The Hardest Part of Explaining Data to Non-Analysts Is Not the Data

You present a dashboard to C-level executives about the company’s financial position. You know every metric. Every assumption. Every limitation.

The slides are clean. The charts are readable. The notes explain everything.

Then someone asks:

“But what does it actually mean for us?”

That moment reveals a problem many analysts eventually encounter:

Explaining data is not the same as helping people make decisions.

Many analysts believe the solution is obvious: just simplify the data.

Advice often sounds like this:

• Present statistics in a simpler way
• Remove jargon
• Choose clearer charts
• Explain the methodology

Though all of this is good advice, none of it addresses the hardest part.

The real challenge is translating between mental models

Data professionals think in terms of:

• Distributions
• Uncertainty
• Probabilities
• Assumptions

Business leaders think in:

• Decisions
• Risks
• Costs
• Impact

Analysts ask: “What does the data say?” Executives ask: “What should we do?”

Or, framing it differently:

Data answers what and why. Business cares about so what and now what.

The Emotional Layer (Often Ignored)

But there is another layer. The emotional one.

Data is rarely neutral.

It can imply that someone made the wrong decision.

It can challenge intuition.

It can create uncertainty instead of certainty.

And when that happens, people start protecting themselves. This is where analysts need to become part psychologist.

If your analysis threatens identity, status, or previous decisions, logic alone rarely wins the argument.

Your job is to communicate insights without triggering defensiveness, while still standing confidently behind your conclusions.

Once I understood this gap between mental models and emotions, I started noticing recurring mistakes, including my own.

Common Communication Mistakes Analysts Make

In communication with non-analysts, avoid:

• Starting with methodology. In the first place, the business is not interested in how you set up the experiment or what statistical methods you applied.
• Showing too many charts. Dashboards with too many charts feel cluttered and hard to read. Leave some space; in many cases, small text notes can do more than sophisticated charts.
• Over-explaining. When you present your analysis, try not to be too zealous with explanations. Nobody wants to hear a lecture on statistics or probability, or an explanation of why the scatter plot is so important to present distributions.
• Being defensive. When your audience asks questions, don’t be aggressive in protecting your analysis. Explain your logic calmly and be ready to admit mistakes if there are.
• Refusing to simplify. When you are deeply immersed in the analysis, it can feel obvious to you, but not to your audience. That’s where the misunderstanding begins: you start speaking in the language of data, while your audience does not know that terminology. Speak their language.

What Actually Works

With so many “don’ts”, what actually works? You can rely on a practical approach that structures communication around five principles:

1. Start with a business question.

Before opening a dashboard, clarify the decision that needs to be made.

Instead of focusing on distributions or statistical outputs, define the core business problem:

• Who is most exposed to credit risk?
• Where are we losing profitable customers?
• Should we invest or postpone?

2. State the conclusion first.

Executives do not want to wait for the “punchline.”

Lead with the insight:

• Borrowers with a past default are ~3× more likely to default again.
• Current interest rates do not fully reflect long-term loan risk, leading to a potential pricing opportunity.

Then explain how you arrived there, if needed.

This is executive communication logic: answer first, justify second.

3. Translate metrics into impact.

We are obsessed with metrics. A metric shows activity, but impact shows value.

The average default rate is reported to be consistent and very low. This is a good indicator from the point of view of that metric. But sometimes it indicates the lending approval parameters are so strict that the bank only lends to people with “iron” credibility, losing a big chunk of the market (for example, young people who have not yet had a credit history).

By reconsidering the lending rules, the bank starts winning new clients. So, a low default rate is a risk metric; finding new profitable markets is the growth impact.

4. Acknowledge uncertainty confidently.

When dealing with data, you can rarely be 100% sure of your conclusions and recommendations. There is always some degree of uncertainty. And it’s better that you explicitly mention this fact when presenting your analysis or preparing the report.

5. Tie insights to decisions.

This transforms your work from just a report that sits in an inbox into a decision that moves the business.

Imagine the bank analysis showed the customers leaving (churning) weren’t the low-value ones; they were the “silent high-value” users who didn’t complain but stopped using the card. That is insight. To make it actionable, this insight should imply the decision. For example, reallocate 70% of their retention budget away from mass marketing and toward a “Concierge Outreach” program for high-value users whose “Love Score” dropped by more than 20% in a month.

The mistakes are common, especially early in an analyst’s career. A common example illustrates this well.

Imagine a discussion about whether to buy new equipment or continue repairing the old one.

Presenting a full Net Present Value calculation with discount rates and depreciation schedules may be technically correct, but irrelevant to the site manager’s core concern:

“Will the equipment reliably operate next month?”

The fix is easy: there is no need for a detailed explanation and NPV calculation. You can just talk about breakeven and explain that the old equipment cost us $4000 per week in repairs and downtime. The new would cost $3500 a month in financing. It would pay for itself in less than a year simply by staying operational.

The goal isn’t just simplifying numbers. The real goal is to bridge two different worlds.

Moving from probability to responsibility. From correlation to consequence. From insight to decision.

The moment you shift from “What does the data say?” to “What should we do?”, communication becomes clearer and your role becomes more strategic.

A Quick Self-Check Before Your Next Presentation

• Can I summarize my main insight in one sentence?
• What decision should change after this presentation?
• Have I translated metrics into business impact?
• Did I lead with the conclusion rather than the method?
• Have I acknowledged uncertainty clearly?
• Am I explaining to impress or to enable action?