Instagram User Analytics

Creation of tables provided as attachments

Primary key
Here we have 7 tables interrelated with one another.
1. The marketing team wants to reward the most loyal users, i.e., those who have been using the
platform for the longest time.
Task: Identify the five oldest users on Instagram from the provided database.

Here we arranged the table in

ascending order of created_ at so
that we can get oldest users and limit
it to 5 to get only get top 5.

2. The team wants to encourage inactive users to start posting by sending them promotional
emails.
Your Task: Identify users who have never posted a single photo on Instagram.

We will first get all active users by

getting id of those who have posted
photos and then we will run query for
those ids which are not in active
ones.

3. The team has organized a contest where the user with the most likes on a single photo win.
Your Task: Determine the winner of the contest and provide their details to the team.
 Here I created a temporary table to count likes then joined it twice with photos and users table
with same photo id as in like to the same in photos and then userid corresponding to it in
photos to id in users

4. A partner brand wants to know the most popular hashtags to use in their posts to reach the
most people.
Your Task: Identify and suggest the top five most commonly used hashtags on the platform.

We created a temp table to get

count of tag used with their id of
top 5 by using limit and joined it
with tags table to show details of
the tag.

5. The team wants to know the best day of the week to launch ads.
Your Task: Determine the day of the week when most users register on Instagram. Provide
insights on when to schedule an ad campaign.
 Here I used Day of week function to count day no starting from 1 as Sunday and printed in
descending order after grouping and then limited it to 1 entry only.
6. Investors want to know if users are still active and posting on Instagram or if they are making
fewer posts.
Your Task: Calculate the average number of posts per user on Instagram. Also, provide the total
number of photos on Instagram divided by the total number of users.
7. Investors want to know if the platform is crowded with fake and dummy accounts.
Your Task: Identify users (potential bots) who have liked every single photo on the site, as this is
not typically possible for a normal user.
 For this we created a temporary table
which will get user id and their
counts of liking photo result of which
we will use to join only with those ids
which have their count = to count of
photos so that we will get those who
have liked all and then just used join
to get details.

I used MySQL workbench as introduced to us in course due to its free availability and easy to work
interface.

Project Overview
Overall this project is made to make us familiar and comfortable with SQL queries and real life problems as
data analyst we may need to deal. In this project we got glimpse of how real life problems are and how to
deal with them and it was a full brush up of everything that was introduced in course.

INSIGHTS
1. Loyal Users: Identified the five oldest users to target for loyalty rewards, fostering community commitment.

2. Inactive Users: Recognized users with no posts as a potential audience for tailored re-engagement marketing campaigns.

3. Contest Winner: Highlighted a user with the most likes on a single photo, showcasing high engagement and serving as
motivation for others.

4. Hashtag Usage: Discovered the top five popular hashtags to enhance brand visibility and engagement in marketing e orts.

5. Ad Campaign Timing: Pinpointed the best day for user registrations to optimize ad launch strategies for maximum exposure.

6. User Engagement Metrics: Calculated average posts per user to assess engagement trends, aiding in strategic decisions to
boost activity.

7. Fake Account Detection: Identified potential bot accounts, emphasizing the need for integrity and trust on the platform to
reassure investors.
RESULTS
Through this project we were able to apply whatever we have learnt in the lectures. This project is far much
resourceful than sources available online as before beginning project practising online had not helped
enough to make feel confident of the joins and other little hard parts but after working on this project now it
seems much easy to work on SQL and deal with problems.