_{Turns out our riders figured Walmart out already…}

I've been thinking about Walmart lately—not the store, but the revolution. When computers started emerging in the 1980s, Walmart didn't just "adopt computers." They rebuilt their entire operation around what computers made possible: real-time inventory tracking, SKU-level logistics, supply chains that moved like nervous systems. Even though their competitors bought the same hardware, Walmart reimagined what a retail operation could be.

Amazon did the same thing with the internet. They actually didn't build "an internet company," they built an operation that could only exist because of the internet. Logistics became their core competency because e-commerce removed the bottleneck of orders. 

(I’ve been listening to a lot of Acquired recently, telling stories of how companies became great. It’s helping me put today’s AI craze into perspective of what is actually possible based on history.)

I firmly believe AI is the biggest paradigm shift since the personal computer (bigger than the internet!), and it’s happening 10 times faster. But most organizations are responding to AI the way Walmart's competitors did—bolting the new technology onto old assumptions.

The deepest assumption we're all carrying is that execution is hard and requires immense expertise. That's why we take years to study and analyze things before daring to execute any change.

But analysis isn't limited to data wizards anymore. Take a look at this amazing dashboard put together by Chris Pangilinan, most recently VP and SVP at New York City Transit, in just 2 hours.  In 2021, this would’ve taken 24 months to build navigating procurement, management, and development. 

Let that sink in: 2 years has become 2 hours… overnight. 

Here's what most agencies do when they identify a problem: they scope a study, issue an RFP, hire a consultant, wait six months (minimum) for findings, scope another RFP for the technology, procure it, implement it, and — if everything goes perfectly — they've "solved" the problem five years after they spotted it. Yet by then the problem has changed, the community has moved on, the data is stale, and the next problem is already three years deep. COVID made all of this glaringly obvious to us.

This isn’t a technology problem. This is an operating model built on the assumption that analysis is expensive and expertise is scarce. Walmart's competitors made the same assumption about inventory tracking — that it required specialists, custom processes, long lead times. Walmart rebuilt the operation around the reality that computers made tracking instant. The question became: what do you do with real-time information?

Transit is sitting on the same turning point. You already have the data — ridership patterns, on-time performance, census demographics, NTD submissions full of operational intelligence. The bottleneck was never the data. It was the months of work between "here's what the data says" and "here's what we're doing about it." AI collapses that gap. Chris didn't need a vendor or a procurement cycle to build that dashboard. He needed two hours and the judgment to know what mattered.

Now imagine that capability spread across your agency. Your planners testing route alternatives in an afternoon instead of a quarter. Your grants team identifying funding matches in minutes instead of weeks. Your GM walking into a board meeting with scenario analyses that were built that morning, not that fiscal year (okay, we can dream right?). 

The agencies that figure this out won't just adopt AI — I’m working with agencies who are actively rebuilding their operations around the reality that analysis is instant and expertise can be amplified. Just like Walmart didn't simply "buy computers"—they became a different kind of company—these agencies are challenging the very idea of how public service could be organized.

AI isn't about building a better bus. It's about you serving your community more effectively by introducing higher quality service faster than you ever could before.

So stop asking yourself how AI helps you get more done. Ask yourself: are you ready to rebuild around what AI makes possible?

TransitGPT is where I share practical AI insights for transit leaders. Subscribe if this resonates or forward to a colleague who needs it. And hit reply — I read everything.

Go shatter your glass ceiling.