If you’re new to data analytics, you’ve probably run into the same vague pitch everyone gives: “we find insights in data” or “we help businesses make decisions.” Sure, but that doesn’t tell you much about what a data analyst does on a typical workday.

So instead of another definition, let’s explore a real dataset — India’s UPI payment system — and see what a data analyst does with it, step by step.

Step 1: Start with a question, not a spreadsheet.

Analysts don’t open with a pile of numbers and hope something jumps out. They start with a question someone genuinely needs answered. For UPI, that question is: did digital payments just change how Indians pay, or did they also change how much Indians spend? Everything that follows exists to answer that one question. That’s really the whole job in a sentence.

Step 2: How Does a Data Analyst Choose the Right Metrics?

Numbers on their own don’t mean much without the right context. For UPI, the useful metrics are transaction volume, transaction value, and average transaction value.

Transaction volume shows how many payments happened. Transaction value shows how much money moved. Average transaction value is the value divided by volume.

Volume alone can be misleading. A million ₹10 chai payments and a million ₹10,000 purchases look the same if you’re only counting transactions. Average transaction value shows whether growth comes from more purchases or bigger ones.

Step 3: How Data Analysts Find Turning Points, Not Just Totals

This is where the real work happens. One clear example from UPI’s data: in FY 2022-23, person-to-merchant (P2M) payments overtook person-to-person (P2P) transfers for the first time. A good analyst doesn’t just log that number and move on — they recognize it as a turning point. It’s the moment UPI stopped being mainly a way to pay back a friend and became mainly a way to buy things.

Step 4: Compare across categories, not just across time.

Solid analysis never stops at “X grew.” It asks, “grew compared to what?” UPI accounts for roughly 85% of India’s digital payment volume, but only about 9.5% of its value — and that gap is the real story. It shows UPI dominates small, everyday payments while big transfers still move through other channels. Without that comparison, “UPI is huge” tells you almost nothing. “UPI is huge for small transactions” tells you a lot.

Step 5: Know the difference between what the data proves and what it merely suggests.

This is the step beginners skip most often. It is arguably the most important one. UPI’s growth and India’s overall spending have both increased over the same years. So, it’s tempting to say UPI caused people to spend more.

However, inflation, e-commerce growth, rising smartphone ownership, and higher incomes were also increasing at the same time. Any one of these factors could explain the rise in spending.

A careful analyst makes this clear: correlation isn’t causation. The data shows that UPI changed how India pays. But whether it changed how much India spends is something the current data cannot confirm.

Step 6: Say it clearly, uncertainty and all.

Finding the pattern isn’t the finish line — someone still has to explain it to people who don’t have time to dig through NPCI tables themselves. A good analyst states the facts plainly (UPI processed 24,162 crore transactions worth ₹314 lakh crore in FY 2025-26) and is just as plain about the parts that are interpretation, not fact — using words like “likely” or “the data doesn’t confirm this” instead of dressing up a guess as a certainty. That honesty is what makes an analysis worth trusting. Data analytics is also changing as AI becomes part of the analyst’s workflow. While AI can automate many data-related tasks, human judgment is still important for asking the right questions, interpreting results, and communicating insights. To understand how AI is changing the data analyst role, read our guide on “Will AI Replace Data Analysts in 2026?“.

The takeaway: data analytics isn’t really about having access to big numbers. It’s a discipline — ask a real question, pick the right metrics, find genuine patterns, make fair comparisons, and be upfront about what you don’t actually know. UPI just happens to be one of the richest live datasets in India to practice all of that on right now.

(Data from NPCI, RBI Annual Reports 2024-25/2025-26, and RBI Payment Systems Report, as of August 2026.)