How to use AI in data storytelling
- Kat Greenbrook
- Apr 26
- 3 min read
Updated: Jul 28
AI has made its way into most data teams. Nobody has settled on how much of the work it will end up doing, but whether it's a lot or a little, it's likely to become part of your workflow. Which raises a question for the communication side of data work: can AI help there too?
It can, but it's not as easy as you might think.
Why just "write me a data story" doesn't work
The first thing many people try is pasting in the numbers and asking for a story. What comes back might sound okay, but it tends to be generic, because the model is missing everything that makes a data story resonate with its audience.
A data story works because of its context. It's written with an understanding of what your organisation is trying to achieve, who your audience is, and what you want them to do. Most often, that information lives in your head (or someone else's head). AI can help you write a data story, but only if you give it that context first.
How AI helps with data storytelling
Used well, AI helps to widen your view. Here are a few places in the data storytelling process where I've found AI to be useful:
Seeing your work in its wider setting. The goal of your work contributes to bigger organisational outcomes, and other work contributes to your goal. AI can help you map both. This is the beginning of how a data story creates organisational impact.
Calibrating to an audience. Given a clear description of who you're talking to, AI can help you think through what they may already understand, what motivates them, and how you might need to adjust your communication.
Finding reasons you hadn't considered. When you've seen a change in your data, it's sometimes easy to explain it entirely through your organisation's own actions. While these may be a driver, AI is good at suggesting potential causes from outside that frame—the market, policy shifts, customer behaviour, the wider environment.
Surfacing what you don't know. Every analytics project carries unknowns: things the data can't tell you. AI can name these as open questions, which may be helpful before you get in front of your audience.
Drafting more than one version of the story. From the same information, AI can produce two structurally different narratives, each with a different tension at its centre and a different call to action at its end. Comparing them side by side helps you see previously invisible editorial choices.
You decide which of AI's suggestions are right for your context.
How YOU can use AI for data storytelling
I've created a free Templates + Prompts file: customisable AI prompts built around the templates from The Data Storyteller's Handbook.
The prompts build on each other (each output becomes a future input) and they work in whichever AI tool you already use. A worked example is also available, showing the full chain from start to finish. And if you'd rather skip using AI, the templates stand on their own.
AI can complement your data storytelling process, but you have to be the storyteller.
Kat Greenbrook is a data storytelling consultant, author, and workshop facilitator based in Wellington, New Zealand. She is the founder of Rogue Penguin and the author of The Data Storyteller's Handbook.
