When people hear the words “data analysis”, many picture complicated formulas, powerful computers and charts that only mathematicians can read. In reality, analysis is simply an organised version of something we all do every day. A data analyst observes in exactly the same way, but uses numbers, tables and clear methods to do it. In this article we explain in plain language what data analysis is, which stages the work goes through, which tools are used and how you can start learning. At the end there is a small exercise you can do in an ordinary spreadsheet.
What data analysis is
Data analysis is the process of answering a question using collected facts (numbers, records, responses) and helping someone make a decision. The goal is not a pretty chart but a useful conclusion: what is happening, why it is happening and what to do next.
Analysis can be roughly divided into three levels:
- Descriptive — shows what happened. For example, how much bread the shop sold last month.
- Diagnostic — looks for the reasons. For example, why sales are lower on Mondays.
- Predictive — estimates what may happen. For example, how much stock to order before a holiday.
Beginners usually start at the first level, and that is perfectly fine.
Everyday examples
Analysis is not only for large companies. Let us look at some familiar places.
The neighbourhood shop
Say Dilshod runs a small grocery shop and writes down in a notebook how many items of each product he sells every day. If, after a few months, those notes are moved into a spreadsheet, interesting patterns appear: dairy products sell out in the morning, while sweets are bought more at weekends. As a result Dilshod plans his orders by day of the week, and less food goes to waste.
The school
A teacher, Nodira, records her pupils’ test results in a spreadsheet over the term. When she calculates the average score for each task, she sees that the questions on fractions were answered noticeably worse than the rest. Now she knows exactly which topic to revisit.
The clinic
If the reception desk notes every day how many patients came in and how long they waited, it becomes possible to identify the hours with the longest queues. On that basis the doctors’ schedule can be adjusted to cut waiting times.
The farm
A farmer, Bahodir, records which variety he planted in each field, the irrigation dates and the size of the harvest. Comparing several seasons, he can see which variety does better on which soil.
The people in these examples do not call themselves analysts, yet they are doing exactly that: asking a question, gathering data and drawing conclusions.
The stages of analytical work
Professional analysts and beginners alike follow roughly the same sequence.
- Define the question. Everything starts with a clear question. “Is the shop doing well?” is vague. “Which three products sell best, and on which weekdays do they run out fastest?” is specific. The sharper the question, the more useful the answer.
- Collect the data. Work out where the information lives: a notebook, a till program, a survey, an electronic register. Sometimes you will need to collect it yourself.
- Clean the data. This is the most time-consuming stage, and the most important. Real data is full of errors: one product spelled two ways (“Bread”, “bread ”), dates in different formats, empty cells, duplicated rows. Uncleaned data leads to wrong conclusions.
- Analyse. Group the data, calculate totals, averages and percentages, compare periods.
- Visualise. Present the result as a chart or table. A good chart is understood at a glance.
- Conclude and recommend. Finally, write in plain words what you found and what you suggest doing.
Remember that these stages sometimes loop. If a new question appears during analysis, you can go back and collect more data.
Tools: from simple to advanced
Analysts use many tools, but they are easy to understand when sorted into levels.
Spreadsheets
Excel, Google Sheets or LibreOffice Calc are the starting point. They offer sorting, filtering, formulas (SUM, AVERAGE, COUNTIF), pivot tables and charts. In many small and medium-sized organisations all analysis is done in spreadsheets alone. Google Sheets is free and runs in the browser, so there is nothing to install.
SQL
When there is a great deal of data stored in a database, analysts use SQL. It is a language for asking the database “give me the records that match these conditions”. It reads a lot like English and is logical, so people who are not programmers learn it too.
Python
Python is a programming language that analysts use together with libraries such as pandas. It is handy for automating repetitive tasks, processing large files and doing more complex calculations. For a beginner it is the next step, not the first one.
BI dashboards
Tools such as Power BI and Looker Studio are used to build interactive reporting dashboards. A manager opens the dashboard, chooses a period or region and sees the figures for themselves.
An important tip: do not try to learn everything at once. Being confident with spreadsheets is already a valuable skill, and the rest builds on it gradually.
Personal qualities that help
Many people think: “I’m not good at maths, so analysis isn’t for me.” In fact, basic analysis needs only school arithmetic — adding, dividing, percentages and averages. Personal qualities matter even more:
- Curiosity — the habit of asking “why is that?”.
- Attention to detail — spotting a small error, such as text typed instead of a number in one row.
- Patience — cleaning data can be tedious, but without it there is no reliable result.
- Logical thinking — telling cause apart from coincidence.
- Explaining simply — making a result clear to someone who does not like numbers.
- Honesty — showing the result as it is, even when it is not what people hoped for.
Women who have managed a household budget, tracked family spending or followed their children’s school marks often already have these qualities. If you are returning to work after a break, do not underestimate that experience.
How to start learning
This plan is realistic for most people and works step by step:
- Start with spreadsheet basics. Cells, formulas, sorting, filters, conditional formatting. Half an hour of practice a day brings noticeable progress within a few weeks.
- Work with your own data. Household spending, electricity meter readings, a reading diary — learning on familiar data is far more engaging.
- Master pivot tables and charts. This skill is asked for in many jobs.
- Learn basic statistics: mean, median, maximum and minimum, percentage change.
- Then move on to SQL. The core commands do not take long to learn.
- Do small projects and keep them. Each one should have a question, data and a conclusion. Together they become your future portfolio.
- Find like-minded people. Studying in a group, asking questions and reviewing each other’s work keeps motivation alive.
You can start even with slow internet or an old computer: Google Sheets works on a phone, and LibreOffice is free and runs well on modest machines.
Mini exercise: your first analysis in a spreadsheet
Now let us move from theory to practice. The exercise takes 30–40 minutes.
Scenario: you are analysing one week of sales for a small shop.
Step 1. Create a table. Make four columns: “Date”, “Day”, “Product”, “Quantity (items)”. Enter made-up figures for one week for three products — bread, milk and tea. That gives 21 rows. For example: Monday, Bread, 40; Monday, Milk, 25; Monday, Tea, 8.
Step 2. Add mistakes on purpose. In one row type “bread ” (with a trailing space) instead of “Bread”; in another, type the word “thirty” instead of a number. This is what real data often looks like.
Step 3. Clean it. Turn on the filter and look at the list of values in the “Product” column — you will see two different “breads”. Make them identical. Fix the number written as a word.
Step 4. Calculate. Find total sales for each product with SUMIF, for example: =SUMIF(C:C,"Bread",D:D). Then work out the average daily quantity for each product. (Depending on your regional settings, the formula may use semicolons instead of commas.)
Step 5. Build a pivot table. Put “Day” in rows, “Product” in columns and “Quantity” in values. Now you can see at a glance which product sold best on which day.
Step 6. Make a chart. Create a column chart from the pivot table.
Step 7. Write a conclusion. Three sentences are enough: which product sold most, which day had the lowest sales, and what you would advise the shop owner.
Common mistakes
- Starting without a question. Opening data just to “see what turns up” often wastes time.
- Skipping the cleaning. One duplicated row or misspelled name can spoil the whole result.
- Confusing correlation with cause. Two figures rising together does not mean one causes the other. In summer, sales of both ice cream and electric fans go up, but neither causes the other — the heat drives both.
- Big conclusions from little data. Planning a whole year from one week of observations is risky.
- Cluttered charts. Ten colours and five measures on one chart tire the reader out.
- Careless handling of personal data. When working with information about patients, pupils or customers, remove names and phone numbers if you do not need them, and never forward files to outsiders. Check the legal requirements in this area with official sources such as lex.uz.
Conclusion
Data analysis is not magic; it is organised thinking and the sensible use of simple tools. It is useful in a shop, a school, a clinic and a field alike. To begin, you do not need an expensive computer or advanced mathematics: curiosity, patience and a little daily practice are enough. Open a spreadsheet today, ask one question and try to answer it with data — that will be your first analysis.
If you would like to learn this field in a structured way, alongside other learners, take a look at our association’s free programmes: they include data analytics, computer literacy and Data Science. Once you have chosen a programme, you can sign up through the application page.