Your Data Has All the Answers. You're Just Asking the Wrong Questions.
Here’s something I’ve learned from years of staring at databases: data doesn’t lie, but it will absolutely mislead you if you let it.
Not on purpose. Data doesn’t have an agenda. But it answers exactly the question you ask — no more, no less — and most of the time, we’re asking the wrong one.
The dashboard that told me nothing
When I first built my personal finance dashboard (a Python/Streamlit thing that pulls my bank transactions through Plaid into a local SQLite database), the first question I asked was the obvious one: how much am I spending?
The answer came back instantly. Charts. Totals. Month-over-month comparisons. Very satisfying.
Also completely useless.
Knowing I spent $2,300 last month tells me almost nothing about why, or whether that’s a problem, or what to do about it. It’s data without context. A number floating in a void.
So I asked a better question: where does my money go when I’m not paying attention?
That one hit different. Turns out “not paying attention” for me means food delivery, streaming services I forgot I had, and a deeply concerning number of convenience store stops. The data hadn’t changed. The question had. And suddenly it was telling me something I could actually act on.
The question shapes the answer
This isn’t just a personal finance problem. It’s the central challenge in any data work.
In BI and reporting, the most common mistake I see isn’t bad data or broken pipelines — it’s a well-built dashboard answering the wrong question confidently. Someone asked “how many units did we sell?” when they should have asked “which units are we selling instead of the ones we want to sell?” Same data. Completely different insight.
The question you ask determines:
- What data you even look at
- How you slice it
- What counts as a pattern vs. noise
- What you do next
Ask a narrow question, get a narrow answer. Ask a lazy question, get a lazy answer that feels complete because it has a number attached to it.
So how do you ask better questions?
Honest answer: you get better at it by being wrong a lot and noticing when your data isn’t actually helping you make a decision.
A few things that’ve helped me:
Start with the decision, not the data. What would you do differently if the number were higher vs. lower? If the answer is “nothing,” you’re measuring the wrong thing.
Ask “compared to what?” A number in isolation is almost meaningless. Spend more than last month? More than average? More than someone in a similar situation? The comparison is where the story lives.
Follow the surprise. When data tells you something unexpected, that’s not a data quality problem — that’s the most interesting thing in the room. Don’t smooth it over. Dig in.
Data is everywhere. Most organizations are drowning in it. Most individuals generate more of it in a day than they’ll ever look at. The bottleneck was never the data itself.
It was always the question.