How AI Made Me Better at Excel (Not Replaced Me)

A prospect contacted us last week saying they found Office Mango through ChatGPT.

They'd been talking through their Excel challenges with AI, and we came up in the list of companies it suggested could help!

It’s made me want to reflect on my journey with AI and share my thoughts about how I think about AI + Excel.

I'll be honest - I was skeptical about AI and Excel at first. Really skeptical.

But over the past few months, I've found myself using AI more and more. Not to replace my expertise, but to enhance it.

Everyone's terrified AI will replace Excel experts. I'm quietly confident it's going to make what we do (and what you can do) even more valuable.

Here's why...

The Foundation Still Matters

I've noticed something interesting about how people use AI for Excel.

They'll ask ChatGPT to:

solve a problem —> get a solution —> copy it into their spreadsheet —> celebrate when it works.

But then something changes. A new column gets added. A different scenario comes up.

And suddenly, they're stuck.

They don't know how to modify what AI gave them because they don't understand what it actually does.

The Power Tool Problem

Think about it like this: AI is a power tool. It's incredibly effective when you know how to use it properly.

You know a drill will put a screw in. But if you use it like a hammer, you'll be there a while.

The screw might eventually go in, but is it actually secure?

Is the tool being used correctly?

Or are you just forcing something that sort of works?

Without foundational knowledge, you can't guide the tool properly. You're just hoping it does what you need.

What ‘Foundation’ Actually Means

For me, Foundational Knowledge means I can ask AI the right questions.

I know what functions exist. I understand how Excel structures data. I can recognize when a solution makes sense vs. when it's overcomplicated.

When AI suggests a formula, I can evaluate it:

• Will this actually work for my scenario?

• Is there a simpler approach?

• What happens when my data changes?

The Blind Trust Problem

This is especially dangerous with macros and VBA.

I've seen people ask AI to generate macro code, paste it in, and have no idea what it's actually doing to their data.

It works once.

But they can't debug it.

They can't modify it.

And they definitely can't explain it to anyone else.

Foundation isn't about being an Excel expert. It's about knowing enough to guide AI effectively and evaluate what it gives you.

And if you take nothing else away from this - DON’T BLINDLY TRUST VBA CODE!

You need to know the source, you need to know what it does - and if you don’t know what it does, you need to find someone you trust to explain it.

AI as Your Debugging Partner

Here's where AI really shines for me and the best use-case I’ve found

Not creating solutions from scratch, but to improve what already exists.

I don't ask AI to "build me a sales tracker."

I ask it to "debug this formula that's returning #N/A errors in specific scenarios."

Big difference.

Debugging Complex Formulas

You know that feeling when you've built a complex nested formula, and it works for 95% of your data, but breaks in edge cases?

That's where AI is brilliant.

I can paste my formula, describe the problem, and AI will often spot the issue faster than I would staring at it for 20 minutes.

But here's the key: I understand the formula well enough to evaluate whether AI's fix actually solves the problem or just masks it.

Another idea around debugging would be to share the formula with AI and to have it generate edge cases for you to test!

Optimisation, Not Creation

I use AI to optimise existing solutions all the time:

• "Is there a more efficient way to structure this calculation?"

• "Can this be simplified without losing functionality?"

• "What's causing this spreadsheet to slow down?"

AI is fantastic at suggesting improvements to something that already works. It's far less reliable at building something from nothing.

Why "Create This For Me" Doesn't Work

When people ask AI to create entire solutions, they get generic templates that don't account for their specific business context.

AI doesn't know your data structure, your edge cases, your reporting requirements, or your team's skill level.

But when you've already built something and need to refine it? That's where AI becomes incredibly valuable.

The Partnership Approach

I build the foundation. AI helps me refine it.

I understand the business problem. AI suggests technical improvements.

I evaluate the suggestions. AI accelerates the process.

It's a partnership, not a replacement.

Expanding Your Knowledge, Not Replacing It

One of the best things AI has done for me is accelerate my learning.

I'm not using it to replace my Excel knowledge - I'm using it to expand what I know faster than I ever could before.

AI as a Learning Accelerator

When I come across a new Excel function or technique, I can ask AI:

• "Explain how LAMBDA functions work with practical examples"

• "What are the real-world use cases for XLOOKUP vs INDEX-MATCH?"

• "Show me three different approaches to solving this problem"

Instead of spending hours reading documentation or watching tutorials, I get targeted explanations that directly address my specific question.

And, because of how AI works, you don’t need punchy and keyword-heavy search terms - you can have a conversation and go through a really natural learning curve with new knowledge.

Getting Ideas and Exploring New Approaches

AI is brilliant for brainstorming alternative solutions.

I'll describe a problem I'm trying to solve, and AI might suggest an approach I hadn't considered:

• "Have you thought about using Power Query for this instead of formulas?"

• "This could be solved with conditional formatting rather than complex IF statements"

It expands my thinking beyond my usual patterns.

Understanding the "Why" Behind Suggestions

But here's the critical part: I always ask AI to explain WHY its suggestion works.

Not just "here's the formula" - but "why does this approach work better than alternatives?"

Understanding the reasoning behind solutions means I can adapt them to future problems, not just copy-paste for this one scenario. (again, building that Foundational Knowledge)

The Danger of Blind Copying

People think AI is like a magic Formula generator.

They copy solutions without understanding them. It works once. Then they're stuck when they need to modify it or apply the same logic elsewhere.

AI should be teaching you, not just doing the work for you.

If you can't explain how the solution works, you haven't actually learned anything - you've just borrowed someone else's knowledge temporarily.

And really, if you rely on AI in every scenario, for every task….are you really that valuable to your organisation?

The Prompt Advantage (Foundation + AI)

If you have foundational Excel knowledge, you can write prompts that get exponentially better results from AI.

You don’t need an answer….you need the right answer for your specific situation.

The Difference in Prompts

Someone without Excel foundation asks: "Create a formula to track sales"

Someone with Foundational Knowledge asks:

"I need to track Sales by Region and Category. The dataset is on the sheet ‘SalesData’ and this report is on ‘MonthlyDashboard’. I have a dynamic range extracting Unique Categories from the dataset and putting them in a vertical list starting in cell C4 on the MonthlyDashboard sheet. Then, as headers, the Unique Regions are extracted and displayed horizontally. In cell B2 is the current Month for the Dashboard to display, with the Date column on the dataset starting in cell I2. Ensure that this formula is fully dynamic to allow for future Categories and Regions.”

See the difference?

The second prompt gets a solution that actually solves the real problem, not just a generic template.

You could also let the AI know the Header Structure of your datasets so it can figure our that Dates are from Cell I2 for example as well (as an additional tip!)

Knowing Exactly How to Instruct AI

When you understand Excel deeply, you can give AI the context it needs:

• Your current data structure

• The specific edge cases you're dealing with

• Performance constraints (file size, calculation speed)

• Your team's skill level for maintaining the solution

• Integration requirements with other systems

AI can't guess these details. You need to know what information matters.

Why Excel Experts Get Better AI Results

I can look at AI's suggestion and immediately evaluate:

• "Will this scale when my data grows?"

• "Is this the most efficient approach, or just the first one that works?"

• "What happens when someone else needs to maintain this?"

• "Are there potential errors I need to account for?"

Beginners can't ask these questions because they don't know what to look for.

And I’ll very rarely use the first thing it comes back. I’ll spot something, challenge it, guide it, it’s a back-and-forth exercise to get the result you need.

The Exponential Value

Foundational Knowledge + AI doesn't is actually Foundational Knowledge * AI.

You can solve problems faster than ever before.

You can explore solutions you might not have considered.

You can optimise and refine with incredible speed.

But only if you have the foundation to guide the AI effectively and evaluate what it gives you.

The Competitive Advantage

In a world where everyone has access to AI, what separates you?

It's not the AI - everyone has ChatGPT.

It's your ability to use it effectively.

The people who combine Excel expertise with AI fluency? They're going to be exponentially more valuable than those who rely on one or the other alone.

So, What Does This Mean for You?

I don’t think AI isn't replacing Excel experts, I think the gap between beginners and experts is going to grow.

If you have Foundational Knowledge and know how to use AI effectively, you're exponentially more valuable.

If you're just copying AI solutions without understanding them, you're becoming less valuable every day.

I think you have two options:

Option 1:

Keep using AI as a magic formula generator.

Copy-paste solutions you don't understand.

Hope they keep working.

Become increasingly dependent on tools you can't control.

Option 2:

Build your foundational Excel knowledge.

Learn to use AI as a partner, not a replacement.

Develop the skills that make you irreplaceable.

Ready to Build Your Foundation?

If you're ready to stop copying AI solutions and start using AI to multiply your Excel expertise, let's talk.

We offer a free "Let's Explore" consultation where we can discuss:

• Where your Excel foundation needs strengthening

• How to use AI effectively for your specific challenges

• What training approach would work best for your situation

No obligation, no sales pitch - just honest advice about developing skills that will make you exponentially more valuable.

📧 Email: info@OfficeMango.co.uk
🗓️ Book directly: Let's Explore Consultation

Because in a world where everyone has access to AI, your competitive advantage is knowing how to use it properly.