Data Storytelling: How to Make Numbers Compelling (Not Boring)

Turn spreadsheets into stories that drive decisions — techniques from 24 years of presenting to boards, credit committees, and investors

I once watched a colleague present 47 slides of flawless analysis to a credit committee. Every number was accurate. Every chart was properly labelled. The recommendation was sound.

They said no.

The problem wasn’t the data. It was the delivery. He presented numbers. He should have told a story with numbers. That’s the difference between data presentation and data storytelling — and it’s the difference between getting polite nods and getting decisions.

After 24 years in banking — presenting to boards at JPMorgan, credit committees at RBS, investors at Commerzbank — I’ve learned that the analysts who get promoted aren’t the ones with the best spreadsheets. They’re the ones who make data mean something.

What Is Data Storytelling (And Why It Matters)

Data storytelling is the practice of building a narrative around data to help your audience understand and act on insights. It combines three elements: the data itself, the visualisation, and the narrative that connects them.

Here’s why it matters:

Data alone doesn’t persuade. Stanford research found that statistics presented with stories are 22 times more memorable than statistics alone. Numbers tell people what. Stories tell people why it matters.

Decisions are made emotionally. Neuroscientist Antonio Damasio’s research shows that people with damage to emotional brain centres can’t make decisions — even with perfect logic. Your CFO may think they’re purely analytical, but they’re not. Nobody is.

Attention is limited. The average executive spends 2-4 minutes reviewing a slide before moving on. If your data doesn’t land immediately, it doesn’t land at all.

Data storytelling isn’t about dumbing down your analysis. It’s about making your analysis accessible to people who don’t have time to interpret it themselves.

Related: How to Present to a CFO: The Finance-First Framework

The Data Storytelling Framework: Lead With Insight

Most presenters structure data slides like this:

Here’s the data → Here’s what it shows → Here’s what we should do

That’s backwards. By the time you reach your point, you’ve lost them.

Effective data storytelling reverses the order:

Here’s the insight → Here’s the data that proves it → Here’s what we should do

This is the “lead with the headline” approach. Your audience knows immediately what they’re looking at and why it matters. The data becomes evidence, not a puzzle to solve.

Example: Before and After

Before (Data-First):

“Q3 revenue was £4.2M. Q2 was £3.8M. Q1 was £3.5M. Year-over-year we’re up 12%. The EMEA region grew 18% while Americas grew 6%…”

The audience is doing mental maths, trying to figure out the point.

After (Insight-First):

“EMEA is now our growth engine — up 18% while Americas stalls at 6%. If we shift Q4 marketing budget accordingly, we can capture another £400K.”

Same data. Completely different impact.

Related: The Executive Summary Slide: How to Write the Only Slide That Matters

5 Data Storytelling Techniques That Work in Business

These are the techniques I use with clients — from biotech fundraising decks to banking board presentations.

1. The Comparison Anchor

Numbers mean nothing without context. “£2.3 million” is abstract. “£2.3 million — that’s 3x what we spent last year for half the results” creates meaning.

Always anchor your data to something your audience already understands:

• Compare to last year / last quarter
• Compare to competitors or industry benchmarks
• Compare to targets or forecasts
• Compare to a familiar reference point

Example: “Our customer acquisition cost is £47. The industry average is £62. We’re 24% more efficient — and here’s why that matters for our Q1 targets…”

2. The Single Number Focus

When everything is important, nothing is important. Pick the one number that matters most and build your slide around it.

I learned this the hard way. Early in my career, I’d cram every relevant metric onto a slide. The result? Decision-makers couldn’t see the forest for the trees.

Now I ask: “If they remember only one number from this slide, what should it be?” That number gets visual prominence. Everything else supports it.

3. The Trend Line Story

A single data point is a fact. Multiple data points are a trend. Trends tell stories.

Weak: “Churn rate is 4.2%”

Strong: “Churn has dropped from 6.1% to 4.2% over eight months — the interventions are working”

When presenting trends, always explain the inflection points. What happened in March that changed the trajectory? That’s where the story lives.

4. The “So What” Test

For every data point, ask yourself: “So what?”

“Revenue grew 12%” — So what?
“Revenue grew 12%, which means we’ve hit our trigger for the expansion budget” — Now I understand why this matters.

If you can’t answer “so what” for a piece of data, it probably doesn’t belong in your presentation.

5. The Contrast Frame

Show what the data could have been — or what it will be if nothing changes.

Example: “At current trajectory, we’ll miss target by £800K. With this intervention, we close the gap entirely.”

Contrast creates stakes. Stakes create attention.

Related: Team Dashboards That Tell a Story (Not Just Show Numbers)

Common Data Storytelling Mistakes (And How to Avoid Them)

After reviewing hundreds of data presentations, I see the same mistakes repeatedly.

Mistake 1: Showing all the data. Your analysis might require 50 data points. Your presentation needs 5. The rest belongs in the appendix. Include only what’s necessary to support your narrative.

Mistake 2: Letting the chart speak for itself. No chart is self-explanatory to a busy executive. Always add a headline that states the insight, not just a label that states the topic. “Q3 Revenue by Region” is a label. “EMEA Drives 70% of Q3 Growth” is an insight.

Mistake 3: Choosing the wrong chart type. Pie charts for trends. Bar charts for composition. Line charts for 15 data points. Match the visualisation to the story you’re telling:

• Trends over time → Line chart
• Comparison between categories → Bar chart
• Part-to-whole relationships → Pie or stacked bar (with few segments)
• Correlation → Scatter plot

Mistake 4: Burying the lead. The most important insight should be visible within 3 seconds. If your audience has to hunt for the point, they won’t.

Mistake 5: No clear action. Data without a recommendation is just information. Always end data slides with what you want the audience to do with this information.

Data Storytelling in Practice: A Real Example

A biotech client came to me with a fundraising deck. Their data slide looked like this:

Title: “Clinical Trial Results”
Content: A table with 12 rows of efficacy data, p-values, confidence intervals, and patient subgroup breakdowns.

Scientifically rigorous. Completely ineffective for investors who see 20 decks a week.

We restructured it:

Title: “87% Response Rate — 2x the Standard of Care”
Content: One large number (87%), one comparison bar showing vs. standard of care (43%), and a single line of supporting text about statistical significance.

The detailed data moved to the appendix. The story stayed on the slide.

They raised £18 million.

Related: Storytelling in Presentations: The NLP Techniques That Captivate Any Audience

Frequently Asked Questions About Data Storytelling

How do I tell a story with data without oversimplifying?

Simplifying isn’t dumbing down — it’s respecting your audience’s time. Keep the full analysis available (in appendix or backup slides) but lead with the insight. If someone wants to drill into methodology, they’ll ask. Most won’t.

What if my audience wants to see all the numbers?

Some audiences do — especially technical or financial reviewers. In these cases, structure your presentation in layers: executive summary with key insights first, then supporting detail, then full data appendix. Let them choose their depth.

How do I present data that tells a negative story?

Lead with the insight anyway — but frame it constructively. “We’re 15% behind target” is a problem. “We’re 15% behind target, and here’s the recovery plan that closes the gap by Q4” is a story with a path forward. Never hide bad data; contextualise it.

How many data points should one slide have?

As few as possible to make your point. For most business presentations, that’s 1-3 key metrics per slide. If you need more, ask yourself if you’re actually making multiple points that deserve multiple slides.

Should I use AI tools for data visualisation?

AI can help generate initial visualisations, but always review and refine. Tools like Copilot are good at creating charts quickly but often miss the storytelling elements — the headlines, the annotations, the “so what.” Use AI for speed, then add the human insight layer.

---

Your Next Step: Apply the Insight-First Framework

Data storytelling isn’t a talent — it’s a technique. Start with one change: on your next data slide, write the insight as your headline, not the topic.

Instead of “Q3 Sales Performance,” write “Q3 Sales Exceeded Target by 12% — Here’s What Drove It.”

That single shift transforms how your audience receives the information.

---

Mary Beth Hazeldine spent 24 years in corporate banking at JPMorgan, PwC, Royal Bank of Scotland, and Commerzbank — where she learned that the analysts who get promoted aren’t the ones with the best spreadsheets, but the ones who make data mean something.