September meetup: In demand - data science behind predicting what sells

Leeds Data Science Meetup's September 2026 session centres on a single talk: David Hoyle, Research Data Scientist Specialist at dunnhumby, presenting "In demand: The data science behind predicting what sells and what's next for that data science."

Hoyle will walk through how demand modelling of shopper behaviour actually works — what goes into those models, what they're used for, and the specific data science, mathematics and engineering challenges involved in building them at scale for the world's largest retailers. The second half of the talk looks ahead: new applications of demand models are creating new problems, and Hoyle argues that data scientists will need to add skills to their toolkit to meet them.

Hoyle brings serious credentials. His background is in theoretical physics — PhD, then research at Cambridge, then leading machine learning groups at Exeter and Manchester. He spent 15 years in the private sector before joining dunnhumby, including a stint as joint Head of Data Science at AutoTrader UK and a period at Lloyds Banking Group. That breadth — academic rigour, financial services, automotive, now retail — shapes how he approaches the problem.

The event is hosted by Enablis at 1 York Place in Leeds city centre. Note the venue change flagged by the organisers: this is not the usual location, so check the directions on the Meetup page before you travel. Doors open at 6 pm with food on a first-come, first-served basis. The talk runs from 18:50 to 20:00.

Leeds Data Science Meetup is organised in partnership with Jumping Rivers. If you have a news item, job opportunity or want to volunteer as a speaker for a future session, contact them at lds@jumpingrivers.com or raise it in the comments section at the end of the evening.

Why attend

Demand modelling at retailer scale is a genuinely hard problem — the kind that sits at the intersection of statistics, engineering and business constraints. Hoyle has spent 35 years on exactly this kind of work and is currently doing it at dunnhumby, whose clients include some of the largest grocery retailers in the world. You'll get a practitioner's account of how these systems are built and where they fall short.

Beyond the talk, the format gives you 45 minutes of refreshments and informal networking before things kick off. The crowd is typically working data scientists, analysts and engineers from across Leeds and West Yorkshire.

Agenda

• 18:00–18:45 — Refreshments (food served on a first-come, first-served basis)
• 18:45–18:50 — Welcome
• 18:50–20:00 — David Hoyle: In demand: The data science behind predicting what sells and what's next for that data science

Speakers

• Dr David Hoyle — Research Data Scientist Specialist at dunnhumby. PhD in Theoretical Physics; former researcher at the University of Cambridge; former joint Head of Data Science at AutoTrader UK. Nearly 35 years' experience in statistical and mathematical modelling.