The Information You Trust Most Is Probably the Least Useful
Organizations don’t treat all information equally.
Organizations don’t treat all information equally. They have preferences. They invest in certain kinds of information — dashboards, reports, AI summaries, synced systems — and they neglect others. The question worth asking is whether those preferences track what’s actually useful, or whether they track what’s measurable, producible, and easy to defend in a meeting.
The answer, in most organizations, is the latter. The information that gets the most attention, the most tooling, and the most trust is frequently the information that changes nothing. And the information that would actually improve a decision — the rough estimate, the dissenting read, the thing someone knows but hasn’t written down — gets treated as a liability.
This isn’t a data quality problem. It’s a judgment problem about what information is for.
The Two Dimensions That Matter
Information can be evaluated on two axes. The first is accuracy — is it correct? The second is usefulness — does it help you make a decision, resolve a tradeoff, reduce uncertainty, or take meaningful action?
These two dimensions are not the same thing, and organizations consistently conflate them.
Accuracy is objective. A number is either right or wrong. Usefulness is contextual — it depends entirely on what you’re trying to do, what you already know, and what the decision in front of you actually requires. An accurate number can be completely useless. An imprecise estimate can be exactly what the situation needs.
Mapping these two dimensions against each other produces four categories of information. Most organizational information falls into the wrong ones.
The Four Quadrants
Accurate and useful is the gold standard. This is the information that actually changes what you do — that clarifies a tradeoff, resolves an ambiguity, or reduces a real risk. It is rarer than organizations believe, and it is almost never the information they invest the most in producing.
Accurate and useless is where most organizational information lives. The weekly report nobody acts on. The dashboard with metrics that don’t affect any decision. The perfectly formatted AI summary of a meeting where nothing was decided. The synced system with no corresponding meaning. None of it is wrong. None of it matters. This quadrant is the hidden cost center of the modern workplace — it consumes attention, produces the feeling of control, and changes nothing.
Inaccurate and useful is the quadrant organizations are afraid to admit exists. Rough estimates. Early signals. Gut reads based on pattern recognition. Back-of-the-envelope calculations that point in the right direction without being precise. Data-obsessed cultures outlaw this quadrant and become slower and blinder as a result. Directionally correct is often more valuable than precisely irrelevant.
Inaccurate and useless is where meaning goes to die — outdated documentation, conflicting metrics, inherited assumptions nobody questions, AI outputs built on stale context. This quadrant silently corrodes decision-making. But the real danger isn’t that it exists. It’s that when context is missing, organizations cannot tell the difference between this quadrant and the first one. Accurate-and-useful and inaccurate-and-useless look identical when you’ve lost the context that would let you distinguish them.
That last point is the one that matters most. Context is what determines which quadrant any piece of information actually occupies. And context is the thing organizations lose fastest.
What Organizations Get Wrong
The preferences are consistent and consistently backwards. Organizations systematically overvalue accuracy, volume, completeness, and visibility. They undervalue usefulness, meaning, interpretation, and the rough signal that doesn’t fit neatly into a report.
The result is an information environment optimized for the second quadrant — accurate, useless, expensive to produce, and impossible to act on. Teams spend their cognitive budget processing information that was never going to change anything, while the information that would — the constraint nobody named, the assumption the model is built on, the thing the person in the room knows but didn’t write down — goes unrecorded and unshared.
AI makes this worse, not better. AI accelerates production of the second quadrant. It generates accurate summaries of useless meetings, confident analysis of irrelevant questions, and perfectly formatted noise. Without a model of utility, AI doesn’t solve the information problem. It compounds it.
The Question That Changes Everything
The Information Utility Index reframes a single question.
Most organizations ask: do we have information?
The more useful question is: does this information matter — and to what decision, in whose hands, at what moment?
That shift sounds small. It isn’t. It moves the evaluation of information from a production question to a judgment question. It forces explicit reasoning about what a piece of information is actually for before investing in producing, storing, or distributing it.
The organizations that will manage information well in the next decade are not the ones with the most data or the most sophisticated systems. They are the ones that develop a shared sense of what information has utility — and the discipline to stop producing, distributing, and trusting the kind that doesn’t.
More information is not the answer. It has never been the answer. The answer is better judgment about which information deserves the weight you’re giving it.
Incontextable — Alexander J. Cooper