Every leadership team seems to be having the same conversation right now. What’s our AI strategy? Which tools do we adopt? How fast can we move? The pressure is real, and the spending shows it.

Now for the number that should give every executive pause. Gartner research found that only about one in fifty AI investments delivers transformational value, and only one in five delivers any measurable return at all.

The technology is remarkable. The returns, for most organizations, are not. So what separates the one in fifty from everyone else?

It’s rarely the tool. It’s almost always the culture the tool landed in.

Think about what AI adoption actually asks of your people. It asks them to change how they work, admit what they don’t know, experiment in front of their peers and trust that efficiency gains won’t be used against them. Those aren’t technology questions. They’re alignment questions, and if the answers are murky, no license purchase will fix it.

A common scenario: an organization rolls out an AI tool with an enthusiastic announcement and a training webinar. Six months later, usage is thin. Ask around and you hear the real answers. Nobody explained how this fits our mission, so it feels like a fad. My manager still measures me the old way, so why would I change how I work? And the quiet one underneath them all: if this thing makes my job faster, does my job survive?

Those questions don’t get answered by IT. They get answered by leadership, and ideally before the rollout rather than after.

This is where alignment work earns its keep. Before adopting any significant technology, a leadership team should be able to answer three questions in one voice. What problem are we solving, in plain language, and how does it connect to who we are? What will we do with the time and capacity this creates? People deserve to know whether efficiency means growth or means cuts, and silence always gets interpreted as the worst case. And what will we protect? Every organization has work that should stay human: the client relationship, the difficult conversation, the judgment call. Naming what won’t be automated builds as much trust as explaining what will.

There’s also a practical step most organizations skip: bringing the people closest to the work into the decision. Collaborative visioning isn’t a luxury here. The employees who’ll use the tool daily can see failure points no vendor demo reveals, and people support what they help create. That was true long before AI. It’s truer now.

The organizations getting real value from this technology didn’t move fastest. They moved aligned. Their people understood the why, trusted the intent and helped shape the how. The tool was the easy part.

So before your next technology conversation, have a culture conversation. Get your leadership team on the same page about purpose, capacity and what stays human. Then adopt whatever tools you like. You’ll be building on solid ground instead of hoping the ground forms beneath you.