8 SQL Techniques to Perform Data Analysis for Analytics and Data Science


SQL is an essential cog in an information science professionals armory. If you havent yet chosen up SQL, Im speaking from experience– you merely can not anticipate to sculpt out a successful career in either information or analytics science.

SQL is a must-know language for anyone in data or analytics science
Here are 8 cool SQL strategies for information analysis that ever information and analytics science expert will like working with



And why is SQL so crucial?

As we move into a brand-new years, the rate at which we are producing and consuming data is skyrocketing day by day. To make wise decisions based upon data, organizations all over the world are working with information professionals like service experts and information researchers to mine and discover insights from the huge gold mine of information.

And among the most important tools needed for this is– you thought it– SQL!

In this post, I will be going over 8 SQL techniques/queries that will make you prepared for any innovative information analysis problems. Do keep in mind that this post presumes a really standard knowledge of SQL.

It is a programming language used for managing the information held in relational databases. A data expert can use SQL to gain access to, check out, control, and analyze the information kept in a database and produce useful insights to drive an informed decision-making procedure.

I would recommend taking a look at the below courses if youre new to SQL and/or organisation analytics:

Keep in mind:- I will be utilizing MySQL 5.7 for moving forward in the short article. You can download it from here– My SQL 5.7 Downloads

We will start our analysis with the most basic query, i.e, counting the number of rows in our table. We will do this by using the function– COUNT().

Lets First Understand the Dataset.

Fantastic! Now we understand the variety of rows in our table which is 10. It might appear to be amusing utilizing this function on a little test dataset however it can assist a lot when your rows encounter the millions!

A great deal of times, our data table is filled with duplicate worths. To achieve the unique worth, we use the DISTINCT function.

SQL Technique # 1– Counting Items and rows.



What is the best method to discover data analysis? By performing it side by side on a dataset! For this purpose, I have actually produced a dummy dataset of a retailer. The client information table is represented by ConsumerDetails.


Total_amt_spend– The overall amount of money invested by the customer in the shop.

Our dataset includes the following columns:.

Table of Contents.

Industry– It represents the market from which the consumer comes from.

Lets First Understand the Dataset.
SQL Technique # 1: Counting Rows and Items.
SQL Technique # 2: Aggregation Functions.
SQL Technique # 3: Extreme Value Identification.
SQL Technique # 4: Slicing Data.
SQL Technique # 5: Limiting Data.
SQL Technique # 6: Sorting Data.
SQL Technique # 7: Filtering Patterns.
SQL Technique # 8: Groupings, Rolling up Data and Filtering in Groups

Area– The locality of the consumer.

Call– The name of the customer.

In our dataset, how can we discover the special industries that consumers come from?

You guessed it right. We can do this by utilizing the DISTINCT function.

You can even count the number of distinct rows by utilizing the count along with distinct. You can describe the below inquiry:

SQL Technique # 4– Slicing Data.

SQL Technique # 2– Aggregation Functions.

We require to discover the consumers who live in specific regions (Shakti Nagar and Shanti Vihar) and invest an amount higher than Rs. 2000.

In our dataset, just Shantanu and Natasha fulfill these conditions. As both conditions need to be satisfied, the AND condition is much better suited here. Lets inspect out another example to slice our data.

Lets learn the sum of the amount spent by each of the consumers:.

The basic discrepancy comes out to be 829.7 which implies there is a high variation between the expenditures of customers!

Aggregation functions are the base of any type of data analysis. They offer us with an introduction of the dataset. A few of the functions we will be going over are– SUM(), AVG(), and STDDEV().

The next type of analysis is to recognize the extreme values which will assist you understand the data better.

Lets state that the store wishes to discover the customers originating from a locality, particularly Shakti Nagar and Shanti Vihar. What will be the inquiry for this?

Calculate basic deviation.

To calculate the average of the numerical columns, we use the AVG() function. Lets find the average expense by the consumers for our store:.

In the above example, sum_all is the variable in which the worth of the sum is kept. The amount of the amount of money invested by consumers is Rs. 12,560.


The maximum quantity of cash invested by the consumer in the store is Rs. 3000.

The minimum amount of money invested by the retail store consumer is Rs. 350

The optimum numerical worth can be identified by utilizing limit() function. Lets see how to apply it:.


Another method to compose the exact same statement would be:.

The average quantity spent by customers in the retailer is Rs. 1256.

Comparable to the max function, we have the MIN() function to identify the minimum numerical worth in a given column:.

This time the retail shop wants to retrieve all the consumers who are spending between Rs. What will be the query for this?

Youll have noticed theres something missing if you have looked at the dataset and then the typical worth of expense by the customers. The average does not rather offer the complete image so lets find another crucial metric– Standard Deviation. The function is STDDEV().

SQL Technique # 3– Extreme Value Identification.

We use the SUM() function to determine the sum of the numerical column in a table.


Great, we have 3 consumers! We have actually utilized the WHERE provision to filter out the information based on the condition that customers need to be living in the locality– Shakti Nagar and Shanti Vihar. I didnt utilize the OR condition here. Rather, I have used the IN operator which allows us to specify numerous values in the WHERE provision.

Now, let us focus on among the most fundamental parts of the data analysis– slicing the information. This area of the analysis is going to form the basis for sophisticated queries and help you recover information based upon some kind of condition.

Just Rohan is clearing this requirements!

Lets say we want to see the information table consisting of millions of records. We cant use the SELECT statement directly as this would dump the total table onto our screen which is cumbersome and computationally extensive. Instead, we can use the LIMIT clause:.

The above SQL command helps us show the very first 5 rows of the table.

SQL Technique # 5– Limiting Data.


Great! We have actually reached midway in our journey. Let us construct more on the knowledge that we have actually gained up until now

What will you do if you simply wish to choose only the fifth and 4th rows? We will make usage of the OFFSET provision. The OFFSET provision will skip the specified variety of rows. Lets see how it works:

Let us see an example where we sort the data according to the column Total_amt_spend in ascending order:.

SQL Technique # 6– Sorting Data.


The keyword can be used to sort the information into rising or descending order. The ORDER BY keyword sorts the data in ascending order by default.

Arranging information assists us put our data into point of view. We can perform the arranging process by utilizing the keyword– ORDER BY.

Amazing! To order the dataset into coming down order, we can follow the below command:

The LIKE operator is used in a WHERE provision to look for a defined pattern in a column.

Incredible, we have 6 regions ending with this name. Notification that we are utilizing the LIKE operator to carry out pattern matching.

%– It represents 0 or more variety of characters.
_– It represents a single character.


In the earlier sections, we discovered how to filter the information based on one or multiple conditions. Here, we will discover to filter the columns that match a defined pattern. To move forward with this, we will first understand the LIKE operator and wildcard characters.

Next, we will attempt to fix another pattern-based issue. We desire the names of the customers whose second character has “a” in their respective names. Again, I would recommend you to take a minute to think and comprehend the issue of a reasoning to fix it.

The Wildcard Character is used to substitute one or more characters in a string. These are used along with the LIKE operator. The two most common wildcard characters are:.

SQL Technique # 7– Filtering Patterns.

Lets breakdown the problem. After the 2nd character, there can be any number of characters so we replace those characters with the wildcard “%”.

Lets try to break down the issue. We need all the regions that end with “Nagar” and can have any variety of characters before this specific string. For that reason, we can make use of the “%” wildcard prior to “Nagar”:.

In our dummy retail dataset, lets state we want all the areas that end with “Nagar”. Take a minute to understand the problem statement and think of how we can fix this.

We have 6 individuals satisfying this bizarre condition!

We see that the count of consumers belonging to the numerous industries is more or less the exact same. So, let us move forward and find the amount of costs by clients grouped by the market they belong to:.

Now, the retailer desires to discover the industries whose total_sum is higher than 2500. To resolve this issue, we will again group by the information according to the industry and after that use the HAVING clause.

We can observe that the maximum amount of cash invested is by the clients coming from the Manufacturing industry. This appears a bit simple, ideal? Let us take a step ahead and make it more complicated.

The HAVING clause is much like the WHERE clause however just for filtering the grouped by data. Keep in mind, it will constantly come after the GROUP BY declaration.


We have actually finally gotten to one of the most powerful analysis tools in SQL– Grouping of information which is performed utilizing the GROUP BY statement. The most helpful application of this declaration is to discover the circulation of categorical variables. This is done by utilizing the GROUP BY declaration in addition to aggregation functions like– COUNT, SUM, AVG, etc

. Lets attempt to comprehend this better by taking up an issue declaration. The store wishes to discover the Number of Customers representing the industries they belong to:.

SQL Technique # 8– Groupings, Rolling up Data and Filtering in Groups.

We have only 3 categories that satisfy the conditions– Aviation, Defense, and Manufacturing. But to make it more clearer, I will also include the ORDER BY keyword to make it more intuitive:

End Notes.

An information expert can use SQL to gain access to, read, control, and examine the data stored in a database and generate useful insights to drive an informed decision-making procedure.

Related Articles.

I really hope that these SQL questions will help you in your day to day life when you are evaluating intricate data. Do you have any of your tips and techniques for evaluating data in SQL? Let me know in the comments!

We have lastly arrived at one of the most effective analysis tools in SQL– Grouping of information which is performed utilizing the GROUP BY statement. These are the building obstructs for all information analysis questions in SQL. I truly hope that these SQL queries will assist you in your day to day life when you are examining complex data. Do you have any of your tips and tricks for analyzing information in SQL?


I am truly glad you made it up until now. These are the building obstructs for all data analysis queries in SQL. You can likewise use up advanced questions by using these basics. In this post, I used MySQL 5.7 to develop the examples.

You can also read this short article on our Mobile APP.


15 gadgets that will sell out in 2020