[Project] Consumer Behaviour & Shopping Habits (Power BI)

Why I worked on this project

Customer data is often reduced to averages. Average spend. Average order value.

Average customer.

But the more I worked with data, the more obvious it became that the “average customer” rarely exists in real life.

This project started with a simple curiosity: Do different customers actually behave in different ways — and if so, what does that mean for business decisions?

The question behind the data

From a business point of view, understanding customer behaviour goes far beyond knowing how much people spend.

The real questions are:

• Do all customers shop with the same frequency?

• Are high spenders also loyal customers?

• Which behaviours are worth investing in - and which are not?

Treating all customers as one group may simplify reporting, but it often hides patterns that matter most.

How I explored customer behaviour

I used Power BI as an exploratory tool rather than a reporting one.

Instead of focusing on a single KPI, I looked at customer behaviour from multiple angles:

• purchase frequency,

• total spend,

• basket size,

• shopping patterns over time.

By combining these perspectives, I could start identifying behavioural differences between customers rather than forcing them into one “average” profile.

The focus was on finding patterns, not confirming assumptions.

What patterns started to appear

As the data came together, several clear behavioural differences emerged.

Some customers purchased frequently but spent relatively little per transaction. Others bought less often, but their orders were consistently higher in value.

There were also groups whose overall contribution looked similar at first glance - but came from completely different shopping habits.

Seeing these patterns side by side highlighted how misleading single-metric analysis can be when working with customer data.

Why this matters for business decisions

Understanding how customers behave changes the type of decisions a business can make.

Insights like these can inform:

• targeted marketing strategies instead of broad campaigns,

• customer retention efforts focused on the right segments,

• more realistic expectations around growth and loyalty.

Rather than asking “Who spends the most?”, the more useful question becomes: “Which behaviours are most valuable to the business - and why?”

Tools used

• Power BI for data modelling, exploration, and visual storytelling

The emphasis was on clarity and interpretation rather than advanced technical features.

Want to see the technical details?

The full project, including the Power BI report and supporting materials, is available on GitHub

Final thought

This project reinforced something I keep coming back to in analytics: good customer insights rarely come from a single number.

Behavioural context matters - and without it, even accurate metrics can lead to the wrong conclusions.

This project reminded me how easily important behavioural differences disappear when we rely too heavily on averages.