The Actual Problem
The job is real. The demand is real. The beginner supply has exploded — and companies can tell immediately.
Data Analytics is not a bad career choice. It is a career where the gap between a ₹3L outcome and a ₹15L outcome is entirely determined by one thing: whether you built generic skills or domain-specific capability. Your course taught you tools. Tools are a commodity. The students getting hired are the ones who can translate a business problem into a data question — not just run a SQL query or build a Tableau dashboard. This page tells you how to become that person before you graduate.
The Key Numbers
₹3–4L
generic fresher data analyst — Python cert + Excel + no domain depth. This market is oversupplied and salaries are falling.
₹10–15L
fresher with domain depth in fintech, healthtech, or e-commerce — same tools, completely different interview outcome
₹15–25L
analyst at an MNC or GCC with genuine specialisation and a business-impact portfolio — accessible within 2 years
2 years
time to build the domain depth that separates the ₹3L offer from the ₹15L offer — if you start deliberately now
The Real Differentiator
What Separates the Hired from the Rejected
Companies reviewing 200 identical CVs a day have developed a very fast filter. Here is exactly what it looks like from both sides.
❌ Gets Rejected
- 5 courses, 5 certificates, no real project
- Portfolio has Titanic survival analysis and Iris flower classification
- Says "proficient in Python, SQL, Tableau, Power BI, Excel" — every single candidate says this
- Cannot explain what business decision their project was meant to support
- Applied to 200 roles with the same generic CV
- Domain knowledge: zero — could be working on any industry
✓ Gets Hired
- 2 projects, both in the same specific industry, both answering a real business question
- Can explain: "I analysed churn patterns for a fintech app — here is the decision it would have driven"
- Knows the KPIs that matter in their target industry (CAC, LTV, NPS, gross margin)
- Applied to 20 roles, all in one domain, with a tailored CV that shows industry language
- Has one tool stack used well — not six tools used generically
- Answers "why" — not just "what"
The Skill Stack
What to Learn, What to Skip, and What to Build Deep
Most courses teach you everything at the same depth. That is wrong. Here is the actual priority stack.
| Skill | Priority | What Actually Matters |
|---|---|---|
| SQL Must | Non-negotiable floor | Joins, aggregations, window functions, subqueries. This is tested in every data analyst interview. Spend 6–8 weeks on this alone before anything else. |
| Python Must | Non-negotiable floor | Pandas for data manipulation, Matplotlib/Seaborn for visualisation. You do not need machine learning. You do not need deep learning. Stop at analysis and visualisation until you have domain depth. |
| Excel / Sheets Must | Underrated — always used | Pivot tables, VLOOKUP/INDEX-MATCH, basic financial modelling. Most real analyst work at small-to-mid companies is done in Excel. If you cannot do it fast in Excel, you will struggle on the job. |
| Tableau or Power BI One, not both | Pick one and go deep | Learn whichever tool your target companies use — check job descriptions. Do not learn both. One tool used with business storytelling skill is worth more than two tools used generically. |
| Statistics Must | Undertaught, high impact | Hypothesis testing, A/B test interpretation, correlation vs causation, distributions. This is what separates analysts who answer "why" from analysts who only show "what." Most courses skip this or rush it. |
| Machine Learning Skip for now | Not required for most roles | Data analyst ≠ data scientist. Companies hiring analysts do not need you to build models — they need you to extract and interpret business insights from existing data. Adding ML to your CV without domain depth signals you followed a generic curriculum. |
| Domain Knowledge Must | The actual differentiator | Learn the KPIs, business metrics, and decision-making language of one specific industry. This is covered in the Domain Guide below. No course teaches this — you have to build it yourself. |
The Domain Guide
Pick One Domain. Go Deep. This Is the Whole Strategy.
Every domain below has different KPIs, different interview questions, and different portfolio expectations. Pick the one that matches your background or genuine interest — not the one that sounds impressive.
Fintech / Banking
Highest paying
KPIs to Learn
CAC, LTV, churn rate, NPA (Non-Performing Assets), credit utilisation, default probability, transaction conversion rate
Portfolio Projects
Credit card churn analysis, loan default prediction (logistic model), transaction anomaly detection, customer segmentation by spending behaviour
Fresher Salary
₹10–16L
Paytm, PhonePe, Razorpay, HDFC, ICICI GCC, Zerodha
"Fintech companies hire analysts who understand why customers churn — not just that they churn. If you can frame a data problem as 'we lost ₹X crore in LTV because of Y behaviour,' you are already ahead of 90% of applicants."
E-commerce / Retail
Most open roles
KPIs to Learn
GMV, AOV (Average Order Value), cart abandonment rate, return rate, repeat purchase rate, fulfilment cost per order, SKU-level margins
Portfolio Projects
Basket analysis (market basket / association rules), seasonal demand forecasting, customer cohort retention analysis, discount campaign ROI analysis
Fresher Salary
₹8–14L
Flipkart, Meesho, Nykaa, Myntra, Amazon India, BigBasket
"E-commerce has the most public datasets available for building your portfolio — Amazon review data, Olist Brazil dataset, and Kaggle's retail sets are all free. There is no excuse for not having a domain-specific project if you are targeting this vertical."
SaaS / Product Analytics
Fastest growth
KPIs to Learn
DAU/MAU ratio, activation rate, feature adoption, Net Revenue Retention (NRR), MRR churn, time-to-value, funnel drop-off rates
Portfolio Projects
User funnel analysis (where do signups drop off), A/B test analysis on a feature change, cohort retention curves, north-star metric identification exercise
Fresher Salary
₹10–18L
Freshworks, Zoho, Chargebee, Postman, Leadsquared, Clevertap
"Product analytics is the highest-ceiling domain for data analysts in India right now — because SaaS companies grow internationally and pull Indian analyst salaries up with them. The KPIs are specific and learnable. The projects are fully buildable on public datasets."
Healthtech / Pharma
Stable + growing
KPIs to Learn
Patient acquisition cost, treatment outcome rates, bed occupancy, diagnostic accuracy rates, claims denial rates, drug trial dropout rates
Portfolio Projects
Hospital readmission analysis (public US CMS dataset), prescription pattern analysis, patient satisfaction survey analysis, clinical trial data summarisation
Fresher Salary
₹8–14L
Practo, PharmEasy, Manipal Health, Apollo, IQVIA India
"Healthtech is ideal for students with a science background who pivoted to data — it uses your prior domain knowledge directly. Analysts here earn less at entry than fintech, but the career is stable and the work is genuinely meaningful."
The Portfolio Framework
How to Build a Portfolio That Gets You Interviews Before You Graduate
You do not need ten projects. You need two strong projects in one domain, both answering a real business question. Here is exactly how to build them.
1
Pick your domain first — then find a dataset
Do not start with a dataset and ask "what can I analyse here." Start with a business question in your target domain. Example: "Why do fintech customers stop using an app after 30 days?" Then find data that can answer it. Free sources: Kaggle, UCI ML Repository, data.gov.in, RBI Open Data, WHO datasets. The question comes before the data.
Time: 1–2 days to frame the question
2
Structure every project as a business recommendation, not a technical exercise
Every project should end with: "Based on this analysis, the business should do X because the data shows Y." Not: "I used pandas to clean the data and seaborn to visualise it." Recruiters do not care about your technical process — they care whether you can turn numbers into decisions. Your project write-up should read like a brief to a business manager, not a data science tutorial.
Format: Problem → Analysis → Recommendation
3
Use domain language, not generic data language
If you are targeting fintech, your write-up should say "customer LTV dropped 18% in Q3" — not "I observed a decrease in the target variable." If you are targeting e-commerce, say "cart abandonment spiked on mobile during checkout — the friction point is likely the payment step." Domain language in your portfolio signals to recruiters that you understand their business, not just data tools.
Rule: If a non-analyst in that industry can't follow it, rewrite it
4
Two projects is enough — build one per semester starting now
Do not wait to build a perfect portfolio at the end of your degree. Build one project now, publish it on GitHub, and write a short LinkedIn post explaining what you learned. Build the second project 4–6 months later with a harder business question. Two strong domain-specific projects beat ten generic tutorial replications every single time. Recruiters can tell within 30 seconds which one they are looking at.
Timeline: First project within 8 weeks of reading this
Salary Reality
What You Will Earn — The Honest Numbers
The difference is not talent. It is whether you built domain depth or stayed generic.
| Profile | Typical Role | Fresher Salary | After 2–3 Years |
|---|---|---|---|
| Generic — certificates, no domain | Junior Data Analyst, tier-3 company or BPO analytics | ₹3–5L | ₹5–8L — slow growth, high competition |
| Mixed — some domain, weak portfolio | Analyst at mid-size company in target sector | ₹6–9L | ₹10–14L with specialisation |
| Domain-deep — strong portfolio, business framing | Analyst at fintech, e-commerce, SaaS, or GCC | ₹10–16L | ₹18–28L |
| Specialised + 4–5 years experience | Senior Analyst / Analytics Manager at product company | N/A | ₹25–45L |
Honest Guidance
Should You Continue on This Path? The Honest Filter
Data analytics is a real career. But only for the right candidate with the right approach.
- ✓You are genuinely curious about why businesses make the decisions they make — not just how to run a query. The career rewards people who think in problems, not people who think in tools.
- ✓You are willing to pick one domain and go deep — even if it means saying no to the generic "data science" label for now. Specificity is the strategy.
- ✓You are ready to start building your portfolio before your degree finishes — not after. The students with a GitHub and two domain-specific projects get interviews. The students waiting to graduate do not.
- ⚠If you enrolled because data analytics felt like the safe, employable option and you have no preference for any domain — this is the honest warning. "Safe and generic" is exactly the profile companies are rejecting at scale. That needs to change before you apply for jobs.
- ⚠If you are on your fourth or fifth certification with no real project yet — stop adding certifications. Your time and money are better spent on one domain-specific project than on another course. The market is not counting your certs. It is asking: can you solve a real business problem?
Your Next Step
Which domain is right for your specific background?
The right domain depends on your degree, your prior experience, and what industry roles are accessible from your location. Submit your specific situation and I will give you a targeted answer — not a generic recommendation.
📋 Submit Your Specific Query 💬 Join the Free Career Community 🗓️ Book a 1:1 Session with AnshulConnect with Anshul Wadhwa
Resources & Next Steps
Put your career queries here, join the community, book a session, or follow for daily guidance.