Michelin: From Siloed Data to a Global AI Fabric With Dataiku
Michelin uses Dataiku to scale AI across R&D, manufacturing, and services, unlocking data-backed insights across 50+ global manufacturing sites.
Learn More
time saved by reducing manual data gathering and ad-hoc analyst requests
Euronext is the leading European capital market infrastructure and covers the entire capital markets value chain, from listing, trading, clearing, settlement and custody to solutions for issuers and investors. Inside the business, teams across product, sales, and analytics rely on market share data to monitor performance, explain movements, and support decisions that quickly move up the management chain.
While dashboards cover recurring reporting needs, many day-to-day questions require exploration beyond standard views — slicing by new dimensions, testing hypotheses, or investigating unexpected shifts in market share. Answering them requires custom analysis, deep knowledge of the data model, and time from a small group of data analysts. Business teams must translate questions into technical requests — and then wait for results to come back.
Euronext wanted to make these insights more accessible, without compromising accuracy or governance. The goal was simple: Help more teams get reliable answers faster, reduce repetitive manual work for analysts, and build a scalable foundation that prevents technical debt over time.
By embedding a market share analytics agent directly into existing workflows, Euronext enabled business and analytics teams to work together in a new way — combining self-serve access, orchestrated intelligence, and built-in governance to accelerate decisions without sacrificing trust.
More people supported — with analysts still in control: The product team can now ask and get answers to market share questions directly through natural language, while the analytics team maintains ownership of definitions, calculations, and validation logic. This expands access to insights without creating inconsistency or eroding confidence in the numbers.
Orchestrated intelligence — from question to insight in seconds: Instead of relying on manual data pulls and ad-hoc requests, the agent translates business questions into governed queries and retrieves results from trusted datasets automatically. This shifts day-to-day market share exploration from analyst-led support to a faster, workflow-driven model that fits how teams already operate.
Governance at scale — visibility, traceability, and trust built in: Every response is backed by transparent query logic and traceability, giving users confidence in what they’re seeing and giving analysts oversight into how insights are produced. This ensures the organization can scale access responsibly — balancing speed with accuracy, control, and explainability.
Faster answers, less manual effort: Questions that once took hours can now be resolved in seconds. For the product team and supporting analysts, this translates into up to a 20% reduction in time spent on recurring market share queries — freeing analysts to focus on higher-value work.
Many of the product team’s most common market share questions sit just outside standard dashboards. Even small variations — a different time period, scope, or market cut — previously required the contribution of a skilled data analyst, slowing decisions and pulling analysts into repetitive work.
To make this possible, Euronext built a Market Analytics Assistant Agent in Dataiku — tailored to the firm’s own market share logic, business definitions, and ways of working — so teams could get accurate, market-specific answers, not generic interpretations, without manual analyst intervention.
The agent translates questions into governed SQL queries, runs them on trusted datasets, and returns structured answers with full traceability — including a clear summary of how it understood the request and the business parameters it applied (time period, scope, market cut, definitions), so users can see exactly how results were produced.
Previously, delivering both speed and this level of transparency was nearly impossible — answering these questions required manual analyst work, and the logic often lived in individual expertise rather than reusable, visible workflows. Now, users can validate outputs and see exactly where numbers come from, while analysts stay in the loop to monitor, refine, and ensure confidence in every result used for decision-making.
Questions that previously depended on analyst availability and prep time can now be resolved in seconds — giving business teams always-on access to trusted market share insights, while analysts spend less time on routine requests and more time on higher-value analysis, with the same level of rigor and confidence in the results.
Beyond efficiency gains, the agent transforms day-to-day market responsiveness into a strategic advantage. Teams can detect shifts as they emerge, investigate market signals in minutes, and act earlier when something looks off — informing faster, higher-confidence decisions that directly impact commercial performance, market positioning, and market share outcomes.
By making it easier to explore market share data quickly and consistently, teams can investigate trends as they emerge, reduce reliance on manual interpretation, and respond earlier when something looks off.
Market share questions come from many roles: product, sales, management. The agent in Dataiku helps us get answers directly without the support of a data analyst or advanced data expertise.Thomas Lagrange Head of Global Analytics, Euronext
Euronext’s experience shows why Dataiku is an enterprise-grade platform for AI agents — not as isolated chatbots, but as part of a governed analytics platform.
Key differentiators include:
The agent in Dataiku reduces the time analysts spend on repetitive questions, so they can focus more on higher-value analysis and improving the quality of insights.Thomas Lagrange Head of Global Analytics, Euronext
Euronext is using the Market Analytics Assistant Agent as a blueprint for scaling AI across the organization. The approach is deliberate: Start with a high-value, high-visibility use case, earn trust through transparency and evaluation, and then expand with confidence.
As answer quality stabilizes and adoption grows, the project will extend beyond the product team to other asset classes and business units. While the underlying data will differ, the same agent pattern — natural language access, analyst-controlled logic, and transparent outputs — can be reused to accelerate delivery and avoid rebuilding from scratch.
At the same time, Euronext continues to consolidate analytics and AI work into a single, governed platform. By reducing tool fragmentation and strengthening access controls, teams can move faster from question to insight while maintaining the rigor required in a regulated environment.
Across all areas of the bank, Standard Chartered is accelerating the development of AI solutions, creating a culture of decision making driven by analytics and unlocking the value of data to power better business outcomes.
Read moreMichelin uses Dataiku to scale AI across R&D, manufacturing, and services, unlocking data-backed insights across 50+ global manufacturing sites.
Learn MoreUncover how Mitsubishi Electric centralized analytics and expanded AI access to improve decision quality while reducing operational complexity.
Learn MoreRoche enhanced patent work with Dataiku, using GenAI and the Themis PatAI agent to unify tools, reduce costs, and boost efficiency for attorneys and paralegals.
Learn MorePrologis democratized AI with Dataiku, leveraging AutoML features to enable analysts to build and deploy models with no coding knowledge.
Learn More