Selling Shovels and the AI Revolution

The most striking aspect of the AI revolution isn't the technology. It’s that AI has suddenly become a part of both business and home life. This duality drives change forward in a way we haven't really seen since the launch of Apple's iPhone and Google's Android. Digital transformation is never-ending, but the lessons we learn with each new wave of technology revolution are largely the same.

"When there's a gold rush, sell shovels"…

At the dawn of the mobile revolution, big enterprises were eager to roll out custom mobile apps for their workers, and I was a part of the team building them for some of the biggest. There are tons of parallels with AI today. The first thing we saw back then was that IT teams were initially interested in quick wins to demonstrate the technology. Nobody wanted to be the one without a mobile story to tell. These didn't gain much traction or adoption because they didn’t really impact the business. Next, we noticed business leaders in enterprise Sales and Marketing teams initiating mobile projects without involving their IT colleagues. This led to rapid, but unsustainable change due to the absence of infrastructure needed for long-term maintenance and support. It also led to more than a few uncomfortable conversations. Mobile projects really took off when companies started really investing in transformation across the business and IT in a structured and scalable way.

So what are some lessons from the mobile revolution that also apply to AI projects today?

Start with People: The first lesson is to start with people. Not with requirements, not with technology, not with ideas or systems or processes. The only way to really know what people actually need is to learn from them. Do ride-alongs, interviews, and co-creation sessions. Get out the sticky notes and a white board. Plan for training, enablement, partnerships, and hiring to build culture. This is a critical step in any business transformation, and it’s the first step of effective change management that will ultimately fuel adoption when you’re ready for it.

• Application to AI: This principle is crucial in the AI domain, where successful applications must be deeply aligned with user needs and contexts. Engaging with end-users from the outset ensures that AI solutions are not only technically feasible but also practically valuable.
• Real World Example: It's inspiring to see what LVMH (Louis Vuitton Mo?t Hennessy) is doing with the Stanford University’s Institute for Human-Centered Artificial Intelligence Institute (Stanford HAI) adopting a thoughtful approach to AI. “Stanford HAI’s mission focuses on how to properly design and build human-centered AI to have positive human impacts”, and with LVMH they are creating educational programs for end users, data scientists, and leaders, and incorporating Human-Centered Design principles in their new AI initiatives (check it out).

Scale Fast: Real transformation happens when IT and business leaders are working together to create secure and scalable solutions that solve real business problems and drive immediate value. The challenge with AI is turning experiments into something much bigger: “AI shouldn’t just change how you do things, it has to inspire you to?change what you do” (HBR).

• Application to AI: AI initiatives should be designed with a cross-functional approach, ensuring that they are not just technologically advanced but also strategically aligned with business objectives and capable of driving immediate value.
• Real World Example: Coca-Cola has a great approach to company-wide collaboration, having launched a digital council with C-level members and liaisons assigned to each operating unit. This approach has led to a strategic roadmap that prioritizes use cases that serve as good building blocks for other, more diverse applications. “We’re embracing the need to take risks, experiment with AI across our system, and build on what we learn to drive scale” (check it out)

Intent to Roll-Out: Capture user feedback, and do it In-App. Pilot with the intent to roll-out quickly by learning as you go. Some ideas will catch on, some ideas will fail, but you won’t know until end users are using it. You’ll be surprised by what they think: one of the best things I ever heard from a user while rolling out an offline-capable mobile field sales solution was “this is amazing, it means I’ll get to spend more time with my kids at night”.

• Application to AI: AI projects should be deployed with a clear path to scaling and integration into operational processes. This approach ensures that lessons learned during pilot phases are quickly incorporated, enhancing the solution's effectiveness and driving adoption. Planning to adapt to user feedback from the very beginning is the best way to make sure you don’t struggle with adoption later on.
• Real World Example: I love this study that attempted to determine if “an artificial intelligence chatbot assistant can provide responses to patient questions that are of comparable quality and empathy to those written by physicians?” because what they learned from users was that the chatbot responses “were preferred over physician responses and rated significantly higher for both quality and empathy.” (check it out)