How AI can help you make better charts

Ever spent hours searching for the right data, wrestling with spreadsheets, or wondering if you picked the best chart type? You’re not alone. Luckily, now you can use AI to jump over any data hurdles and smooth out your chartmaking process.

In this blog, you’ll discover six practical ways AI can transform your data visualization workflow. From finding trustworthy data sources to perfecting your final design, these tips will save you time and help you create charts that are clearer, memorable, and easier to understand.

What you’ll learn in this guide

• The basics: how does it work?
• Use AI to find data
• Analyze your data
• Clean and fill missing data with AI
• Choose the best chart for your story
• Format your data for a chart
• Polish your chart

The basics: how does it work?

Artificial intelligence comes in many shapes. For this blog, we’re only going to work with chat-based tools like ChatGPT, Claude, and Gemini.

What these tools have in common is that you can interact with them through a text box, like a chat. The secret to getting better answers from AI tools is to ask better questions. Like in a regular conversation, the clearer you are, the more helpful the responses will be.

Those questions or instructions are called prompts, and here are four simple rules for writing good ones:

1. Set the scene: Tell the AI what you’re working on. Instead of “analyze this data,” try “I’m creating a report on climate change for non-scientists.”

2. Be specific about what you need: Rather than “help with my chart,” say “suggest 3 ways to visualize temperature changes over time.”

3. Show, don’t just tell: Share examples of what you like, including snippets of text, files, or images. “I want help making charts in the style of Edward Tufte.”

4. Request the format you need: Ask for “5 bullet points” or “a data table I can paste into a spreadsheet” and the AI will deliver exactly that.

So, how exactly can AI make your chart-making process smoother? Let’s explore six practical ways to put it to work.

1. Use AI to find data

AI models can be great data scouts. They can help you find good data sources lying around or help you extract data from a website, for example, by scraping it. For this blog, we’re going to attempt to create charts with historical inflation data.

Say you want to create charts showing how inflation has changed over time, but you don’t know where to even start looking for that data.

Here’s a prompt that can solve this:

📋 Copy

Pro tip: Always ask for “publicly accessible” data to ensure you can actually download and use what you find!

2. Analyze your data

AI can be your storytelling partner. It can spot patterns you might miss, explain trends, and even connect your numbers to real-world events. It’s like having a data analyst on speed dial.

The previous prompt suggested OECD data, so let’s have a look at what the numbers say:

📋 Copy

This prompt should give you a summary of the key patterns in the data, historical context to understand certain behaviors, and a country-based overview.

Using AI you can get other insights, like:

• General trends and patterns
• Top and bottom values
• Rankings
• Changes over time

3. Clean and fill missing data with AI

Sometimes, you’ll want to complement your data with extra information, like a new series or a categorical column. Other times, you’ll want to drop columns from your dataset that you’re not using.

For example, this inflation data has lots of unnecessary columns, and is missing a column with regions to do comparisons between continents. Instead of manually editing on a spreadsheet, you can try a prompt like this:

📋 Copy

You can tailor this prompt further by adding more information, like specific countries or regions.

4. Choose the best chart for your story

After analyzing our data, we can use AI to find the best chart to tell our story. You can brainstorm and test different chart types and ask your tool for pros and cons.

A good approach is to mention the goal of your chart. For example:

📋 Copy

You can even ask it to suggest templates available on Flourish. The previous prompt would look like this:

📋 Copy

If you’re curious, here are the 3 charts recommended:

• Line chart (multi-series)
• Area chart (stacked)
• Small multiples lines chart

Explore our templates

5. Format your data for a chart

Once you have data, you’ve analyzed it, and you have a good idea of how you want to visualize it, then it’s time to put it in the right format to make a chart.

In Flourish, different chart types have different data requirements, so there isn’t a one-size-fits-all data format, and sometimes, you need to do more than just swap your rows and columns to get your data to work.

For this project, we’ll aim to create a grid of charts to show inflation rates by region, but we need to be mindful that the structure of a grid of charts looks very different than that of a simple line chart.

To go from our original dataset into a grid-friendly dataset, we need to do 3 things:

1. Use the new Regions column as our categorical variable to split the data by
2. Replicate each year by region - Transform from one row per year to multiple rows (one per region per year)
3. Selectively populate country data - Only show values for countries that belong to each specific region

Rather than wrestling with pivot tables or writing complex formulas, we can simply ask AI to handle this transformation with a clear prompt:

📋 Copy

And with that, we’re able to transform a simple line chart into a region-based grid of charts!

6. Polish your chart

You can use AI as your personal chart critique to take your charts to the next level. The following prompt will return a list of enhancements that can improve this grid of charts:

📋 Copy

The key suggestions it came up with are:

• To standardize the Y axis for better comparisons
• To add a filter or to remove some series from the data for more readability
• To color regions in a single color to make a visual grouping and reduce clutter
• To add annotations pinpointing key events that might have had an effect on this metric

All great points that follow best practices and focus on clearer storytelling.

Here’s the final chart with it’s improvements:

Let’s recap

In this blog, we’ve covered how you can use AI to:

• Find reliable data sources faster
• Analyze patterns you might miss
• Enrich datasets with useful context
• Choose the right chart type for your story
• Wrangle data into the perfect format
• Enhance your final design with expert-level critique

Now it’s your turn! Try these prompts in your next chart and supercharge your Flourish workflow. Feel like sharing? Tag us on social media using #madewithflourish and let us know how you are using AI to make your chart better and more impactful!