The Data Told You. Now Your Team Decides.

There's a management story that has been told in business schools for fifty years. A family is sitting comfortably on a porch in Coleman, Texas on a hot afternoon. Someone suggests driving to Abilene for dinner, about fifty miles away, in the heat, in a car with no air conditioning. Nobody really wants to go. But nobody says so. One by one they all agree. They make the miserable trip, eat a mediocre meal, and drive back exhausted.

On the way home someone admits they did not actually want to go. Then everyone admits the same thing. Nobody wanted to go to Abilene. They all went anyway because each person assumed everyone else wanted to go and did not want to be the one to object.

Management theorist Jerry Harvey called this the Abilene Paradox. Groups often take actions that no individual member actually wants to take, because each person suppresses their real preference to avoid conflict or to go along with what they assume the group wants.

This is worth keeping in mind the next time your team sits down to act on a piece of data that everyone agrees is important.

The Knowledge Is the Beginning, Not the End

The last several years have produced remarkable tools for understanding what is actually happening inside a business. Cloud cost attribution. AI inference tracking. Customer-level unit economics. Gross margin by segment. For SaaS companies navigating the AI era, the ability to see which customers are profitable, which features earn their cost, and where the margin is actually coming from has gone from impossible to achievable.

That visibility is valuable. It's also just the beginning.

Data tells you what is happening. It does not tell you what to do about it. That step, the translation from knowledge to decision, is where leadership earns its place. And it's harder, messier, and more human than any dashboard can capture.

A Table Worth Sitting At

Imagine a SaaS company that has just received a clear picture of its customer economics for the first time. The data shows that a meaningful segment of customers is costing more to serve than they pay in subscription revenue. The company is subsidizing them, quietly, every month. The leadership team sits down to decide what to do.

The CFO speaks first. "Dump them. Plain and simple. We are running a business, not a charity."

The sales manager pushes back. "We cannot do that. Our sales force is compensated on new customer acquisition and retention. If we start exiting customers, our people take a pay cut for something that was not their fault."

The HR manager sees an opening. "That is a fair point, but it's a compensation design problem, not a reason to keep losing money. We could restructure the plan. Pay people to bring in profitable customers. Stop penalizing them for customers who turn out to be unprofitable through no fault of their own."

The CTO offers a different angle. "The primary reason we are losing money on these customers is AI usage. They are heavy users of our AI features. I can build usage limits. Cap their consumption and the cost problem largely solves itself."

The marketing officer nods but adds a caution. "That might work technically, but we have to think carefully about how we communicate it. These customers did not sign up expecting limits. We need to offer them an alternative, a different plan, a usage tier, something that does not feel like a penalty for being engaged users."

And then the CEO says something that stops the room. "Our first customer, the one who took a chance on us before we had a product worth buying, is one of the names on this list. Do we not owe that person something for their loyalty?"

What Just Happened in That Room

Every person at that table was right. Not partially right. Not right from a limited perspective. Right in a way that the data alone could never capture.

The CFO is right that unprofitable customers are a real problem that compounds over time. The sales manager is right that incentive structures drive behavior and you cannot change outcomes without changing incentives. The HR manager is right that the compensation design is the lever, not the customer relationship. The CTO is right that the cost is largely technical and has a technical solution. The marketing officer is right that how you communicate a change matters as much as the change itself. And the CEO is right that loyalty has value that does not appear on a cost report.

The data showed the problem. The team brought the solutions. None of those solutions were in the dashboard.

This is the conversation that the knowledge makes possible. Without the data, the company stumbles along subsidizing customers it cannot identify, making no decision because there is nothing concrete to decide. With the data, the conversation can finally happen. The data does not make the decision. It earns the right to have the meeting.

The Abilene Paradox Waiting to Happen

Here is where it gets dangerous.

Imagine the same meeting, but with a different culture. The CFO speaks. Everyone in the room knows the CFO controls the budget. The sales manager has a concern but does not want to seem like he is protecting his commission. The HR manager is new and does not feel established enough to challenge the CFO. The CTO thinks usage limits are a bad idea but does not want to be seen as an obstacle. The marketing officer worries about being the one who slows down a decision. The CEO is thinking about the loyal first customer but does not want to seem sentimental in front of the team.

So they all nod. They dump the customers. The sales team is demoralized. The first loyal customer calls the CEO personally, hurt and confused. The CTO quietly knows the problem will recur because the underlying cost structure was not addressed. And six months later, someone mentions in a hallway conversation that nobody actually wanted to do it that way.

They went to Abilene.

The antidote to the Abilene Paradox is not better data. It is a culture where people feel safe enough to say what they actually think. Where the CFO's first instinct is treated as a starting point, not a verdict. Where the sales manager's pushback is welcomed as information, not resistance. Where the CEO's concern about a loyal customer is understood as a legitimate strategic input, not sentimentality.

The data creates the conditions for a good decision. The culture determines whether a good decision actually gets made.

What AI Cannot Do Here

The right visibility tools can tell you which customers are unprofitable. They can model the financial impact of different scenarios. They can draft a communication plan for usage limits or calculate the cost of restructuring a compensation model.

What no tool can do is tell you what the loyal first customer is worth. It cannot weigh the demoralization of a sales team against a quarterly margin improvement. It cannot sense that the HR manager has something important to say but is holding back. It cannot recognize when a room full of nodding heads is actually a room full of people who do not want to go to Abilene.

Those are human judgments. They require relationships, context, organizational history, and the kind of pattern recognition that comes from years of watching how people and organizations actually behave. No model has that. No dashboard surfaces it.

This is not a limitation that will eventually be fixed. It is a permanent feature of the boundary between information and wisdom.

The Two Steps That Matter

The SaaS companies navigating the AI era well are the ones that have figured out both steps in sequence.

The first step is visibility. Know what your product actually costs to deliver. Know which customers are profitable and which are not. Know which features earn their cost and which are eroding your margin. Without this step, the second step is impossible, because you are making decisions in the dark.

The second step is decision making. Bring the right people into the room. Create the conditions where every perspective gets heard. Watch for the Abilene Paradox. Recognize that the friction and discomfort of a real debate is the system working correctly, not failing. And understand that the data gives you the map. The leadership team decides which road to take.

Neither step works without the other. Data without decision making is just an expensive report. Decision making without data is just an expensive guess.

The knowledge was the beginning. Now your team decides.

*Richard Allen is a contributing writer on strategy and organizational effectiveness for Beakpoint Insights.