Contains work that I have done as a candidate professional data analyst in project work using Microsoft Excel, SQL, Python, Looker Studio, Tableau, Power BI, business presentation, and summary of data analysis
What was learned?
Learn the formulas used for data analysis purposes
Create a pivot table according to the importance of the data needed
Create a simple dashboard to make it easier to explain data
Pivot Table Analysis of Sales in Several Country
Analyst Results :
The highest income is in May, while the lowest income is in February
Recommendation :
The month of May produces the highest total income, so the campaign strategies carried out this month can be tried to be applied to other months
Analyst Results :
United Kingdom is the country with the highest total sales revenue, while Greece and Netherlands is the country with the lowest total sales revenue.
Recommendation :
The population of the United Kingdom, when compared with the Netherlands and Greece, is around 1: 6 to 1: 7, but the United Kingdom's income is 193 times greater than that of these two countries. So it is recommended if you want to make price changes, or carry out a marketing strategy tailored to the destination country, because each country has a different per capita income.
Simple Analysis of Dataset Film in Malaysia
Analyst Results :
Success indicators including satisfaction level (rating score), liking level (total likes) and recommendations (total share) were not achieved
Recommendation :
Research the film genres that are most popular with audiences, before making a new film to be shown in the future
What was learned?
Solve business cases and questions using SQL queries
Practice creating basic queries, applying functions, aggregation and grouping, combining rows or columns in tables, subqueries, and CTE (Common Table Expression)
Setup dBeaver
Setup databases: SQLite and PostgreSQL
Identifying primary keys and foreign keys and reading database schemas
Know the relationships between tables such as one to many, many to many, and one to one
Knowing database terminology
📘1
📘2
Still like the previous problem, now look for people whose last name contains the word "LI", but try to sort them by last name. If there are people with the same last name, then prioritize sorting by first name.
Calculate the average basketball size for each city
📘3
Find how many sales/amount each staff member made in August 2005, use the staff and payment table
📘4
The company wants to create a marketing campaign via email in Canada. Search for all customer emails originating from Canada, use CTE if possible
📘5
Look for the 5 film genres with the highest gross revenue, then sort them from the best-selling ones first. Hint ; Genres are found in the film_category table.
📘6
Which city has the highest number of orders? HINT : Use count(distinct column_name) for count aggregation
What was learned?
Introduction to Variables in Python
Primitive data types in Python
Operations on numeric and string type values
Comparison and boolean operations, If else, for loop
Data Structure in Python: List and iteration
Data Reading, Working Directory
Read data with pandas and create connection sql and python
Data Cleansing: Remove columns that are not needed, tidy up column names, create columns with datetime type, map the values of a column, clean missing values / observations, correct unusual extreme values
Data Filtering and Subsetting : .ilco and .loc
Data Grouping & Aggregation,Data restructuring and combination, Combining data (concat and merging)
Library for data visualization
Univariate Visualization: test scores, histogram, lineplot, boxplot, Lineplot or Barplot
Bivariate & Multivariate Visualization: Scatter Plot, Heatmap,
Multiple visualization in Python : Subplots and Facet Grid
Customer Funnel Analysis: Funnel Analysis, Trend Analysis
Exploratory Data Analysis (EDA)
K-Means algorithm: standardization, analysis using k-means, analysis of cluster data, determining clusters (Elbow Method and Silhouette Score)
Conducting RFM Analysis: Recency Calculation, Volume Calculation, and Frequency
Basket Size Analysis, Retention Cohort Analysis, Time Series Analysis, and Price Elasticity
Retention Cohort Analysis
Analyst Results :
January 2019 produced the highest retention.
The growth of new customers slowed down, in January 2019 new customers returning to make transactions reached 49%. However, from February to December 2019, customer return transactions experienced a decline and growth was unstable, namely in the range of 10% -36%.
If we look at it as a whole, from January to December 2019, in the final period customers made very low transactions again and found it difficult to increase, only in the range of 8% -23%.
Recommendation :
Campaigns in January produce the highest retention, it would be a good idea in the future to try to implement a campaign model like January.
Time Series Analysis
Analyst Results :
The highest order is at 16.00, and the lowest is at 05.00
The highest order for the daily period was on the 24th, and the lowest was on the 31st
The highest orders for the weekly period are on Tuesday, and the lowest are on Saturday
The highest orders for the monthly period were in August, and the lowest were in September
Recommendation :
At the time, day or date with the highest orders, there can be more discounts on goods or product delivery, stock of goods, and maintenance of the online shop so that customers make lots of purchases and the system doesn't go down when many customers access it simultaneously.
Market Basket Size Analysis
Analyst Results :
The highest basketball market size is on the 18th, and the lowest is on the 31st
The highest basketball market size is in the United Kingdom, and the lowest is in Saudi Arabia
Recommendation :
If we look at it as a whole, from the beginning to the end of the date, there is a decrease in orders, so future evaluations need to be carried out, such as making differences in product prices and discounts at the beginning and end of the month. Because the majority of workers, at the end of the month the money is usually used for other more important things. So by making different prices, the hope is to increase product sales.
Price Elasticity
Analyst Results :
Price increases: The number of orders decreases
Price decreases: The number of orders increases
Recommendation :
It is best if you want to increase product prices, adjusting them to each country, because each country's income per person is different. Apart from that, you can think again if you want to increase the price, because the more the price rises, the average number of purchases will decrease.
Conducting RFM Analysis
Analyst Results :
The customer category based on total value price with the highest is in the cheap group, and the lowest is in the expensive group
Customer categories based on the total number of ordered items are in the little group and the lowest are in the lot group
75% of customers are in the satisfied category, and 25% are in the dissatisfied category when providing reviews
Days since last transaction customers with the highest score is 1, and the lowest is score 2
Recommendation :
On average, customers buy goods at low prices, so that in the future the stock or types of products can be increased further, so that buyers are more interested in making transactions.
Funnel Analysis
Analyst Results :
From page views to page addtocart decreased by around 97% or from 2256640.0 to 63309.0
From page addtocart to page transactions decreased by around 34% or from 63309.0 to 21639.0
The significant decrease from addtocart to transactions could be caused by visitors not being interested in the product (perhaps the total product and shipping costs are expensive, payment methods are difficult, or there is no free shipping, etc.) so they are reluctant to make transactions
Recommendation :
From the view page to the addtocart page, the loss of visitors is more than 90%, it is necessary to check whether there is anything strange on this page. Do errors often occur or the appearance and price of the product are less attractive and too expensive.
Microsoft Excel
Analyst Results :
Highest total attrition by gender: No
Count of job roles and total years with most cur managers: Life Sciences
Total current employee oh hour rate :bachelor's Degree
Most total relational satisfaction: Married
Tableau
Analyst Results :
Most wise attrition department: R&D
The highest number of ratings based on job satisfaction rate is number 4
The highest employee by age is 32 years old, and the average age of all employees is 37 years old
Education field wise highest attrition: Life Sciences
Power BI
Analyst Results :
Sales order ma'am, highest customer region: West
Category by highest shipment: Office supplier
Sum of days for shipment scheduled by customer state : California
Highest day shipment per city: New York
Quantity orders per customer segment: Consumer
Looker Studio
Analyst Results :
Most total sales of goods and income: United Kingdom
Total items sold based on the highest advertising sources: Paid Ads and Instagram
Highest selling product code: 84755
CustomerID with the most items purchased: 15502
Highest total income: May
What is being done?
Presenting data that has been analyzed as needed, using Power Point to be presented to relevant stakeholders, including
Introduction dataset
Objective
Data explanation
Data analysis results
Conclusions or recommendations
What is being done?
Create a summary of material about data analysis starting from types of data practitioners, duties of a data analyst, industries that need data analysts, and skills a data analyst must have
Create a summary of material about key business metrics starting from what is a business metric, relationship between KPIs and metrics, types of KPIs, main KPIs in business, how to monitor metrics and KPIs in business, difference between descriptive analytics and diagnostic analytics, and tools commonly used to monitor metrics and kpis