You close the tab. What next?

If you are feeling daunted, you are not alone. There's no need to learn Excel, Python, Power BI, and SQL all at once (or in that order). This post will tell you what to learn next and what to leave out. The suggested order is Excel, SQL, BI (Power BI/Tableau), and then Python. And here's why. Read on to discover why.

Let's dive right in!

What is data analytics?

Let's cut through the hype and get to the heart of it: a data analyst answers questions. Why are sales dropping? Which ad campaign gets the most customers? Which product gets returned the most?

A data analyst collects and cleans data, finds patterns, and explains what they mean (usually to someone who cannot read numbers). The last part is one of the most important skills to learn because it can separate you from other analysts and is the reason companies pay analysts six-figure salaries. (Think about it: A brilliant analysis that only the C-suite understands is not very useful.)

The Roadmap: At-a-glance (Time required: approx)

These are all ball-park figures for someone who can dedicate a few hours every day to learning.

Step 1: Excel, your foundation

Excel may not be the most glamorous tool out there, but it is one of the most widely used tools in the corporate world and a great starting point for data analytics.

What to learn:

• Cleaning (duplicates, blanks)
• Formulas (lookups, etc)
• Pivots (your first baby-analysis)
• Charts (and how to make beautiful ones)
• Power Query (transforming data)

wscubetech

Step 2: SQL, the language of databases

In reality, data does not live in Excel. More often than not, it lives in databases (think of Excel's equivalent). SQL is the language you use to extract this data (hence why it stands for Structured Query Language).

What to learn:

• SELECT, WHERE, ORDER BY
• GROUP BY, aggregate functions (SUM, etc)
• JOINS
• Cleaning/filtering using queries
• Window functions

One roadmap recommends moving on to this step only after completing at least one project on your own. (Trust me, this is good advice.)

Placementpreparation

Step 3: Power BI, turning analysis into stories (or dashboards)

Do you want to make your analysis more accessible to non-technical people? Dashboards are a great way to do this since they allow you to showcase visuals (charts, tables, etc) and let people interact with them.

What to learn:

• Importing and cleaning data (Power Query)
• Basic modelling
• Visualisations (charts, tables, etc)
• Some DAX (formulas)
• Dashboard design: Readability, focus

Power BI vs Tableau? Both are BI tools that can be learned as a beginner. You should choose one and stick to it (ideally for life). They are similar, but Power BI is slightly better for people who know Excel.

Placementpreparation

Step 4: Python, making you a complete data analyst

Python is not a beginner-friendly programming language. It is, however, a great way to make you marketable to more employers since it has applications in data science, too. It is not recommended that you learn Python as a first step, but it is a worthwhile investment if you ever want to level up in your data analytics career.

What to learn:

• Python basics (variables, loops, etc)
• Pandas (working with data)
• Cleaning data, basic analysis
• Making simple plots with Python (Matplotlib/Seaborn)
• Basic statistics (mean, median, mode, standard deviation, etc)

Step 5: Projects and portfolio, what employers want to see

Projects (and not certificates) are what set you apart from the rest of the pile. Aim to complete 3–5 projects:

1. An excel project with a pivot table
2. A SQL project with some queries
3. A Power BI project
4. A Python project with some analysis
5. An end-to-end project (applying all you have learned)

For each of these, write a short description of the problem, action, and findings. Display it on a webpage (portfolio) or your GitHub profile.

Sample 6-month plan

This is basically an excerpt from any 6-month data analyst roadmap (they all have slight variations). As mentioned, this depends on how much time you can dedicate to learning analytics every day. Do not get disheartened if you are not able to meet these targets; keep practising and you will get there.

Common mistakes by beginners

• Trying to learn Excel, Python, Power BI, and SQL all at once. Stick to one tool at a time.
• Wasting time on tutorials: Try to practise while watching tutorials. Try to code along.
• Neglecting statistics: It is not important to learn complex statistical concepts, but you should know averages, trends, etc
• Using 'clean' data: This will do more harm than good. Try to practice on messy data.
• Creating dashboards without insights: Remember that dashboards are meant to showcase insights, so start with an insight
• Not focussing on communication: Always remember to explain your analysis in simple language.

Do you need a degree or coding background?

Not necessarily. A beginner can learn data analytics without any coding experience. You can learn Excel, SQL, Power BI/Tableau, and Python one after the other. That said, many employers do value a graduate degree (especially for managerial positions). A combination of a degree and a strong portfolio is ideal for most entry-level analyst jobs.

Pw

Learning with guidance in Delhi

Self-learning works if you are someone who is disciplined. If you are someone who prefers guided practice, institutes in Delhi can teach you practice tasks and give you feedback on your mistakes, saving you months of confusion.

The bottom line

You should not try to learn everything this month. Start with Excel, followed by SQL, then a BI tool (Power BI/Tableau), and finally Python. Choose a dataset this week and answer one question with it. That is data analytics in a nutshell, and all big skills have to start somewhere.