [competitor]" or "what is the best HR scheduling software for French retail." These queries require comparison-structured content to answer well. Businesses that provide this content get cited. Businesses that do not get mentioned generically or excluded entirely. Kadence's willingness to publish direct, factual comparisons rather than generic feature lists was the primary driver of the speed of improvement.
The Mistral-specific optimization track provided a genuine competitive moat. None of Kadence's competitors had invested in Mistral-specific AI visibility at the time of the engagement. By deploying French-language structured data and ensuring presence in French-language sources with known Mistral sampling rates, Kadence built a position in a model that is increasingly important for French enterprise buyers. This moat will persist until competitors replicate the approach. For startups in France, Mistral visibility is an underexploited competitive lever that AISOS specifically addresses. See our AI visibility glossary entry for a discussion of model-specific optimization.
The founder's direct involvement in approving comparison content was critical. Comparison pages require precise, verified claims about competitor products. The founders of Kadence had firsthand knowledge of the competitive landscape and could approve or correct claims quickly. Startups with direct founder involvement in content accuracy routinely produce better AI-cited content than those where content is delegated entirely to marketing teams without subject matter authority. Speed and accuracy of content approval is one of the strongest predictors of engagement outcomes.
Lessons Learned
The most important lesson from the Kadence engagement is that in competitive startup categories, being technically superior is insufficient for AI visibility. AI assistants recommend based on the signals they can read, not on actual product quality. A startup with better technology but weaker AI signal infrastructure will consistently lose AI recommendations to competitors with stronger signals but inferior products. Building AI visibility infrastructure is not a marketing nice-to-have. For startups competing against better-funded rivals, it is a strategic equalizer.
The bilingual AI visibility requirement for French companies adds meaningful complexity but also provides an advantage for companies willing to invest in it. Most AI visibility guidance is written in English for English-language markets. French companies that adapt the methodology to include Mistral-specific and French-language source strategies gain access to a layer of AI recommendations that their English-only competitors cannot reach. AISOS has developed specific frameworks for this bilingual implementation that are not available elsewhere in the market.
Finally, the engagement confirmed that the startup funding stage affects AI visibility strategy. At seed and early Series A stages, the highest-ROI investment is comparison content and entity definition, not broad content volume. As the company grows and has more customer outcomes to document, structured case studies become the next highest-leverage asset. Early-stage companies should sequence their AI visibility investment in this order rather than attempting to replicate the full-scale programs of Series B and C companies. Contact AISOS to design the right sequence for your current stage and competitive context.