Introduction
Data is the foundation of every data analytics project. Businesses collect information about sales, customers, products, website visitors, payments, and daily operations. However, collecting information is only the beginning. A data analyst must understand what type of data they have, where it came from, and whether it is suitable for analysis.
For example, a company may have thousands of sales records. These records can help answer important questions: Which product sells the most? Which month generates the highest revenue? Which customers purchase repeatedly?
To answer these questions correctly, an analyst needs to understand data types and data collection methods.
In this lesson, you will learn these concepts through simple examples, practical business situations, interview questions, and exercises.Understanding Data Types and Data Collection in Data Analytics
1. What is data?
Data is a collection of facts, measurements, observations, or information that can be used to understand a situation or solve a problem.
Examples include:
Sales revenue: ₹25,000
Customer name: Rahul Sharma
Product category: LED Driver
Order date: 10 October 2026
Customer feedback: “The product quality is good.”
Website visits: 1,500 visitors
When data is organized, checked, and analyzed, it can help businesses make better decisions.Understanding Data Types and Data Collection in Data Analytics
2. Main types of data
Data can be classified in different ways. Two particularly useful classifications for beginners are its format and its content.
A. Structured data
Structured data follows a predefined format, usually arranged in rows and columns.
Example:
Order ID | Product | Quantity | Sales |
|---|---|---|---|
101 | LED Driver | 10 | ₹5,000 |
102 | SMPS | 5 | ₹3,000 |
103 | DALI Driver | 8 | ₹12,000 |
This data can easily be organized and analyzed in Excel, SQL databases, or business intelligence tools.
Common examples: sales reports, inventory lists, customer databases, and monthly expense sheets.
B. Semi-structured data
Semi-structured data does not always follow a fixed table format, but it contains labels, keys, or other organizational information.
Examples include JSON files, XML files, and some email data.
A simplified JSON example:
{
"product": "LED Driver",
"quantity": 10,
"price": 500
}An analyst may receive this type of data from websites, software applications, or APIs.
C. Unstructured data
Unstructured data does not fit neatly into a predefined table.
Examples include:
Product photographs
Videos
Voice recordings
Customer reviews
Documents and free-text feedback
This information can still be valuable, but it may need additional processing before analysis.
3. Quantitative and qualitative data
Another useful classification is based on what the data represents.
Quantitative data — numerical information
Examples: sales amount, product quantity, delivery time, customer age, and monthly expenses.
Qualitative data — descriptions or categories
Examples: product colour, customer comments, payment method, product category, and customer satisfaction level.
Quantitative data can be further divided into:
Discrete data: Countable values, such as 15 orders or 8 products.
Continuous data: Measurable values, such as temperature, weight, or delivery time.
A number is not always a measurement. For example, an order ID such as 101 is an identifier, not a quantity you would normally add or average.
4. What is data collection?
Data collection is the systematic process of gathering information for a specific purpose. Analysts may collect data through surveys, interviews, observations, company records, databases, sensors, or software systems.Understanding Data Types and Data Collection in Data Analytics
Suppose a lighting business wants to understand why sales decreased last month. It might collect:
Monthly sales figures
Number of customer inquiries
Product prices
Order cancellations
Customer feedback
Delivery times
The objective is to gather relevant information before drawing conclusions.
5. Primary data vs secondary data
Primary data | Secondary data |
|---|---|
Collected directly for the current purpose | Previously collected by someone else or for another purpose |
Example: conducting a customer survey | Example: using a published industry report |
Can closely match your question | May not exactly match your question |
Can require more time and cost | Can be faster and less expensive to obtain |
Both types can be useful. Before using secondary data, check its source, date, definitions, completeness, and suitability for your question.Understanding Data Types and Data Collection in Data Analytics
6. Common data collection methods
Surveys and questionnaires: Collect responses from customers or users.
Interviews: Ask people detailed questions about their experiences.
Observation: Record what happens in a real situation.
Business records: Use invoices, sales reports, inventory systems, and order histories.
Website analytics: Review page views, traffic sources, and user interactions.
APIs and software systems: Retrieve data from applications and online services.
The right method depends on the business question, available resources, data quality, and ethical or privacy requirements.
7. Real-world example
Imagine that SmartTechMitra wants to understand which educational blogs attract the most visitors.
The analyst could collect:
Blog title
Publication date
Page views
Average engagement time
Traffic source
Number of clicks
The data could be organized in Excel and used to compare blog performance.
But the analyst should not immediately conclude that the blog with the most page views is the best. They may also need to examine engagement, search intent, and the purpose of each article.
8. Practice exercise
Choose a small business and write down five types of information it could collect.
For each item, identify:
Whether it is structured or unstructured
Whether it is quantitative or qualitative
Whether it is primary or secondary data in your chosen example
How the business could use it
Example: Monthly sales revenue is quantitative and usually structured. If it comes from the company's own sales records, it is internal historical data.
9. Interview questions and answers
Q1. What is data analytics?
Data analytics is the process of examining data to identify patterns, answer questions, and support decisions.
Q2. What is structured data?
Data organized according to a predefined format, often in tables with rows and columns.
Q3. What is the difference between primary and secondary data?
Primary data is collected directly for a specific purpose; secondary data was already collected and is reused.
Q4. Give two examples of unstructured data.
Videos and free-text customer reviews.
Q5. Why is data collection important?
It provides the information needed for analysis. Poor-quality or irrelevant data can lead to misleading conclusions.
10. Frequently asked questions
Can beginners learn data analytics without coding?
Yes. You can start with spreadsheets and basic analytical thinking before progressing to SQL, Python, and other tools.
Is Excel useful for data analytics?
Yes. Excel is useful for organizing, cleaning, calculating, summarizing, and visualizing structured data.
What should I learn after data types and data collection?
A useful next step is data cleaning, followed by sorting, filtering, formulas, and summary analysis.
Conclusion
Understanding data types and data collection is an essential first step in data analytics. Before calculating totals or creating charts, learn what each field represents, where the information came from, and whether it is reliable enough for your purpose.Understanding Data Types and Data Collection in Data Analytics
In the next lesson, you will apply these fundamentals by cleaning and organizing a dataset in Excel.
Blog 2: Data Analytics Practical Day 2 — Clean and Organize Data in Excel
For SmartTechMitra · Practical / Hands-on · Beginner Level
SEO settings
SEO title: Data Cleaning in Excel for Beginners: Day 2 Practical Guide
Focus keyword: Data Cleaning in Excel
URL slug:
data-cleaning-in-excel-practical-day-2Meta description: Learn data cleaning in Excel step by step. Remove duplicates, fix missing values, correct formatting, and prepare a dataset for data analysis.
Category: Data Analytics Course
Tags: Excel Data Cleaning, Data Analytics Practical, Remove Duplicates, Excel Tutorial, Data Preparation
Introduction
Before a data analyst creates charts, calculates sales totals, or builds a dashboard, the data must be organized and checked.
Real-world data is rarely perfect. A sales spreadsheet may contain duplicate orders, missing customer names, inconsistent product categories, extra spaces, or numbers stored as text. If these problems are ignored, the results of an analysis may be misleading.
In this practical lesson, you will learn how to clean a small sales dataset using Microsoft Excel. You do not need programming knowledge to complete the exercise.
1. What is data cleaning?
Data cleaning is the process of identifying and correcting data-quality problems so that a dataset is more consistent and suitable for analysis.
Common problems include:
Duplicate records
Missing values
Inconsistent spelling
Incorrect date formats
Extra spaces
Numbers stored as text
Invalid or impossible values
Cleaning does not mean changing data simply to make it look better. Every correction should be justified, and uncertain values should be investigated rather than guessed.Understanding Data Types and Data Collection in Data Analytics
2. Create your first practice dataset
Open Microsoft Excel and enter the following sample data into cells A1 to F8.
Order ID | Product | Quantity | Price | City | Status |
|---|---|---|---|---|---|
1001 | LED Driver | 10 | 500 | Delhi | Delivered |
1002 | SMPS | 5 | 600 | Noida | Delivered |
1003 | DALI Driver | 4 | 1500 | Ghaziabad | Pending |
1002 | SMPS | 5 | 600 | Noida | Delivered |
1004 | LED Driver | 500 | Delhi | Pending | |
1005 | dali driver | 2 | 1500 | Delhi | Delivered |
1006 | SMPS | 3 | 600 | Noida | Cancelled |
This is a deliberately imperfect practice dataset. It contains a repeated order, a missing quantity, and inconsistent capitalization in the product name.
Your objective: identify these issues and prepare a cleaner version without losing important information.
3. Step 1 — Make a backup
Before editing:
Open Excel.
Select File → Save As.
Save a copy named
Data_Analytics_Day_2_Practice.xlsx.Keep the original data on a separate worksheet named
Raw Data.Create another worksheet named
Clean Datafor your corrections.
This makes it easier to compare the original data with the cleaned version.
4. Step 2 — Format the dataset as a table
Select your dataset, including the headings.
Press Ctrl + T.
Confirm that My table has headers is selected.
Click OK.
Excel will format the dataset as a table and add filter controls to the headings.
You can now sort or filter the data more easily.
5. Step 3 — Identify duplicate records
The sample dataset contains Order ID 1002 twice with identical details.
To remove the duplicate from your working copy:
Select the dataset in the
Clean Dataworksheet.Open the Data tab.
Select Remove Duplicates.
Choose the columns you want Excel to compare.
Click OK and review the result.
For this exercise, comparing all six columns identifies the repeated row because every field matches.
Important: Do not remove records solely because the product name or customer appears more than once. A customer can place multiple legitimate orders. Confirm the business rules before deleting anything.
6. Step 4 — Handle missing values
Order 1004 has no quantity.
To investigate it:
Select the Quantity column.
Open the filter menu.
Look for blank values.
Check the original invoice, order record, or other reliable source.
Enter the correct quantity only when it is verified.
If the correct value cannot be found, keep the field blank or mark it according to your reporting rules. Do not enter 0 automatically because a blank quantity and a zero quantity mean different things.
7. Step 5 — Standardize product names
The dataset contains both DALI Driver and dali driver.
These represent the same product in this example, so you can standardize the capitalization.
For a simple dataset, use Find and Replace:
Press Ctrl + H.
Enter
dali driverin the Find field.Enter
DALI Driverin the Replace field.Choose Replace All after checking the matches.
For larger datasets, Excel's PROPER, UPPER, and LOWER functions can help standardize text. However, abbreviations such as DALI, LED, and SMPS may need special handling because automatic capitalization can change their intended format.
8. Step 6 — Check numbers and prices
Review the Quantity and Price columns.
Confirm that quantities are numeric.
Confirm that prices are numeric and use the expected currency.
Check for accidental spaces or symbols stored as text.
Look for unexpected negative values or unusually large amounts.
For this dataset, you can format the Price column as currency using Home → Number → Currency.
Formatting changes how a value appears; it does not necessarily fix a value that is stored incorrectly. Verify the underlying data when calculations fail.
9. Step 7 — Calculate sales value
Add a new column called Sales Value.
In the first data row, enter:
=[@Quantity]*[@Price]If your range is not formatted as an Excel table, a normal cell formula might look like this:
=C2*D2Copy the formula down the column.
For order 1001, the result is:
\(10 \times ₹500 = ₹5,000\)
For order 1002, the result is:
\(5 \times ₹600 = ₹3,000\)
Remember that an order with a missing quantity will not produce a reliable sales value until the quantity is resolved.
10. Step 8 — Check the final dataset
Use this checklist before analysis.
Data cleaning checklist
0 of 7
11. What can you learn from the cleaned data?
Once the dataset is verified, you can start asking useful questions:
How many orders were delivered?
Which product generated the most sales?
How many orders are pending?
What is the total sales value of valid orders?
How many orders were cancelled?
One important distinction: a cancelled order should not automatically be counted as completed revenue. Your reporting rules should define how delivered, pending, and cancelled orders are treated.
12. Practical homework
Complete these tasks in your Excel workbook:
Create the
Raw DataandClean Dataworksheets.Identify the duplicate order and explain why it is a duplicate.
Investigate the missing quantity.
Standardize the product names.
Add the Sales Value column.
Filter the data to show only delivered orders.
Write three observations about the dataset.
Extra challenge: Create a PivotTable that summarizes the number of orders by product.
13. Interview questions and answers
Q1. What is data cleaning?
Data cleaning identifies and corrects data-quality issues before analysis.
Q2. How do you remove duplicates in Excel?
Select the dataset, open Data → Remove Duplicates, choose the appropriate comparison columns, and review the result.
Q3. How should missing values be handled?
Investigate the source, correct the value if verified, or document that the value remains missing. Avoid guessing.
Q4. What is the difference between a blank and zero?
A blank means no value is recorded; zero is an actual numerical value.
Q5. Why should raw data be preserved?
Keeping an original copy allows you to audit changes, compare results, and recover from mistakes.
14. Frequently asked questions
Can I learn data cleaning in Excel without coding?
Yes. Many common tasks can be completed with Excel menus, filters, formulas, and tables.
What should I learn after data cleaning?
Learn sorting, filtering, basic formulas, conditional formatting, and PivotTables before progressing to dashboards and SQL.
Is removing duplicates always safe?
No. Some repeated-looking rows represent legitimate transactions. Identify the correct business key and verify the records first.Understanding Data Types and Data Collection in Data Analytics
Conclusion
Data cleaning is one of the most practical skills in data analytics. By learning to identify duplicates, investigate missing values, standardize text, validate numbers, and calculate sales values, you can prepare more reliable data for analysis.
The key habit is to verify first, correct carefully, and document your changes.Understanding Data Types and Data Collection in Data Analytics
0 Comments