The Data Is Wrong: The Statement That Gives Me the Heebie-Jeebies

"The data is wrong."

Few statements give me the heebie-jeebies quite like this one. Not because it's never true - sometimes data genuinely is wrong. But because it almost always ends a conversation that should be starting. Analysis surfaces something uncomfortable, someone says those four words, everyone nods, and the uncomfortable finding gets quietly shelved. Nothing changes.

The dismissal pattern

The sequence is remarkably consistent:

1. Analysis reveals something uncomfortable
2. Someone says "that data is wrong"
3. Everyone nods
4. The finding is ignored
5. Nothing changes

"The data is wrong" functions as a shield. It's far easier to question the data than to question the decisions the data is challenging.

What's actually happening

When I hear that phrase, one of four things is usually going on:

The data disagrees with intuition. Which might mean the data is wrong. Or might mean the intuition is wrong. Dismissing it without investigation means you'll never know which - and you'll keep operating on the intuition regardless.

The data surfaces a problem. Problems are uncomfortable, and data that exposes them can feel like an accusation. "The data is wrong" is sometimes just "I don't want this to be true."

The data contradicts a prior decision. If a strategy was approved on certain assumptions, data that challenges those assumptions threatens the decision itself. Easier to question the messenger.

The data actually is wrong. This does happen. But it should be the conclusion of an investigation, not the starting assumption that prevents one.

The signals going unheard

Research consistently shows that most organisations use a fraction of the data they already collect. The rest sits in systems, untouched. Every "the data is wrong" dismissal adds to that waste - data that could reveal something gets labelled unreliable and left alone.

The irony is that the data most likely to be dismissed is often the data most worth examining. Comfortable data gets accepted. Uncomfortable data gets questioned. Over time, organisations quietly train themselves to ignore exactly the signals that challenge the status quo.

The alternative: investigate, don't adjudicate

When someone says "the data is wrong," the move is not to accept or reject - it's to investigate.

Ask: "Wrong how?" Is the issue completeness (missing records)? Accuracy (incorrect values)? Timeliness (stale information)? Getting specific transforms a dismissal into a solvable problem.

Ask: "Can we verify?" Is there another source to check against? Can we sample and validate a subset?

Ask: "If the data were right, what would that mean?" Sometimes this question is hard to answer honestly - and that difficulty tells you more about the resistance than about the data.

Ask: "What would convince you?" If there's a standard of proof that would settle it, pursue it. If no evidence could ever be sufficient, the objection isn't really about data quality.

The goldmine you already own

The data sitting in your CRM, your call reporting system, your customer records - it isn't perfect. No data is. But imperfect data analysed thoughtfully beats confident intuition operating blindly.

The patterns are there. The trends are visible. The problems are already signalling. "The data is wrong" is what keeps those signals from being heard.

What I say now

When I hear it, I've learned to say: "Let's find out."

Not "you're wrong." Not "trust the data." Just: let's find out.

Investigation usually lands in one of two places:

1. The data has a specific, fixable problem. Good. Now we know what to fix, and we try again with better data.

2. The data is sound, and someone doesn't like what it shows. Also useful. Now we can have the real conversation - about what the data means and what we're going to do about it.

Both outcomes move things forward. "The data is wrong" as a final statement moves nothing.

Your data isn't perfect. It's still better than guessing. Stop dismissing it and start investigating.