🍒 The low-hanging fruit for AI

Hi, and happy Tuesday.

What do IBM's $4 billion Watson project and McDonald's AI drive-thru have in common?

They both fell into the same trap that's catching leaders everywhere… 

Going for the moonshot.

Talk of AI is everywhere. 

At board meetings. In vendor decks. On every newsfeed.

And every time, the pressure builds: 

“AI is reinventing markets. We need an AI strategy.”

So you brainstorm. Automate customer support? Rebuild data pipelines? Launch a chatbot?

But nothing feels right.

So the conversations repeat… and repeat.

It’s not that you don’t want to act.

It’s that with a hundred possible starting points, it’s impossible to know which one is worth the bet.

Here’s what we’ve seen work in the field:

1. The first step is friction, not a moonshot
The highest-value entry points are the repetitive processes that make your team groan. One global electronics company cut a 3-week manual documentation cycle down to just 2 days by automating updates across 50 countries.

2. The best use cases pass four filters
Across industries, the best candidates are work that is routine, voluminous, data-rich, and clearly ROI-positive. For all the hype, AI is just autocomplete [6 min video]. For where the technology is right now, building robust systems that meet this criteria is the lowest-hanging fruit - and where organizations are winning.

3. The early wins outweigh the master plans
Automation opportunities deliver ROI fast because the metrics are obvious. In one project, a retailer automated inventory reconciliation, reducing reporting time from days to hours. That quick result built confidence across the organisation and paved the way for larger initiatives.

4. The momentum grows with every win
Each successful use case frees capacity and sharpens your playbook for the next one. Over time, this builds real AI maturity, not through a single moonshot, but through repeatable results that compound. One financial services team, for example, automated customer verification and cut onboarding time by 60%. That quick win gave them the credibility and confidence to expand further.

To paint the picture, compare this moonshot temptation:

To this similar, but tangible, realistic and achievable alternative: 

See what I mean?

That’s the roadmap: 

1. Start where friction hurts, 

2. Filter with clear criteria: routine, voluminous, data-rich, and clearly ROI-positive

3. Win small, stack wins, and

4. Let the strategy emerge.

If you know some colleagues that may need a little nudge to get going with AI, here’s my offer:

I’m considering hosting a few complimentary AI Masterclasses based on my book, "Do More With Less: The AI Playbook for Amplifying Talent & Output". It’s a short “done with you” program that gets early wins for participants with the tools they already have (e.g. Copilot and Excel).

If you know anyone that might be interested, just have them complete this form: https://docs.google.com/forms/d/e/1FAIpQLSf8ZbGGY1ZdaiUxxCOTTS3CIFmkO9f-oSzU0cfYLjO_cQn7dw/viewform

Talk soon,

Dino

Catch up on our recent videos: