Mistakes I made while learning data analysis (and what they taught me)

When I started learning data analysis, I thought mistakes were something to avoid.

I wanted to do things properly. Follow the right steps. Use the right tools. Produce the “right” output.

What I didn’t realise back then was that most of my progress came because of mistakes, not despite them.

Trying to analyse everything at once

One of my biggest mistakes was believing that good analysis meant analysing everything.

More metrics. More visuals. More calculations.

I felt that leaving something out meant I hadn’t done my job properly. So I tried to cover every possible angle - and often ended up with analyses that were technically correct, but hard to explain and even harder to use.

What this taught me was simple, but not easy to accept: more analysis doesn’t automatically mean better insight.

Learning to focus - and consciously leave things out - was one of the most important shifts in my thinking.

Starting with tools instead of questions

Another mistake I made early on was jumping straight into tools.

Opening Power BI. Writing SQL queries. Thinking about visuals before understanding the problem.

At the time, it felt productive. In reality, it often meant I was answering questions nobody had actually asked.

Now I know that tools should come after clarity, not before it. Without understanding the question, even the most polished dashboard doesn’t help much.

Trying to prove that I “know enough”

For a while, I treated every project as a kind of test.

I wanted to show that I could:

• use the right techniques

• apply what I had learned

• prove that I belonged in this field

That mindset pushed me towards complexity for the sake of complexity.

Over time, I realised that analysis isn’t about proving knowledge - it’s about supporting decisions. And decisions usually need clarity, not impressive-looking solutions.

Letting go of the need to “show everything I know” made my work much more effective.

Thinking mistakes mean failure

Early on, every wrong assumption or messy analysis felt like failure.

Now I see mistakes differently. They usually point to:

• unclear questions

• missing context

• wrong priorities

In other words, they’re feedback - not proof that something went wrong beyond repair.

This shift helped me become more comfortable with uncertainty and iteration, which are a natural part of analytical work.

What these mistakes changed for me

Thanks to these early mistakes, my approach to data analysis today is much calmer and more intentional.

I spend more time thinking before analysing. I focus more on meaning than volume. I’m less worried about doing everything - and more focused on doing the right things.

If there’s one thing learning data analysis has taught me so far, it’s this:

Mistakes don’t slow you down - misunderstanding what matters does.