Hello hello!

So, you’ve finally got your models built, and your numbers are accurate.

But now you are dreading the 3 hours you are about to spend in PowerPoint, trying to make a waterfall chart that doesn't look like a tax form.

And the crazy thing?

By the time you've finished the visuals, the conversation has changed already.

One person wants to see a different scenario. Another wants it broken down by region. So, you end up back in Excel, rebuilding everything.

But what if I told you there's a way to create board-ready financial visuals in 10 minutes instead of 3 hours?

And I'm not talking about templates or fancy BI tools that take weeks to set up.

Today, I'm going to show you exactly how to do this. You don’t need to be a data visualization expert, and you don’t need to spend thousands on new software.

All you need is 30 minutes this week. It will change how you present financial data forever.

​

​
From Presentation Death Loop to Real Time Visuals

Here's what happens in most companies.

Thursday afternoon: Your team asks for a board presentation showing Q4 performance across all business units.

Now you have the task of turning 47 tabs of Excel into something the board can (actually) understand. All formatted to match the corporate template.

So you spend Friday afternoon in PowerPoint. Copying data from Excel, building a chart, formatting the axes, adjusting the colours and adding labels.

Then you realize the chart type is wrong. So you start again.

Three hours later, you have only five slides done.

Saturday morning: Email from the team. "Can we show this by region instead of by product? And then add a scenario where we grow 15% faster in EMEA?"

So, you're back in Excel. Rebuilding the model. Re-creating the charts. Re-formatting everything.

This isn't strategic finance work. This is design hell.

And you are not built for this.

Here's what’s changed with AI:

You need to stop treating visualization as the last step.

You need to start treating it as part of the modelling process.

What does this mean?

Instead of:

1. Building the models in Excel
2. Copying the numbers into PowerPoint
3. Formatting the charts manually
4. Repeating this when something changes

You do this:

1. Build the model with AI
2. Generate the visuals instantly
3. Adjust it in real-time when the requirements change

The tool I'm going to show you is ChatGPT Canvas. And it's built specifically for this type of workflow.

You don't export the data and import it somewhere else. And, you don't need to switch between five different applications.

Instead, you build the model and the visuals in the same place, so when something changes, everything updates together.

This is not about making prettier charts. It's about reducing the 3-hour gap between ‘model complete’ and ‘presentation ready.’

And once you see how this works, you'll never go back to the old way.

​

​
How to use ChatGPT Canvas in 5 Steps

Now, how do you do this?

To do this in ChatGPT, enable Canvas.

Canvas is a side-by-side workspace where you and the model co-create charts, tables, and slides in one place.

You don’t need to context-switch between your AI chat and your documents.

Step 1: Open Canvas

Type “/canvas” (or select if from your menu by clicking ‘+’).

Step 2: Provide Data

Paste a small table or upload a CSV/XLSX document. State your units (e.g currency; thousands/millions) and your time granularity (e.g Month/Quarter/Year).

If you have created a financial model in the chat already, the AI will use the data you have uploaded previously and generate the visuals based on that.

Step 3: Ask for the visual

Specify your chart type and chart dimensions.

Example prompt:

“Build a variance waterfall for 2025 vs 2024 from this table. Bars: Price, Volume, Mix, FX, Cost. Show both $ and pp impact. Add labels and a total column.”

Step 4: Refine

Request this like, axis labels, rounding, color-blind-safe colours, and annotations (e.g., “flag if the margin is < 20%”).

Step 5: Export

Copy the chart image into your slides or brief.

Pro Tip - Keep the data and the prompt under the chart for your audit trail.

Here's a short demo where I used ChatGPT Canvas to generate a dynamic presentation in html (the language used to create websites):

Prompt (copy/paste)

Good finance visuals to request

• Variance waterfall: Actual vs Plan (or vs LY) with drivers (price/volume/mix/cost/FX).
• Bridge charts: Revenue → EBITDA → FCF.
• Cohort heatmap: customer revenue by start quarter (retention/expansion).
• Small multiples: GM% by product line across quarters.
• Working-capital dashboard: DSO, DIO, DPO, CCC trend with target bands.

Quality checks (always)

• Do the labels show units and rounding? Do the totals reconcile to the table?
• Do the axes start at logical baselines, with no misleading scales?
• Do callouts for thresholds work? (e.g., “Gross Margin % is below 22%”).
• Do the footnotes list sources and the periods covered?

As a result of all of this, you will have no more financial presentations that are overwhelming data dumps that your colleagues struggle to understand.

Instead, you can now transform your data into clear, dynamic, and persuasive visual stories.

Time to let your creativity shine.

​

​

The Bottom Line

You didn't become a finance pro to spend your weekends in PowerPoint. And you definitely do not want the presentation to take longer to put together than the analysis (this is backwards). Canvas changes this equation. Build your model -> generate the visuals -> iterate in real-time. What used to take three hours now takes ten minutes. So here's the framework: Build with Canvas enabled - Start your financial model in ChatGPT with Canvas from the beginning Generate visuals as you go - Don't wait until the end to visualize, create charts during your analysis Iterate in the meeting - When someone asks "what if we grew faster in EMEA?", adjust it live So, this week, pick one financial model you need to present. Build it using ChatGPT Canvas. Generate the visuals. Then, see how it feels to adjust everything in real-time. Try it once, and you'll understand straight away why some finance teams are moving three times faster than everyone else. Make sure you are one of them! ​ Best, Your AI Finance Expert, Nicolas ​ P.S. - What did you think of this approach? Hit reply and let me know if you're planning to try this for your team (I read all replies). ​ P.P.S. - I've highlighted this in my financial models video, along with other use cases using GPT-5. Watch it now!​