Every organisation collects data, but only some turn it into better decisions. Data analytics is the discipline that closes that gap. It is also one of the most accessible entry points into the data world, because the core tools, spreadsheets, SQL and a dashboard tool, can be learned step by step and practised on real datasets. This guide covers the background, the different types of analytics, a stage-by-stage roadmap with syllabus, ethics, projects and an honest outlook. Sources are linked at the end, and anything that is our own estimate or opinion is labelled.
What is data analytics?
Google's data analytics certificate describes the work this way: data analysts collect, transform and organise data to help make informed business decisions. In practice an analyst asks a clear question, gets and cleans the data, analyses it, and presents a finding in a form that decision-makers can use.
A widely used way to describe the maturity of analytics has four stages, often attributed to Gartner's analytics maturity model:
| Type | Question it answers | Example |
|---|---|---|
| Descriptive | What happened? | Monthly sales by region |
| Diagnostic | Why did it happen? | Sales fell because a product was out of stock in two cities |
| Predictive | What is likely to happen? | Forecasting next quarter's demand |
| Prescriptive | What should we do? | Recommending stock levels per store to meet forecast demand |
Most analysts spend the bulk of their time on the first two, which is where good questions, clean data and clear communication create most of the value.
Roles around analytics
| Role | Focus |
|---|---|
| Data analyst | Collects, transforms and organises data to support business decisions; works with spreadsheets, SQL and visualisation tools |
| Business intelligence analyst | Builds and maintains dashboards and reports for ongoing monitoring; Microsoft describes a Power BI data analyst as delivering actionable insights by working with available data and domain expertise, collaborating closely with business stakeholders |
| Analytics engineer | dbt Labs describes analytics engineers as providing clean data sets to end users, transforming, testing, deploying and documenting data, bridging the gap between data engineering and analysis |
| Data engineer | Builds the infrastructure that moves data into the warehouse |
| Data scientist | Statistical modelling and machine learning, the next step for many analysts |
A short history
| When | Milestone |
|---|---|
| 1958 | IBM researcher Hans Peter Luhn uses the term "business intelligence" in an article titled "A Business Intelligence System". |
| 1974 | Donald Chamberlin and Raymond Boyce at IBM describe SEQUEL, the language that became SQL, for IBM's System R. SQL became an ANSI standard in 1986 and an ISO standard in 1987. |
| 1979 | VisiCalc, the first widely used spreadsheet, is released for the Apple II. |
| 1985 | Microsoft Excel is released for the Macintosh. The Windows version followed in 1987. |
| Late 1980s to 1990s | Data warehousing takes shape. Bill Inmon's Building the Data Warehouse (1992) and Ralph Kimball's The Data Warehouse Toolkit (1996) set out two influential design approaches: normalised and dimensional (star schema). |
| 1999 | The first version of CRISP-DM, a widely used process for data mining projects, is presented. Its six phases are business understanding, data understanding, data preparation, modelling, evaluation and deployment. |
| 2003 | Tableau is founded, growing out of a Stanford research project on visual analysis. |
| 2015 | Microsoft Power BI is released to the general public on 24 July 2015, building on Power Query, Power Pivot and Power View. |
The learning roadmap: seven stages
The plan assumes about ten hours a week. The durations are our own planning estimates, not guarantees. For reference, Google states that many learners complete its Data Analytics certificate in three to six months, and the certificate has nine courses totalling roughly 150 hours of stated course time, while IBM's Data Analyst Professional Certificate lists eleven courses.
| Stage | Focus | Estimated time |
|---|---|---|
| 0 | Foundations: thinking with data, asking good questions | 2 to 3 weeks |
| 1 | Spreadsheets for analysis | 3 to 4 weeks |
| 2 | SQL and databases | 5 to 6 weeks |
| 3 | Statistics for analysts | 4 to 5 weeks |
| 4 | Data visualisation and dashboards (Power BI or Tableau) | 5 to 6 weeks |
| 5 | Python for analysis | 5 to 6 weeks |
| 6 | Capstone, storytelling and job preparation | 4 to 5 weeks |
That is roughly seven to nine months at a steady pace.
Syllabus: what to study at each stage
Stage 0: Foundations
- The analysis process: ask, prepare, process, analyse, share and act
- Turning a vague business request into a precise, answerable question
- Types of data: structured and unstructured, quantitative and qualitative
- Metrics and KPIs: how to define a metric so everyone measures the same thing
- Data ethics, fairness and privacy basics
Stage 1: Spreadsheets for analysis
- Formulas and functions, lookups, conditional logic
- Cleaning data: removing duplicates, fixing formats, handling blanks
- Pivot tables and charts
- Basic dashboards and good chart design
Stage 2: SQL and databases
- Relational databases: tables, keys and relationships
- SELECT, WHERE, ORDER BY, GROUP BY and aggregate functions
- Joins across multiple tables
- Subqueries, common table expressions and window functions
- Cleaning and transforming data in SQL
- Data modelling concepts such as fact and dimension tables and star schemas
Stage 3: Statistics for analysts
- Descriptive statistics: mean, median, spread, distributions
- Correlation, and why correlation is not causation
- Sampling, confidence intervals and hypothesis testing
- A/B testing: design, sample size and common pitfalls
- Spotting outliers and anomalies
- Basic forecasting and trend analysis
Stage 4: Visualisation and dashboards
- Choosing the right chart for the question
- Design principles: clarity, colour, labelling, accessibility
- Power BI or Tableau: connecting to data, cleaning and shaping, building relationships, measures and calculations
- For Power BI, the topics in Microsoft's PL-300 certification: preparing the data, modelling the data, visualising and analysing the data, and managing and securing Power BI, each weighted at roughly a quarter of the exam, with security and management somewhat less
- Interactive features: filters, drill-through, tooltips and bookmarks
- Access control, refresh schedules and sharing dashboards safely
Stage 5: Python for analysis
- Python basics, Jupyter notebooks
- pandas and NumPy for cleaning, reshaping and summarising
- Visualisation with Matplotlib, Seaborn or Plotly
- Reading data from files, databases and APIs
- Automating a repeated report
Stage 6: Capstone, storytelling and job preparation
- A full case study: question, data, cleaning decisions, analysis, visuals and recommendation
- Data storytelling: lead with the finding, then the evidence
- Presenting to non-technical stakeholders and handling questions
- Portfolio, resume and interview practice, including SQL and spreadsheet exercises
The standard toolkit
- Spreadsheets: Excel or Google Sheets
- Databases and SQL: PostgreSQL, MySQL, SQL Server or a cloud data warehouse
- Visualisation and BI: Power BI, Tableau, Looker Studio
- Programming: Python with pandas, NumPy, Matplotlib; R is common in statistics
- Collaboration: Git and GitHub, documentation and version control for queries and notebooks
Ethics and privacy
- India: the Digital Personal Data Protection Act, 2023 received the President's assent on 11 August 2023 and is being brought into force in phases. It centres on consent, gives people rights over their data, and sets duties for organisations that process it. Read the official text and take advice before handling personal data.
- GDPR: if you analyse data about people in the European Union, the General Data Protection Regulation applies. It has applied since 2018 and sets principles including lawfulness, purpose limitation, data minimisation, accuracy, storage limitation, integrity and confidentiality, and accountability.
- Good practice: collect only the data you need, protect it, anonymise where you can, document your assumptions and never present a chart that misleads.
Portfolio projects
- Sales performance dashboard: load a public retail dataset, clean it, model it and build an interactive dashboard that answers three business questions.
- SQL case study: analyse customer behaviour in a database with joins and window functions, and publish the queries with explanations.
- A/B test analysis: analyse a sample experiment, state the hypothesis, check the sample size and explain the decision.
- Public data story: use open data about your city or college, and write a one-page recommendation for a decision-maker.
- Automated report: use Python to refresh a monthly report from raw files and export a clean summary.
For each project, state the question, the data source, the cleaning steps, the analysis and what you would recommend. Explaining limitations honestly is a strength, not a weakness.
Common mistakes
- Starting with tools before understanding the question
- Skipping data cleaning and trusting dirty data
- Confusing correlation with causation
- Overloading dashboards with charts nobody uses
- Reporting numbers without context or comparison
- Ignoring who will read the result and what decision they must make
Careers and outlook
The US Bureau of Labor Statistics does not have a single "data analyst" occupation, so these are related occupations. They are US figures and should not be read as Indian salaries.
- Data scientists: median pay of $120,230 a year (May 2025), with employment projected to grow 35 percent from 2025 to 2035.
- Market research analysts: median pay of $78,760 a year, with 7 percent projected growth over the same period.
- Operations research analysts: median pay of $88,940 a year, with 12 percent projected growth.
Pay and demand vary by location, industry and experience. No course, including ours, can promise a job or a salary.
Where the field is heading: our view
These are opinions, not forecasts.
- AI assistants speed up routine work. Writing queries, cleaning data and drafting charts get faster, which raises the value of framing the question and checking the answer.
- Self-service analytics grows. More business users explore data themselves, so analysts spend more time on data quality, modelling and governance.
- Analytics and data engineering overlap. Skills in version control, testing and well-modelled data are increasingly expected.
- Trust and privacy matter more. Regulation and public expectations push teams to document sources and protect people's data.
- The core skills stay constant. SQL, statistics, sound judgement and clear communication will outlast any tool.
Start this week
- Pick a small dataset about something you care about, such as cricket, movies, a local bus route or a household budget.
- Write three questions you would like it to answer.
- Answer one in a spreadsheet with a pivot table, then answer another with a SQL query.
- Make one clear chart and write two sentences on what it shows and what it does not.
- Follow a structured path with feedback and projects so you build skills in a sensible order.
If you would like guided, project-based learning, see the courses at Nexelence Academy or talk to us about where to begin.
Sources
- Google, Data Analytics Certificate and Coursera listing; IBM, Data Analyst Professional Certificate
- Microsoft Learn, Power BI Data Analyst Associate and the PL-300 study guide; dbt Labs, What is analytics engineering?
- History: Wikipedia articles on business intelligence, SQL, VisiCalc, Microsoft Excel, data warehouse, CRISP-DM, Tableau and Power BI (secondary sources; check primary papers where dates matter)
- The four types of analytics are described in many secondary sources, for example Qlik, attributing the framework to Gartner
- Wikipedia, DPDP Act 2023 (check the official text and dates); gdpr.eu, What is GDPR? (an informational site, not an official EU source)
- US Bureau of Labor Statistics, Data Scientists, Market Research Analysts and Operations Research Analysts
- #data analytics
- #SQL
- #Power BI
- #Excel
- #statistics
- #dashboards
- #roadmap
- #syllabus