
Here’s a number: 1.43°C.
That’s roughly how much warmer 2025 was than the pre-industrial average, according to the World Meteorological Organization. It’s one of the most important numbers in the world.
How did it make you feel?
Probably not much. That’s not because you don’t care. It’s because human brains aren’t built to feel decimals. We evolved to respond to faces, voices, places and stories, not to anomalies measured against an 1850–1900 baseline.
This is the central challenge of climate communication. The data is extraordinary, alarming and clear. And on its own, it barely moves people.
Here’s how to change that.
Step 1: Find the “so what”
Every dataset contains a headline. Your first job is to find it.
Ask yourself: what is the single most surprising, important or consequential thing in this data? Then write it in one sentence that a fifteen-year-old would understand.
Not: “Analysis of rainfall records indicates a statistically significant decline in long-rains totals over the period.”
But: “The long rains in this county have become less reliable over the last thirty years.”
If you can’t write that sentence, you haven’t found the story yet.
Step 2: Translate to human scale
Big numbers need translating into things people can picture.
- Billions of tonnes of emissions mean nothing. A comparison to something familiar means something.
- Per person and per household numbers land better than national totals.
- Time is relatable: “every minute”, “every school year”, “since you were born”.
- Place is relatable: “an area the size of Nairobi”, “from here to Mombasa”.
When I talk about the WMO’s finding that 2015 to 2025 were the eleven warmest years on record, I often put it this way: every child who started primary school in 2015 has spent their whole school life in the hottest years ever measured.
Same data. Suddenly it’s about someone’s child.
Step 3: Find the face
Data tells you what is happening. A person tells you what it means.
Pair your numbers with someone who lives them. A rainfall chart becomes powerful when you meet the farmer whose planting calendar it describes. A sea-level graph becomes real when you stand with a family whose well has turned salty.
Always with consent, always with dignity, always letting people speak in their own words.
Step 4: Choose the right visual
The wrong chart can bury a strong finding.
- Change over time: a line chart.
- Comparing categories: a bar chart.
- Parts of a whole: use sparingly. A simple stacked bar often beats a pie.
- Location: a map.
- One dramatic number: sometimes the best visual is just the number, very big.
And the most important rule: write the takeaway as the title. Not “Temperature data, 1990–2025” but “Nairobi’s hottest days are getting hotter.” Never make your reader work out what the chart means.
Step 5: Be honest about uncertainty
Climate science deals in ranges, probabilities and confidence levels. It’s tempting to drop them for a cleaner story. Don’t.
Explain uncertainty simply: “Scientists estimate between X and Y.” “This is very likely, meaning at least a 90% probability.” Honesty about what we don’t know is what makes people trust what we do know.
Step 6: Cite your source
Link to the original data. It protects your credibility, lets others verify and models good practice.
A worked example
The data: satellite and rainfall records show drought conditions have become more frequent in a pastoralist county over two decades.
The weak version: “Drought frequency has increased significantly in the region, according to recent analysis.”
The story version: “When she was a girl, the river near her village ran all year. Today it’s dry for months at a time, and in bad years, most of the year. She’s not imagining it. Satellite records show drought now visits her county far more often than it did twenty years ago. Her family has gone from forty goats to twelve.”
Notice the structure: person, change, evidence, consequence. The data hasn’t been weakened. It’s been given a home.
(A quick note: when you write a story like this, it must be a real person and real numbers from your fieldwork, shared with consent. Never invent people to illustrate data. The moment an audience discovers a fabricated example, every number you’ve ever shared becomes suspect.)
Practise this week
Take one statistic from your latest report. Run it through the six steps. Share the before and after with a colleague and watch their reaction.
Data storytelling is one of the core skills I’m teaching through ImpactComms. Explore my impact communications services, or read Climate Storytelling 101.
Frequently asked questions
How do you turn data into a story?
Find the one surprising “so what”, translate it to human scale, pair it with a person who lives it, choose a clear visual and be honest about uncertainty.
What chart should I use for climate data?
Line charts for trends over time, bar charts for comparisons and maps for location, each titled with its main takeaway.
Cover photo: Deng Xiang on Unsplash


