Thinking like a Data Analyst

🌟 by Jolie Pham | April 2, 2025

Let’s say we’ve just received a new dataset. And as expected, it’s messy, overwhelming, and full of numbers that don’t seem to tell a clear story. As data analysts, our job is not just about crunching numbers, it’s about transform that chaos into valuable insights.

But how do we make sense of raw data and turn it into something actionable?

It’s all start with mindset.

Think Business First and Know your Audience

Why thinking business first is critical?

Data is not simply just data. Every dataset exists because of a business process, and every insight must tie back to a business decision. If we don’t take the time to understand the business context, we risk spending hours on analysis that doesn't provide real value. Worse, we might misinterpret the data and draw conclusions that lead to poor decisions.

Ask yourself

What problem is the business trying to solve?

Who are the key stakeholders, and what decisions do they need to make?

How will the insights drive business strategies?

Evaluate the data quality

Before diving into SQL queries or dashboards, let’s take a step back. Dataset can be misleading if they contain missing values, duplicates, or inconsistencies.

Data Analysts should think about

Do categorical values follow a consistent format? (e.g., “Male” vs. “M”)

Do we see impossible numerical values (e.g., negative age)? We don’t have -5 years old.

Are there any missing values?

Are there any duplicates? If so, do they present true errors or repeated data?

Seeing a bigger picture

Every dataset has a story, but to understand it, we need to see how values are spread across different variables. This helps us identify trends and detect anomalies.

Data Analysts should think about

Does the distribution suggest normality or skewness?

Are there any outliers that need attention? If so, is it data entry errors, or extreme but valid cases?

Spotting patterns

Data Analysts should think about

How does the data change over time?

Are there any sudden spikes or dips that require investigation?

Can we forecast future trends based on the past patterns?

Are there recurring seasonal patterns?

Peak Holiday Seasons

Increased travel, hotel bookings, and gift shopping due to summer/winter vacations.

Black Friday

Massive discounts drive a surge in consumer spending on all items, specifically cosmetics and clothing.

Year-end bonuses

Employees receive extra income, boosting spending on vacations, luxury goods, fine dining, and investments.

Back-to-School Shopping

Parents and students purchase school supplies, clothing, and electronics in preparation for the new academic year.

Who matters?

For dataset containing customer factor, identifying patterns in customer behavior helps businesses tailor their strategies.

Data Analysts should think about

Which characteristics define customer groups?

How do customer behaviors differ across segments? (e.g., frequent buyers vs. One-time purchasers)

Which customer segments contribute most to company's revenue?

Which customer segments are likely to contribute most to profitability in the future?

Which customer segments having the highest and lowest Retention Rate? Which factors contribute to the high retention rate?

Are there specific customer segments showing an extremely high Customer Churn Rate? If so, what underlying factors drive their decision to leave?

Hidden insights from missed opportunities

Not all insights come from what did happen, some of the most valuable ones come from what didn't happen. Businesses lose money not just when customers make bad purchases, but also when potential customers walk away.

Why ignoring missing opportunities is wasteful?

Imagine a hotel only looks at successful bookings but ignores cancellations. They might see a 70% occupanacy rate and think that everything is fine. But cancellations accounts for 25%. That's a significant lost revenue opportunities.

Another common missed opportunities would be Shopping Cart Abandonment. Let's say your marketing team has spent weeks preparing a big sale campaign. Many people actually add products to their carts. And then...nothing happened. They abandoned their carts and leave without purchasing.

This isn't just an inconvenience - it's wasted effort and of course, lost revenue.

So why did they leave?

Too high shipping costs? The checkout process was too complicated?

Whether the reasons are, insights from missed opportunities can truly help businesses optimize their operating performance.

Data Analysts should think about

Why do some customers book once but never return?

What are factors leading to their decision to cancel their bookings?

Conclusions

A real data analyst never just "analyze data" - we solve business problems. We start by understanding the dataset, and stakeholders before diving into our data.

A golden rule is to always be curious about the story behind the data!