Short Term
Intermediate to Advanced4 monthsCertification
Machine Learning
The algorithms, properly understood. You implement the core models before reaching for a library, so you know why one fails and another works.
- Python
- scikit-learn
- pandas
- +2
Certification programme
Data Analytics Course in Mumbai
The complete analyst toolkit in one programme: spreadsheets, SQL, visualisation and Python, applied to real datasets and finished with a dashboard you can defend in an interview.
- Level
- Beginner to Advanced
- Duration
- 4 months
- Commitment
- 6 hrs / week
- Mode
- Classroom or Online
Short Term
- Excel
- MySQL
- Power BI
- Python
- pandas
What you will be able to do
Outcomes
- Clean and reshape messy real-world data
- Write analytical SQL against production-shaped schemas
- Build dashboards that answer business questions
- Communicate findings to non-technical stakeholders
At a glance
Your learning path
Full syllabus
What you actually learn
8 modules · 43 topics · 165 concepts
Thinking Like an AnalystThe part that is not a tool: turning a vague business question into something measurable.
Framing a question
- the ask behind the ask
- defining the metric
- choosing a comparison
- what decision the answer will drive
Metrics
- rates versus counts
- averages and their failure modes
- medians and percentiles
- vanity metrics
Data literacy
- population versus sample
- correlation and causation
- survivorship bias
- Simpson paradox
The analysis workflow
- ask, prepare, process, analyse, share, act
- documenting assumptions
- when to stop analysing
Stakeholders
- interviewing the requester
- managing scope
- delivering an answer they can act on
You can take an ambiguous request and return a defined, answerable question.
Spreadsheets for AnalystsExcel as an analysis tool rather than a place to type numbers.
Cleaning
- text to columns
- TRIM and CLEAN
- removing duplicates
- finding inconsistent categories
- dates stored as text
Lookups and joins
- XLOOKUP
- INDEX-MATCH
- joining two exports
- handling unmatched rows
Summarising
- pivot tables
- grouping
- calculated fields
- percentage of total
- slicers
Power Query
- repeatable cleaning steps
- unpivot
- merging queries
- refreshing next month
Statistical functions
- descriptive statistics
- correlation
- trendlines
- the Analysis ToolPak
You can clean, join and summarise a dataset entirely in Excel.
SQL for AnalysisThe single most-asked skill in analyst interviews.
Core querying
- SELECT and WHERE
- ORDER BY and LIMIT
- NULL handling
- CASE expressions
Joins
- inner and outer joins
- joining several tables
- the accidental fan-out
- verifying row counts
Aggregation
- GROUP BY
- HAVING
- conditional aggregation
- getting the grain right
Window functions
- ROW_NUMBER and RANK
- LAG and LEAD
- running totals
- top N per group
CTEs and structure
- WITH clauses
- breaking a question into steps
- readable analytical SQL
Interview SQL
- the question patterns that recur
- cohort and retention queries
- writing under time pressure
You can answer a business question directly against a database.
Statistics That Actually Get UsedThe statistical ideas an analyst is expected to apply, without the proofs.
Descriptive statistics
- central tendency
- spread and standard deviation
- skew
- quartiles and box plots
Distributions
- normal distribution
- the empirical rule
- long-tailed data
- why averages mislead on skewed data
Sampling and uncertainty
- sampling methods
- sampling error
- confidence intervals
- sample size intuition
Hypothesis testing
- null and alternative
- p-values in plain language
- significance versus importance
- common misinterpretations
A/B testing
- designing a test
- control and treatment
- minimum detectable effect
- peeking and why it invalidates results
Regression
- simple linear regression
- interpreting coefficients
- R-squared
- when regression is the wrong tool
You can quantify uncertainty and test whether a difference is real.
Visualisation and DashboardsCommunicating a finding so the decision actually gets made.
Choosing a chart
- comparison, composition, distribution and relationship
- when a bar beats a pie
- the chart types to avoid
Design principles
- data-ink ratio
- axis honesty and truncated scales
- colour with meaning
- accessible palettes
Power BI
- connecting and modelling
- star schema basics
- DAX measures
- slicers and drill-through
- publishing
Tableau basics
- the interface
- dimensions and measures
- building a dashboard
- where it differs from Power BI
Storytelling
- leading with the finding
- annotation
- structuring a five-slide readout
- presenting to people who disagree
You can build a dashboard a stakeholder reads correctly without you present.
Python for DataWhere Python overtakes Excel: size, repeatability and anything that must run twice.
Python essentials for analysts
- data types
- lists and dicts
- functions
- Jupyter workflow
pandas fundamentals
- Series and DataFrame
- reading CSV, Excel and SQL
- selection with loc and iloc
- filtering
Transforming
- groupby and aggregate
- merge and join
- pivot and melt
- apply and vectorisation
- method chaining
Cleaning at scale
- missing data strategies
- outlier detection
- type coercion
- normalising inconsistent categories
Visualising in Python
- Matplotlib basics
- Seaborn for statistical plots
- when a notebook chart beats a dashboard
Exploratory analysis
- a structured EDA checklist
- distributions and relationships
- documenting what you found
You can do an end-to-end analysis in pandas and explain why it beats a spreadsheet.
Getting Data From AnywhereThe half of the job where the data is not handed to you in a clean file.
APIs
- REST basics
- requests and JSON
- pagination
- rate limits and keys
- flattening nested responses
Web scraping
- BeautifulSoup
- selecting elements
- handling pagination
- legality and robots.txt
File formats
- CSV encoding problems
- Excel with merged cells
- JSON
- Parquet and why analysts meet it
ETL thinking
- extract, transform, load
- idempotent reruns
- incremental loads
- where a scheduled script belongs
Data quality
- validation rules
- reconciliation against a source of truth
- documenting known issues
You can acquire and consolidate data from files, APIs and web pages.
Portfolio and InterviewTurning the skills into an offer, which is a separate skill again.
Building a portfolio
- choosing a dataset with a real question
- end-to-end projects over notebook dumps
- GitHub and README structure
The case interview
- structuring an answer out loud
- guesstimates
- metric definition questions
- what to do when you do not know
Take-home assignments
- scoping the time
- documenting assumptions
- what reviewers actually check
SQL and Excel rounds
- live query rounds
- common test questions
- practising under a clock
Positioning
- analyst versus MIS versus BI roles in Mumbai
- reading a job description properly
- tailoring a CV to it
You can present a project and answer a case interview convincingly.
5 builds you keep
What you will build
- Project 01
Why did last quarter miss target?
A deliberately ambiguous brief and a messy sales export. You define the question, find the driver, and deliver a five-slide answer with the assumptions written down. The actual job, rehearsed.
- Project 02
A cohort retention analysis
Group customers by joining month, track return behaviour, and show where the drop-off is. Built in SQL with window functions, then visualised. The analysis every subscription business lives on.
- Project 03
An A/B test read-out
Given results from a real experiment design, determine whether the difference is real, state the confidence, and make a recommendation. Includes the case where the honest answer is "not enough data yet".
- Project 04
A refreshable operations dashboard
Raw data from three sources, modelled properly, published with scheduled refresh. Built so it still works next month without you touching it.
- Project 05
Capstone: an end-to-end analysis from an API
Pull data you had to acquire yourself, clean and store it, analyse it in pandas, and present a finding. Documented on GitHub as a portfolio piece you can walk an interviewer through.
Hands on with
Tools and technologies
- Excel
- MySQL
- Power BI
- Python
- pandas
Before you enrol
Data Analytics questions
Do I need a maths or engineering background?
No. Commerce, science and management graduates do well here. The statistics module teaches what an analyst actually applies and skips the proofs. What matters more is comfort with logic and a willingness to question numbers.
Excel, SQL, Power BI and Python — is that not too many tools?
It is exactly what the job descriptions ask for, and they build on each other rather than competing. Excel teaches the shape of data, SQL gets it, Power BI communicates it, and Python is for anything that has to run more than once.
Is this the same as data science?
No, and the difference matters when applying. Analytics answers what happened and why, using SQL, visualisation and statistics. Data science predicts what will happen, using machine learning. This course is the analyst path; we teach the other separately.
What salary should a fresher expect in Mumbai?
It varies enough by company type that any single number would be misleading, and we would rather not quote one. What we can say is that analyst roles here ask for SQL first and visualisation second, and that a portfolio project you can defend moves the conversation more than a certificate does.
How much of my time will this need?
Sixteen weeks at six hours of class a week, and you should plan on a similar amount again for project work. The people who come out employable are the ones who did the projects properly rather than just attending.
Where this leads
Pair it with these and you are job-ready, not course-ready
One skill on a CV is a hobby. These are the courses students take alongside this one, in the order that builds an actual role.
Short Term
Beginner to Intermediate2 months
Linux and Operating Systems
What happens beneath your program, taught at the Linux terminal so the theory stays concrete. Doubles as the foundation for cloud and cybersecurity work.
- Linux
- Bash
- VirtualBox
- +1
Short Term
Beginner2 months
Video Editing with AI
Editing is now half craft and half tooling. Learn the craft properly, then use AI to remove the tedious parts: transcription, captions, rough cuts and cleanup.
- DaVinci Resolve
- Adobe Premiere Pro
- CapCut
- +1