How I approach a data analysis project before opening any tools

When I started learning data analysis, my instinct was to open a tool as quickly as possible.

Excel.

SQL.

Power BI.

Doing something felt productive.

Over time, I realised that the most important part of a data analysis project often happens before any tool is opened.

Starting with the problem, not the data

Before looking at tables, charts or queries, I try to understand the actual problem.

Not the technical task - the reason behind it.

I ask myself:

• Why is this analysis needed?

• What decision is someone trying to make?

• What would change as a result of this analysis?

Without clear answers to these questions, even the most detailed analysis can miss the point.

Understanding who will use the analysis

Another step I take early on is thinking about the audience.

Who will look at the results?

• A manager?

• A stakeholder without a technical background?

• Someone who needs a quick answer, not a deep dive?

This matters because the same data can be analysed in many ways - but not every way is equally useful for every audience.

Clarity often comes from tailoring the analysis to the person who needs it.

Defining what “good enough” looks like

One of the most important lessons I’ve learned is that not every project needs maximum depth.

Before opening any tools, I try to define what good enough means for this specific case.

• Is a high-level overview sufficient?

• Do we need precise numbers or clear trends?

• Is this a one-off question or something that will be revisited regularly?

Setting this boundary early helps avoid unnecessary complexity later.

Deciding what not to analyse

This part used to be uncomfortable for me.

Leaving things out felt like doing less than I could.

Now, I see it differently.

Choosing what not to analyse is just as important as choosing what to focus on. It helps keep the analysis relevant, readable and aligned with the original goal.

Only then do tools come in

Once the problem, context and expectations are clear, tools become exactly what they should be - support, not the starting point.

At that stage:

• Excel helps explore and structure ideas

• SQL helps ask precise questions of the data

• Power BI helps communicate insights clearly

But none of them replace the thinking that happens beforehand.

How this approach changed my work

Approaching projects this way made my analysis calmer and more intentional.

I spend less time reworking outputs.I ask better questions earlier. I focus more on decisions than on deliverables.

Most importantly, the analysis feels purposeful — not just technically correct.

Good data analysis doesn’t start with tools. It starts with understanding the problem you’re trying to solve.

This way of thinking shapes how I later use tools like Excel, SQL or Power BI in my projects.