This is an archive episode, first broadcast in 2024 under the show's previous name. It carries a short synthetic announcer introduction; the conversation underneath it is untouched human audio.
Most AI pilots die at the end of the pilot. The reason is usually not the technology.
Patrick Ward spent thirty years in commercial roles, twelve of them at Microsoft, where he worked on their global approach to IoT and AI. He now lectures at UCD Smurfit Business School. His diagnosis is that companies start with the technology and then go looking for a business model to attach it to, which is the wrong way round.
Recorded in November 2024, when the show carried a different name.
What this episode covers
- Lots of proofs of concept, not much scaling. The pattern he kept seeing, and what distinguishes the projects that survive contact with the business.
- Connect a device to the cloud and your business model changes. Whether or not anyone in the company has noticed, and usually they have not. He sets out three fundamental truths about IoT and AI that decide whether what follows is a product change or a business change.
- Get the custodians of the business model in one room. His practical intervention, and why the first hour of it is usually spent discovering that ten people hold ten different views of what the company sells.
- Bolt-on or rethink. The distinction he boils everything down to, and why strapping a digital business onto the side of an existing one tends to fail.
- Why the HVAC sales team will not sell digital services. A concrete example of compensation and habit quietly defeating a strategy that looked sound on the slide. He pairs it with a smart buildings payback case where the numbers did work.
- The research finding that surprised him. Fifteen multinationals in Ireland, and a talent gap that turned out to sit with business leaders rather than with data scientists.
- Twenty use cases in one morning. What happened when ten developers were asked properly, and what that says about where ideas actually are.
- Do not build a business case you cannot measure. His warning on productivity claims, which are the easiest to promise and the hardest to defend afterwards.
- Governance, policy, use cases, training and sharing. The five things he wants in place, and the story of the sceptical CTO whose objection turned out to be the red flag.
- Drinking from the fire hose. His honest answer on keeping pace, which is that you cannot, and what to do instead: identify champions and keep learning deliberately.
- Competitor insight from press releases and job postings. A method that costs nothing and tells you what rivals are actually building rather than what they say they are.
About the guest
Patrick Ward has thirty years of commercial experience, including twelve years at Microsoft, where he transformed their global approach to IoT and AI implementation. He is a part-time lecturer at UCD Smurfit Business School, which has since introduced a postgraduate diploma in AI and business analytics that addresses exactly the gap he describes in this episode. His research covering fifteen multinationals operating in Ireland is where the episode's most surprising finding comes from.
Resources mentioned
- UCD Smurfit Business School - where he lectures, and the new AI and business analytics postgraduate diploma
- AI envisioning sessions - the format he uses to get from nothing to a real list
- Competitor press releases and job postings - his low-cost method for reading what rivals are actually doing
Quotes
"The bigger challenge was: how do we educate our business leaders to think about AI in the context of the overall strategy of the business?"
Patrick Ward, on the real talent gap
"There is risk in it, and they're right, there is risk in it."
Patrick Ward, on hesitant leadership
Books - the Human-First AI Series
- Hello! I'm New Here
- An Absolute Beginner's Guide to AI
- Use AI, Stay Human
- The AI Pilot Handbook
- When AI Becomes Your Responsibility
Connect with Alastair on LinkedIn.