The Data Front

Driving Innovation with Tableau Agent: Insights from a Pilot Program

Beyond the Tech: Human-Centered Learnings from a Tableau Agent Pilot

Working in Data Enablement, I’m always looking for ways to empower teams with data and drive innovation. That’s why I’m thrilled to be one of the first to pilot Tableau Agent at their organisation. Tableau Agent is an agentic AI assistant designed to revolutionise the way we interact with data in Tableau. Over a focused two-week period in December 2024, our teams embarked on a journey to explore the capabilities of this cutting-edge tool, and the results were fascinating.

Agent Onboarding: Setting the Stage for Success

David Chapman, JLP’s AE at Tableau, & I kicked off the pilot with comprehensive onboarding, ensuring the teams understood the purpose and technicalities of Tableau Agent. A dedicated Slack channel provided a space for quick troubleshooting, feedback, and knowledge sharing, fostering a collaborative learning environment. To ensure we focused on real-world applications, we integrated Tableau Agent into genuine daily workflows, avoiding any over-hyped use cases. Daily Slack prompts and weekly questionnaires helped us capture valuable qualitative and quantitative data, and a bit of friendly competition with Tableau swag kept the enthusiasm high!

Agent in Action: From Calculated Fields to Visualisations

The teams embraced Tableau Agent with gusto, using it primarily within Tableau Web Authoring for tasks like creating calculated fields (CASE, LOD, and date manipulations), building basic visualisations (bar charts, KPIs), and streamlining repetitive tasks.

Users reported significant time savings, ranging from 30 minutes to 2 hours per day, particularly for those everyday queries and tasks that might have otherwise slowed them down.

The Verdict: A Game Changer in the Making

The overall sentiment towards Tableau Agent was overwhelmingly positive. While it excelled at supporting newer users and simplifying processes, there were some challenges with advanced workflows, formatting, and context awareness. Encouragingly, Tableau has confirmed that these areas are on their roadmap for future development. With 11 out of 14 users recommending Tableau Agent and many describing its potential as a “game changer,” we’re excited to see how this tool evolves.

Lessons Learned: Fueling Future Adoption

This pilot underscored the importance of several key elements for successful implementation:

• Comprehensive Onboarding: Clear and concise training materials are crucial for ensuring users understand the capabilities and limitations of AI tools
• Dedicated Support: A collaborative space for troubleshooting, feedback, and knowledge sharing fosters a sense of community and encourages adoption
• A great team: Working alongside fellow enthusiasts & experts from Tableau to form a united support & enthusiasm front was key to maintaining the Pilot’s approach (and made the pilot that bit more fun to work on!)
• Integrated Workflows: Focusing on real-world applications and embedding AI tools into existing workflows ensures practical relevance and demonstrates tangible benefits
• Engagement Tactics: Regular check-ins, feedback mechanisms, and a bit of fun can help maintain enthusiasm and encourage thoughtful engagement with new tools

Looking Ahead: AI and the Future of Data Enablement

I believe AI tools like Tableau Agent have the potential to revolutionise the way we work with data, empowering individuals and teams to unlock deeper insights and drive greater efficiency.

While there’s still room for improvement, I’m excited to continue exploring the possibilities and shaping the future of data enablement at John Lewis Partnership – especially as Google’s Gemini is being rolled out!

What are your thoughts on AI tools in the data space? Have you experimented with Tableau Agent or other AI assistants in your workflows? Share your experiences and insights in the comments below!