Popular starting points

Professional CertificateIntermediate

Data Analytics + Power BI

Excel, SQL, data cleaning, Power BI, dashboards, visualisation and business reporting.

Talk to a counsellor about Data Analytics + Power BI

Fees, batch dates and timings are shared by a counsellor. No payment is needed to ask.

What you will cover
  • Excel
  • SQL
  • data cleaning
  • Power BI
  • dashboards
  • visualisation
  • business reporting

Topic list from the institute's programme record.

Certification

Certificate terms for this programme are confirmed by the institute at admission. Any ADIVAS course-completion certificate is an institute certificate; it is not a university degree, an externally accredited qualification or a vendor certification (such as AWS or Microsoft) unless the institute states so in writing.

Curriculum 0.1 · Draft: written for ADIVAS and awaiting academic review and approval. Topics and weeks may change before your batch starts.

Your skill trail

You move through 6 stations. Each week builds one skill you can show, not just a topic you have heard.

  1. Spreadsheet foundations

    Handle data in Excel with correct types, formulas and pivot tables.

    • Week 1 Data, tables and Excel basics
    • Week 2 Formulas and lookups
    • Week 3 Pivot tables and first charts

  2. Cleaning and preparing data

    Turn messy files into trustworthy tables with a recorded cleaning log.

    • Week 4 Cleaning messy data
    • Week 5 Missing data, outliers and validation

  3. SQL for analysis

    Ask questions of a database with filters, joins and aggregates.

    • Week 6 Databases and SELECT
    • Week 7 Aggregates and GROUP BY
    • Week 8 Joins
    • Week 9 Subqueries and analytical queries

  4. Power BI modelling and DAX

    Load, model and calculate business measures in Power BI.

    • Week 10 Getting data into Power BI
    • Week 11 Data model and star schema
    • Week 12 Measures with DAX

  5. Dashboards and interpretation

    Design honest dashboards and interpret numbers with care.

    • Week 13 Visual design and dashboards
    • Week 14 Interpreting data responsibly

  6. Ethics and capstone

    Apply privacy rules and deliver a complete reporting project.

    • Week 15 Privacy, security and sharing
    • Week 16 Capstone: business reporting project

  7. Regional Retail Sales Dashboard

    Your project, reviewed by a trainer.

The ADIVAS mastery ladder: how each skill is recognised
  1. Explored You have seen the idea and tried a first example.
  2. Practising You can do it with hints and keep improving.
  3. Proficient You complete the weekly practical task correctly on your own.
  4. Proven You apply it in a project reviewed by your trainer.

Skill-level tracking is being added to the learner workspace. Your trainer already reviews weekly practical work.

What you will be able to do

  • Organise and analyse tabular data in Excel using tables, formulas, lookups and pivot tables.
  • Clean messy data: fix types, remove duplicates, handle blanks and standardise text, while keeping a record of every change.
  • Query a relational database with SQL using filters, joins, grouping and simple subqueries.
  • Build a tidy data model in Power BI with a star schema and write basic DAX measures.
  • Choose a chart that fits the question, and design dashboards that are readable, honest and not misleading.
  • Interpret results correctly: separate trend from noise, avoid mistaking correlation for cause, and state what the data cannot show.
  • Respect data privacy and ethics: mask personal data, apply role-based access and document sources.
  • Deliver a business report with a dashboard, a short written insight and a documented data pipeline.

Who it suits

  • School Students in Classes 10-12 who like numbers and want to see how data turns into charts and decisions, starting with a spreadsheet.
  • College Commerce, science and engineering students who want practical data skills for projects, internships and further study.
  • Professional Accountants, sales and operations staff and small-business owners who need reliable reports from sales, stock or attendance data.

Prerequisites: Comfort with a computer and basic spreadsheet use (rows, columns, SUM). Basic arithmetic including percentages. No programming needed. Tools: Microsoft Excel (or LibreOffice Calc for early practice), Power BI Desktop (free; Windows only), SQLite or PostgreSQL with a free client such as DB Browser for SQLite, Provided practice datasets: retail sales, attendance and a small hospital queue log (all synthetic).

Weekly plan

Spreadsheet foundations Weeks 1–3

Week 1: Data, tables and Excel basics

Theory
  • Rows, columns, data types (number, text, date)
  • Excel tables and structured references
  • Sorting, filtering and number formats
  • What makes data tidy
Lab
Convert a 300-row retail sales sheet into an Excel table, correct text stored as numbers and dates, and sort and filter to answer five questions.
Homework
Collect a small real dataset (for example your monthly expenses) and set it up as a tidy table.
Deliverable
A tidy sales table and a list of five answered questions.
Weekly checkpoint
Identify the data type problems in a screenshot; one filtering task.
You will be able to
Organise and analyse tabular data in Excel using tables, formulas, lookups and pivot tables.

Week 2: Formulas and lookups

Theory
  • SUMIFS, COUNTIFS, AVERAGEIFS
  • IF and nested conditions
  • XLOOKUP / VLOOKUP and exact match
  • Absolute and relative references
Lab
Add GST-inclusive price, a discount tier and a product category lookup to the sales table; verify three rows by hand.
Homework
Find and fix five formula errors in a provided workbook.
Deliverable
A sales workbook with formulas and a verification note.
Weekly checkpoint
Predict the output of three formulas; fix a broken lookup.
You will be able to
Organise and analyse tabular data in Excel using tables, formulas, lookups and pivot tables.

Week 3: Pivot tables and first charts

Theory
  • Pivot table rows, columns, values and filters
  • Grouping dates
  • Choosing basic charts: bar, line, pie caution
  • Slicers
Lab
Build a pivot of monthly sales by branch and category with a slicer, then chart the trend and the top five products.
Homework
Write three plain-language findings from the pivot and one question it cannot answer.
Deliverable
A one-sheet summary with two charts and three findings.
Weekly checkpoint
Choose the right chart for four questions; read a pivot to answer two questions.
You will be able to
Organise and analyse tabular data in Excel using tables, formulas, lookups and pivot tables. Choose a chart that fits the question, and design dashboards that are readable, honest and not misleading.
Cleaning and preparing data Weeks 4–5

Week 4: Cleaning messy data

Theory
  • Common problems: duplicates, blanks, inconsistent spelling, wrong types
  • Trim, clean, text-to-columns, find and replace
  • Power Query: steps recorded and repeatable
  • Never overwrite the raw file
Lab
Clean a provided customer file with duplicates, blank phone numbers, 'Madurai', 'madurai ' and 'MDU' spellings, and mixed date formats using Power Query.
Homework
Write what rule you used for each issue and what you would ask the data owner.
Deliverable
A cleaned table plus a cleaning log with issue, rule and rows affected.
Weekly checkpoint
Sort cleaning steps into the right order; identify the unsafe step in a given log.
You will be able to
Clean messy data: fix types, remove duplicates, handle blanks and standardise text, while keeping a record of every change.

Week 5: Missing data, outliers and validation

Theory
  • Handling blanks: leave, fill, remove, and why it matters
  • Spotting outliers with sorting and simple summaries
  • Data validation lists and checks
  • Reconciling totals with the source
Lab
Find an invoice entered as 45000 instead of 450, decide what to do with a missing value in three ways, and compare the totals before and after.
Homework
Reconcile the cleaned totals with the original report and explain any difference.
Deliverable
A reconciliation table and a short note on decisions made.
Weekly checkpoint
Decide how to treat a blank and an outlier in two scenarios and justify.
You will be able to
Clean messy data: fix types, remove duplicates, handle blanks and standardise text, while keeping a record of every change. Interpret results correctly: separate trend from noise, avoid mistaking correlation for cause, and state what the data cannot show.
SQL for analysis Weeks 6–9

Week 6: Databases and SELECT

Theory
  • Tables, rows, keys and relationships
  • SELECT, WHERE, ORDER BY, LIMIT
  • AND, OR, IN, BETWEEN, LIKE
  • NULL and how it behaves
Lab
Load the retail database and write ten queries, such as orders in a date range over Rs 5,000 from a given city.
Homework
Translate five business questions into SQL.
Deliverable
A file of ten tested queries with the result row counts.
Weekly checkpoint
Read five queries and state what they return.
You will be able to
Query a relational database with SQL using filters, joins, grouping and simple subqueries.

Week 7: Aggregates and GROUP BY

Theory
  • COUNT, SUM, AVG, MIN, MAX
  • GROUP BY and HAVING
  • WHERE versus HAVING
  • Date functions
Lab
Compute monthly revenue, average order value and the number of customers per city; find cities with over Rs 1 lakh revenue.
Homework
Rewrite a pivot table from week 3 as a SQL query and compare the numbers.
Deliverable
Queries and a table matching the Excel pivot.
Weekly checkpoint
Fix a query that uses WHERE where HAVING is needed.
You will be able to
Query a relational database with SQL using filters, joins, grouping and simple subqueries.

Week 8: Joins

Theory
  • Primary and foreign keys
  • INNER and LEFT JOIN
  • Why row counts change after a join
  • Duplicate rows from many-to-many joins
Lab
Join orders, customers and products, find customers who never ordered with a LEFT JOIN, and explain a row count that doubled.
Homework
Draw the table relationships for the retail database.
Deliverable
A relationship diagram and five join queries with checks of row counts.
Weekly checkpoint
Predict row counts for a join on a tiny example; spot a missing join condition.
You will be able to
Query a relational database with SQL using filters, joins, grouping and simple subqueries.

Week 9: Subqueries and analytical queries

Theory
  • Subqueries and CTEs (WITH)
  • Ranking and top N
  • Running totals and window functions (introduction)
  • Writing readable SQL and comments
Lab
Find the top three products per category, customers above the average order value, and the month-on-month change in sales.
Homework
Explain one query line by line for a non-technical manager.
Deliverable
A query set with comments and a plain-English explanation.
Weekly checkpoint
Write a query with a CTE; interpret a month-on-month table.
You will be able to
Query a relational database with SQL using filters, joins, grouping and simple subqueries. Interpret results correctly: separate trend from noise, avoid mistaking correlation for cause, and state what the data cannot show.
Power BI modelling and DAX Weeks 10–12

Week 10: Getting data into Power BI

Theory
  • Power BI Desktop and the report, data and model views
  • Importing from Excel, CSV and a database
  • Transform steps in Power Query
  • Refreshing and keeping sources documented
Lab
Import the cleaned sales, customers and products tables into Power BI, apply the same cleaning steps and document the source and refresh plan.
Homework
Compare the Power Query steps with your Excel cleaning log.
Deliverable
A Power BI file with three clean tables and a source note.
Weekly checkpoint
Identify which step to change to fix a wrong data type.
You will be able to
Clean messy data: fix types, remove duplicates, handle blanks and standardise text, while keeping a record of every change. Build a tidy data model in Power BI with a star schema and write basic DAX measures.

Week 11: Data model and star schema

Theory
  • Fact and dimension tables
  • Relationships, cardinality and filter direction
  • A date table
  • Avoiding ambiguous relationships
Lab
Build a star schema with Sales as the fact table and Customer, Product and Date as dimensions; test that slicers filter correctly.
Homework
Explain why a flat sheet with repeated customer names is harder to maintain.
Deliverable
A model diagram and a test log of three filter checks.
Weekly checkpoint
Label a model diagram as fact and dimension; fix a wrong relationship.
You will be able to
Build a tidy data model in Power BI with a star schema and write basic DAX measures.

Week 12: Measures with DAX

Theory
  • Calculated columns versus measures
  • SUM, COUNT, DIVIDE, CALCULATE, FILTER
  • Time intelligence: year to date and same period last year
  • Checking measures against SQL
Lab
Create Total Sales, Total Orders, Average Order Value, Sales YTD and Sales Last Year measures, and verify each with a SQL or pivot result.
Homework
Write what each measure means in one sentence a manager would understand.
Deliverable
A measures table with definitions and test results.
Weekly checkpoint
Read three DAX measures and say what each returns; find the context error in one.
You will be able to
Build a tidy data model in Power BI with a star schema and write basic DAX measures. Interpret results correctly: separate trend from noise, avoid mistaking correlation for cause, and state what the data cannot show.
Dashboards and interpretation Weeks 13–14

Week 13: Visual design and dashboards

Theory
  • Choosing charts: comparison, trend, composition, relationship
  • Truncated axes, 3D effects and colour traps
  • Layout, hierarchy, titles and annotation
  • Accessibility: contrast, labels, colour-blind safe palettes
Lab
Build a one-page sales dashboard with KPIs, a trend line, a ranked bar chart and a map or table; then redesign a given misleading chart honestly.
Homework
Get feedback from two people on what they understood in 10 seconds.
Deliverable
A one-page dashboard and a before and after of the misleading chart.
Weekly checkpoint
Critique a dashboard screenshot using a checklist.
You will be able to
Choose a chart that fits the question, and design dashboards that are readable, honest and not misleading.

Week 14: Interpreting data responsibly

Theory
  • Trend, seasonality and random variation
  • Correlation is not causation
  • Small samples and averages that hide groups
  • Writing a finding: claim, evidence, limit
Lab
Analyse a branch sales dataset: examine whether a discount week caused higher sales, compare with a control branch and note confounders.
Homework
Write a 150-word insight with one caution.
Deliverable
A one-page insight note with a supporting chart and stated limits.
Weekly checkpoint
Interpret a chart where an average hides two very different groups.
You will be able to
Choose a chart that fits the question, and design dashboards that are readable, honest and not misleading. Interpret results correctly: separate trend from noise, avoid mistaking correlation for cause, and state what the data cannot show.
Ethics and capstone Weeks 15–16

Week 15: Privacy, security and sharing

Theory
  • Personal data: what to mask or drop
  • Row-level security in Power BI
  • Publishing, sharing and workspace roles
  • Documenting data sources and versions
Lab
Mask phone numbers in the customer table, create row-level security so each branch manager sees only their branch, and test with View as roles.
Homework
Write a one-page data-handling note for a small business.
Deliverable
A secured report and a data-handling note.
Weekly checkpoint
Decide which fields to drop, mask or keep for a given report purpose.
You will be able to
Respect data privacy and ethics: mask personal data, apply role-based access and document sources.

Week 16: Capstone: business reporting project

Theory
  • Scoping a business question
  • Pipeline documentation: source, cleaning, model, measures, visuals
  • Presenting insights in five minutes
Lab
Deliver a complete project from a provided dataset: clean, model, query, build the dashboard and present findings.
Homework
Finish documentation and practise the presentation.
Deliverable
A Power BI file, SQL file, cleaning log, one-page insight and a five-minute demo.
Weekly checkpoint
Capstone review with rubric.
You will be able to
Organise and analyse tabular data in Excel using tables, formulas, lookups and pivot tables. Clean messy data: fix types, remove duplicates, handle blanks and standardise text, while keeping a record of every change. Query a relational database with SQL using filters, joins, grouping and simple subqueries. Build a tidy data model in Power BI with a star schema and write basic DAX measures. Choose a chart that fits the question, and design dashboards that are readable, honest and not misleading. Interpret results correctly: separate trend from noise, avoid mistaking correlation for cause, and state what the data cannot show. Respect data privacy and ethics: mask personal data, apply role-based access and document sources. Deliver a business report with a dashboard, a short written insight and a documented data pipeline.

Sample lessons

Week 8

Joining tables without changing the answer

Real data is split across tables to avoid repetition. A customer table holds each customer once and an orders table holds each order once, linked by a key. A JOIN combines them. The important skill is to know how many rows to expect before and after a join. If the row count changes unexpectedly, the join is probably wrong, and totals will be wrong too.

Join typeKeepsTypical useRisk
INNER JOINOnly matching rowsOrders with customer detailsDrops unmatched rows silently
LEFT JOINAll left rows, matches if anyCustomers with or without ordersNULLs in the right columns
Duplicate keysRepeated matchesNever intendedTotals multiply
Always compare row counts before and after a join.

Step by step

  1. Identify the key in each table, for example customer_id.
  2. Count rows in each table before joining.
  3. Write an INNER JOIN on the key.
  4. Count rows after the join and compare with what you predicted.
  5. Use a LEFT JOIN when you must keep rows with no match, such as customers with no orders.
  6. Check that a total such as revenue still equals the total from the single table.

In real life

A shop owner wants to know which customers have not ordered this year. An inner join drops them, because they have no orders. A left join from customers to orders keeps them and shows the gap.

Worked example

Problem. Orders has 1,000 rows. Customers has 200 rows. You join orders to a customers table that accidentally lists each customer twice. How many rows do you get and why does revenue become wrong?

Solution. Each order matches two customer rows, so the result has 2,000 rows and revenue is doubled. Check for duplicate keys with SELECT customer_id, COUNT(*) FROM customers GROUP BY customer_id HAVING COUNT(*) > 1, remove the duplicates, and the join returns 1,000 rows again.

Common mistakes

  • Joining on a column that is not unique and not noticing the inflated total.
  • Using INNER JOIN and silently losing the customers with no orders.
  • Forgetting the ON condition and producing every possible pair of rows.

Try it yourself

  • Write a query listing customers with no orders using a LEFT JOIN and IS NULL.
  • Count the rows of a join before and after and explain any change.
  • Find the duplicate key in a provided products table.

Week 13

Choosing a chart and not misleading the reader

A chart answers a question. If the question is a comparison, use bars. If it is a trend over time, use a line. If it is parts of a whole with few parts, a bar or a simple stacked bar is usually clearer than a pie. Bars must start at zero because the length is the message. Dashboards should lead with the key numbers, use few colours and always label what is shown, for which period and in which unit.

Clear dashboard

  • One question per chart
  • Bars start at zero
  • Units and period labelled
  • One highlight colour

Misleading dashboard

  • Cut-off axis
  • 3D pies
  • No units or dates
  • Ten colours

Step by step

  1. Write the question the chart must answer in one sentence.
  2. Pick the chart type: bar for comparison, line for time, scatter for relationship.
  3. Sort bars from highest to lowest unless there is a natural order.
  4. Start bar axes at zero and label units, for example Rs lakh.
  5. Use one highlight colour and grey for the rest.
  6. Test by asking a viewer what the chart says in ten seconds.

In real life

A shop owner sees one chart where the vertical axis starts at 95 and sales appear to triple after a discount; the real change is from 96 to 100. Starting the axis at zero shows a small, honest increase.

Worked example

Problem. Branch A sold Rs 96 lakh and Branch B sold Rs 100 lakh. A bar chart has its axis from 95 to 100. What does the viewer think, and how do you fix it?

Solution. The bar for A looks about one fifth the height of B, suggesting B is five times larger. The real difference is about 4 percent. Start the axis at 0, label the values, and the bars look nearly equal, which is the truth.

Common mistakes

  • Truncating a bar chart axis so small differences look large.
  • Using a pie chart with eight slices that cannot be compared.
  • Using red and green only, which colour-blind people cannot separate.

Try it yourself

  • Redraw a chart with a truncated axis honestly and write what changed.
  • Choose the chart type for five questions and justify each.
  • Build a KPI card and a ranked bar chart for one dataset and test it with a friend.

Try the questions

The same skills are asked at three levels: School (Classes 6–12), College and Professional. Switch the level and the language (English or Tamil), and say how sure you are before you check: knowing what you are unsure about is part of learning.

Classes 6–12: everyday examples, no prior knowledge. Practice only. Nothing here is graded or saved.

Question 1 / 4 · Choose a chart for a question

A class wants to show how the temperature changed each day over a week. Which chart is best?

A class wants to show how the temperature changed each day over a week. Which chart is best?
How sure are you?

Projects

Regional Retail Sales Dashboard

A retail chain with five branches in southern Tamil Nadu has a year of sales in messy Excel files and a small product table. The owner wants to know which branches and categories are growing, which products to reorder, and whether last quarter's discount worked. Build the data pipeline and a one-page dashboard with a short written insight.

Tasks

  1. Clean the raw files with Power Query and write a log of every rule used.
  2. Load the tables into SQLite or Power BI and build a star schema with a date table.
  3. Write at least five SQL or DAX measures and verify two against an independent calculation.
  4. Design a one-page dashboard with KPIs, a trend, a ranking and filters.
  5. Write 150 words of findings with one limitation and one recommended next check.

Sample input

Orders: order_id, date, branch, product_id, qty, unit_price, discount; Products: product_id, name, category; some dates typed as text and some branch names misspelt.

Expected output

A dashboard where the total sales KPI equals the SQL total, branch names are standardised, and the insight cites at least two numbers from the data.

How it is assessed

CriterionWeight
Cleaning log and data quality20%
Model and measures verified25%
Dashboard clarity and honesty25%
Insight and stated limits20%
Privacy and documentation10%

Stretch challenge: Add a simple forecast for next month using a moving average and explain why it may be wrong.

How you are assessed

Each week: five short questions mixing data interpretation, a query or formula read, and one chart-reading question, plus a lab file checked against a short checklist and a one-line insight.

Practical work. Learners submit their own cleaned dataset, cleaning log, SQL file and Power BI file. Results are tested by comparing totals across Excel, SQL and Power BI and by reviewing how the dashboard communicates.

What trainers look forWeight
Correct, verified results30%
Data cleaning and documentation20%
Chart choice and dashboard design20%
Interpretation and stated limits20%
Privacy and ethical handling10%

Where this can lead

  • Prepares you for junior data analyst, reporting assistant and MIS-style reporting roles that rely on Excel, SQL and dashboards.
  • Prepares you to support sales, stock and finance teams with clear, checked reports.
  • Builds the base for further study in statistics, Python and data science.

These describe skills you will build. ADIVAS does not guarantee jobs, placement or salary.