What If You Could Have a Conversation With Your Data?

Most organisations don’t lack data—they lack timely, clear answers when someone asks “what happened?”, “why?”, and “what should we try next?”. New AI tools make it tempting to think we can simply “talk to our numbers.” That can help—but only if we’re clear about what makes an answer reliable and usable in a real meeting, not just fast on a screen.

This session starts from one question: What if getting insight felt more like a good back-and-forth with a colleague—where you can refine the question, compare time periods, and understand the “so what?”—without needing to be an analyst?

We’ll separate helpful uses of AI (clarifying questions, explaining changes in plain language, suggesting what to look at next) from risky ones (answers that sound confident but rest on shaky definitions or missing context). You’ll hear what leadership and teams should ask for so “conversation with data” builds trust, not noise: agreed definitions for key numbers, clear ownership of who can see what, and transparency about what the tool can and can’t know.

You’ll leave with a practical way to talk about this inside your company—whether you’re in sales, marketing, operations, finance, or general management—and a short checklist for evaluating vendors and internal pilots without getting lost in technical detail.

Elizabeth Hanson is Head of Engineering, Data at DeepL and has been leading data and engineering teams for the last several years. At DeepL she leads the data organisation; previously she was Director of Engineering at BenchSci and Technology Manager then Senior Technology Manager at Just-Eat Takeaway.com. Her focus is on building and scaling data teams that deliver reliable, self-service analytics and align technical investment with business outcomes. She is based in Bath, UK, and holds a PhD in molecular evolution from the University of Bath.

Know what’s Happening Next — before everyone else does.

Get App

Host or Publisher TechSPARK