Article
The AI Talk: Clément Lemariey and why building an AI strategy will separate winners from the rest
AI
February 13, 2026

Article
The AI Talk: Clément Lemariey and why building an AI strategy will separate winners from the rest
Article
The AI Talk: Clément Lemariey and why building an AI strategy will separate winners from the rest
AI
February 13, 2026
Article
The AI Talk: Clément Lemariey and why building an AI strategy will separate winners from the rest

13/2/2026
min read
AI
First, tell us a little bit about yourself and your company, Kanta.
I’m Clément Lemariey, CEO of Kanta. I started my career as a developer before moving into more strategic roles. My role today is to define Kanta’s long-term vision and strategy. I’m acting as a bridge between sales and tech teams through product management.
Kanta is a Caen-based company (Normandy) dedicated exclusively to accounting firms. Historically, we focused on anti-money laundering. Kanta has since evolved to take a broader, 360-degree approach to compliance. We are positioned as the entry point for accounting firms from the very start of the client relationship.
Can you share one concrete example where Kanta used AI to add real value for the customer?
Today, AI handles around 70% of support requests and automatically resolves about 80% of those. This frees our teams to focus on higher‑value tasks like onboarding, customer follow‑up and identifying upsell opportunities.
Customer support is a clear example. Before AI, our Customer Success teams spent a lot of time answering simple usage questions. Now, human support is reserved for complex and high-context issues where expertise truly matters.
How do you distinguish between AI hype and tools that bring value to your company?
AI is not a hype cycle like Web3, for instance. Its impact on day‑to‑day work is immediate and irreversible.
At Kanta, we’ve embraced an AI‑driven development approach: using AI not just to code faster, but to think better upfront. It helps us plan, identify blind spots, and go deeper at every stage of product design.
As technology gets smarter, which human skill matters most?
Vision and strategy. While execution becomes automated, humans are adding value by deciding what to build. We are capable of guiding AI to work on the right problems, not just using it to work faster.
What is the biggest mistake companies make when rushing into AI?
Not going far enough. Companies without a clear AI strategy risk being outpaced very quickly.
How has being part of Visma, the world’s largest SaaS network, influenced your approach to AI?
Visma is clearly ahead in how it integrates AI into products and development practices. Visma’s AI strategy helped me see AI not as a feature, but as a strategic foundation for how software companies should operate.
Visma played a key role in shaping my mindset around AI. Without the strong internal push on these topics, I probably wouldn’t have gone as deep as I did.
If a founder asked you where to start with AI tomorrow, what would you tell them?
Start with strategy right now, and most importantly, do it yourself! Founders need to build an AI mindset at the top and implement this culture with their teams. They need to understand that to get value from AI, teams must learn to provide proper context, as well as to prompt effectively. They need to understand that the real value of AI comes from structured knowledge. Teams must build a shared knowledge base to give AI the right context and prompt if effectively.
They have to map their processes. Mapping can be manual. However, identifying automation opportunities without AI quickly leads to negative ROI. In Tech, especially, if you don’t adopt AI seriously, competitors will, and you will be left behind.
About the episode
AI, interview, kanta
Voice of Visma
Welcome to the Voice of Visma podcast, where we sit down with the business builders, entrepreneurs, and innovators across Visma, sharing their perspectives on how they scale companies, reshape industries, and create real customer value across markets.
AI, interview, kanta





















































































































































































































