In 1996, more than 100 cities wanted Southwest Airlines to fly to them. Southwest opened service in four.

I keep coming back to that number. Four. Imagine being in the room when the opportunities were laid out. How tempting would it have been to say yes to a few more? And how much conviction would it have taken to say, “Four is what we can do well right now”?

I thought of Herb Kelleher and Southwest after reading Jon Kirchner’s Entrepreneur article about slowing down when everyone is rushing. The pressure he describes feels familiar. Leaders see what AI can do, hear what their competitors are doing, and wonder whether taking more time will cost them their advantage. It’s a fair concern. And it brings me to the question I see at the heart of Southwest’s story:

How fast can we responsibly expand when the opportunity is bigger than our capacity to handle it?

WHAT SOUTHWEST WAS WILLING TO PROTECT

Kelleher put it this way: “Think small and act small, and we’ll get bigger. Think big and act big, and we’ll get smaller.”

What I hear in those words is a leader who understood what it took to make good on a promise. Every new city meant more people to hire and train, more flights to operate, and more customers counting on Southwest to deliver the service they’d come to expect. The opportunity was real. So was the work of being ready for it.

Jim Collins uses Southwest to illustrate what he calls the 20-Mile March: the discipline to keep moving in difficult times and to hold to a pace you can sustain when conditions are good. The second part is easy to underestimate. It takes discipline to keep going when things are hard. It takes a different kind of discipline to hold back when there’s a line of people asking you to do more.

The contrast matters. Major carriers went through bankruptcy in the years that followed. In 2003, the U.S. airline industry recorded its third consecutive annual loss, while Southwest reported its 31st consecutive profitable year. Those airlines faced different pressures, and growth alone doesn’t explain their outcomes. What Southwest’s choice shows is that the time to protect your ability to deliver is when you still have the freedom to choose your pace.

BEFORE YOU SPEED UP THE WORK, TAKE A LOOK AT IT

This is where I see the connection to AI. A tool can be ready to use long before the work is ready for the tool.

In Entrepreneur, Rejna Alaaldin makes a point I think every leader should sit with: before asking how AI can make a process faster, ask whether the process should exist at all.

Think about a report your team spends hours preparing every week. AI might help them produce it in minutes. That sounds like a win. But what if no one uses it to make a decision? Or what if two departments define the same metric differently, and your people spend every Monday reconciling the numbers? A faster report won’t settle which definition the business should use. Your people still have to do that work together.

AI may help you see the mess more clearly. Cleaning it up means deciding what’s worth keeping, agreeing on what information people can trust, and making sure someone knows when to question the result. Otherwise, a problem one team used to catch can become a problem ten teams inherit.

You can start small enough to learn while you do that work. Give one team one task that matters. Ask the people closest to it where it breaks down and what they’d change before using AI. Then see whether the new way of working holds up on an ordinary, busy day. Can people check the output without creating a second job for themselves? Do they know what to do when something looks wrong? Their experience tells you more about whether you’re ready to expand than a promising demo ever could.

That’s what I admire about Southwest’s four cities. Each yes was a commitment to the people who would do the work and the customers who would count on them. As leaders, we can bring that same care to the choices we make about AI. We can be excited about what it makes possible and honest about what it will take to do it well.

What needs to be ready before we ask our people to go faster?

Author(s)