A junior data analyst in Nepal spends most of the week in Excel, SQL and reports that managers read again and again. So a good data analyst course in Nepal spends most of its hours there, and treats machine learning as a later step.
We read data analyst job ads on Merojob and Froxjob, and a 2026 career guide from Kumari Job. Every ad asked for Excel and SQL. None of them asked for machine learning.
What does a junior data analyst in Nepal do all week?
Put the duties from those ads side by side, and the week has four kinds of work.
- Pulling data. You write SQL to get rows out of a database or a data warehouse.
- Cleaning it. You fix missing values, repeated rows, broken dates and names typed in different ways.
- Keeping reports alive. You build a dashboard once, then refresh it and check it for managers.
- Answering quick questions. A manager asks why sales fell in one branch, and wants the answer fast.
The Kumari Job guide says it in fewer words. It calls the junior level "focused on data cleaning and basic reporting".
Cleaning is the boring part, and in Nepal it has its own trap. Many offices record dates in Bikram Sambat, and the fiscal year starts in Shrawan, not January. A report that reads those dates as ordinary English dates gives a wrong answer that still looks right. Nobody will catch it for you, so you learn to check.
Why does SQL decide the job test?
The Kumari Job guide says most entry-level data analyst tests in Nepal now include a basic SQL question. That holds even for roles that look like Excel work on paper. The interview often adds an Excel task or a small case study.
So learn SQL early, and go past SELECT *. The first idea that trips beginners is WHERE against HAVING. WHERE filters rows before they are grouped. HAVING filters the groups after GROUP BY has counted or added them up.
That sounds small. It is the difference between "branches with sales over one lakh on a single day" and "branches with total sales over one lakh". Learn to explain it with a small table of your own.
Is Excel or Power BI more important for data jobs in Nepal?
Excel comes first. The Kumari Job guide calls it the single most used tool in Nepali offices. On a junior ad we read for freshers, advanced Excel and SQL were "essential", while a dashboard tool like Power BI was "a plus".
Power BI comes right after. The same guide calls it the most requested visualization tool in Nepal. An ad for a Power BI analyst role asked for data modelling and DAX, the formula language inside Power BI. That role also wanted two years of experience.
Check your laptop before you plan around Power BI. Power BI Desktop is free, but it runs only on Windows 10 or later. Microsoft also lists a minimum screen size of 1440x900, and says 1024x768 and 1280x800 screens are not supported. Many older and budget laptops have screens smaller than that, so look under Settings, System, Display before you start.
Where does machine learning fit?
Later, but it still counts. The junior BI ads we read asked analysts to forecast performance and build simple business models. That is the road from analyst to junior data scientist.
Our view is plain. A course that starts machine learning in the second week, before you can write a join, gives you a model you cannot feed with clean data. Learn to get the data and trust it first. The model is the easy part after that.
How should you judge a data analyst course in Nepal?
Read the syllabus week by week and count the weeks. Most of them should go to Excel, SQL, cleaning and dashboards. Ask whether you practise on messy data, and whether the SQL reaches joins and window functions.
Here is how we do it at Saarathi Academy in Old Baneshwor, Kathmandu. Our Data Science and Machine Learning course runs for 12 weeks.
Week 3 is Pandas and analyst Excel: pivot tables, XLOOKUP and SUMIFS, and cleaning that includes Bikram Sambat dates. Week 4 is SQL in PostgreSQL, from GROUP BY and HAVING to joins and window functions. Week 8 is a full week of Power BI, with a Nepali fiscal year date table. Machine learning starts in Week 9.
The last week has timed SQL drills and a mock interview with a SQL round and an Excel practical. Because Power BI Desktop needs Windows, Mac users set up their own Windows before Week 8. Each classroom batch has up to 10 students, and the full syllabus is public before anyone pays.
What can you try tonight?
Two free things, about an hour each.
Open Google Sheets and paste any table you have, such as a month of your own spending. Choose Insert, then Pivot table, and total it by category. Google's pivot table help page shows each step.
Then open the PostgreSQL tutorial page on aggregate functions. It runs GROUP BY and HAVING on a small weather table. Change the condition and guess the output before you look.
If you want to check your laptop for Power BI too, Microsoft keeps the Power BI Desktop requirements on one page.
The full 12 week plan is on our Data Science and Machine Learning course page, and the syllabus lists every week.
Top comments (0)