<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Vicky Clarissa Jennie Damara on Medium]]></title>
        <description><![CDATA[Stories by Vicky Clarissa Jennie Damara on Medium]]></description>
        <link>https://medium.com/@damaravicky?source=rss-dff7e725d4ac------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*9GuGqK0dsMIqROc48jkzvw.jpeg</url>
            <title>Stories by Vicky Clarissa Jennie Damara on Medium</title>
            <link>https://medium.com/@damaravicky?source=rss-dff7e725d4ac------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 09:19:46 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@damaravicky/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Project #01 Data Analysis and Visualization Using Python and Google Looker Studio]]></title>
            <link>https://medium.com/@damaravicky/project-01-data-analysis-and-visualization-using-python-and-google-looker-studio-73e48a425bbc?source=rss-dff7e725d4ac------2</link>
            <guid isPermaLink="false">https://medium.com/p/73e48a425bbc</guid>
            <category><![CDATA[data]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[data-visualization]]></category>
            <category><![CDATA[google-data-studio]]></category>
            <category><![CDATA[data-analysis]]></category>
            <dc:creator><![CDATA[Vicky Clarissa Jennie Damara]]></dc:creator>
            <pubDate>Tue, 15 Aug 2023 08:20:38 GMT</pubDate>
            <atom:updated>2023-08-15T12:11:09.081Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QtFWl8lkz6vzstUdX03b2A.jpeg" /><figcaption><strong>Photo by </strong><a href="https://unsplash.com/@kmuza?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText"><strong>Carlos Muza</strong></a><strong> on </strong><a href="https://unsplash.com/photos/hpjSkU2UYSU?utm_source=unsplash&amp;utm_medium=referral&amp;utm_content=creditCopyText"><strong>Unsplash</strong></a></figcaption></figure><p>In this project, we will use Python for data preparation and visualization of the results with Google Looker Studio. The dataset we will use is provided by the VIX Program BI Analyst Bank Muamalat x Rakamin in the form of 4 csv files: customer, orders, product category, and product.</p><p>In this project, we will review the following steps:</p><ol><li>Step 1: Data Preparation</li><li>Step 2: Exploratory Data Analysis</li><li>Step 3: Drawing Insights</li><li>Step 4: Visualizing with Google Looker Studio</li></ol><h3><strong>Step 1: Data Preparation</strong></h3><p>In this step, we will handle missing data, invalid values, remove duplicates, and perform formatting to discover patterns for insights using Python. To begin, we will import essential Python libraries such as pandas, numpy, seaborn, and matplotlib.</p><pre>import pandas as pd<br>import numpy as np<br>import seaborn as sns<br>import matplotlib as mpl<br>import matplotlib.pyplot as plt<br>import matplotlib.ticker as tkr</pre><p>We’ll import the dataset and examine basic information within our dataframe.</p><pre>df_customer = pd.read_csv(&#39;D:\RAKAMIN\INTERNSHIP\BANK MUAMALAT BI ANALYST\FINAL PROJECT BI MUAMALAT\FINAL PROJECT BI MUAMALAT\Customers.csv&#39;)<br>df_customer</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*emy8rxbCIRT3NnzSyjav7w.png" /></figure><pre>df_order = pd.read_csv(&#39;D:\RAKAMIN\INTERNSHIP\BANK MUAMALAT BI ANALYST\FINAL PROJECT BI MUAMALAT\FINAL PROJECT BI MUAMALAT\Orders.csv&#39;)<br>df_order.tail()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/449/1*wYcO6UJx1GLRUG9xS5lErQ.png" /></figure><p>In the dataset, there is a missing value with the order ID ‘3339’ and inconsistency DateOrder within the orders. To rectify this inconsistency, we remove the erroneous entry.</p><pre>df_orders = df_order.drop([3338], axis = 0)<br>df_orders.tail()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/445/1*-IzZ5rsHP9j8mKYFVyR-Bg.png" /></figure><pre>df_product = pd.read_csv(&#39;D:\RAKAMIN\INTERNSHIP\BANK MUAMALAT BI ANALYST\FINAL PROJECT BI MUAMALAT\FINAL PROJECT BI MUAMALAT\Products.csv&#39;)<br>df_product.head()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/442/1*CedWtvc-6seQ1xB_hw0fBQ.png" /></figure><pre>df_productcategory = pd.read_csv(&#39;D:\RAKAMIN\INTERNSHIP\BANK MUAMALAT BI ANALYST\FINAL PROJECT BI MUAMALAT\FINAL PROJECT BI MUAMALAT\ProductCategory.csv&#39;)<br>df_productcategory.head()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/394/1*oWWs5gN_HzxhVg9cFCJBKg.png" /></figure><p>By using the .info() function, we gather essential details such as the total count of rows and columns, the presence of null values, and the data types of each column. Subsequently, we move forward to inspect the existence of duplicate data with duplicated().any() function. Our observation confirms that neither null values nor duplicated entries are present within the dataset. After handling certain inconsistencies, the next step involves merging distinct CSV files into one dataframe through an inner join.</p><pre>df_1 = df_order.merge(df_customer, on = &#39;CustomerID&#39;, how = &#39;inner&#39;)<br>df_2 = df_1.merge(df_product, on = &#39;ProdNumber&#39;, how = &#39;inner&#39;)<br>df = df_2.merge(df_productcategory,<br>                left_on = &#39;Category&#39;,<br>                right_on = &#39;CategoryID&#39;,<br>                how = &#39;inner&#39;)</pre><h3><strong>Step 2: Exploratory Data Analysis (EDA)</strong></h3><h4><strong>How much revenue for each year?</strong></h4><p>During this step, we explore the data to find the revenue for each year and create simple visualizations using seaborn. We start by calculating the total sales by multiplying quantity and price to create a ‘total sales’ column.</p><pre>df[&#39;TotalSales&#39;] = df[&#39;Quantity&#39;]*df[&#39;Price&#39;]<br>df.head()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*o81aRp5m88u954yDWWootw.png" /></figure><p>Additionally, we convert the order date from the object data type to date time and extract the year value as a new ‘year’ column. We perform a groupby operation on the year with the aggregation of the sum of total sales.</p><pre>df_year = df_1.groupby(&#39;Year&#39;).agg({&#39;TotalSales&#39; : &#39;sum&#39;}).reset_index()<br>df_year</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/209/1*EFH7SpoMUl5MH9TJOBfbQw.png" /></figure><p>For better understanding, we use seaborn to create a barplot.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/650/1*MKJrbodA15JljluZLy8CPQ.png" /></figure><p><strong>The barplot illustrating the revenue from year to year indicates a significant decrease of 7.84% in 2021, resulting in revenue totaling only $841,540. This reflects a gap of $71,670 compared to the previous year.</strong></p><h4><strong>How about month to month revenue?</strong></h4><p>We begin by creating a ‘month’ column by converting the order date. Execute a multi-groupby operation based on year and month, aggregating the sum of total sales and sorting the months in ascending order.</p><pre>df_month = df_2.groupby([&#39;Year&#39;, &#39;Month&#39;]).agg({&#39;TotalSales&#39; : &#39;sum&#39;}).reset_index()<br>sort_order = [&#39;January&#39;,&#39;February&#39;,&#39;March&#39;,&#39;April&#39;,&#39;May&#39;,&#39;June&#39;,&#39;July&#39;,&#39;August&#39;,&#39;September&#39;,&#39;October&#39;,&#39;November&#39;,&#39;December&#39;]<br>df_month.index = pd.CategoricalIndex(df_month[&#39;Month&#39;],categories=sort_order,ordered=True)<br>df_month = df_month.sort_index().reset_index(drop=True)<br>df_month.head()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/285/1*5aB7jTbLFatEEH0HRAKQ9w.png" /></figure><p>This information is visualized through a line plot created with seaborn.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hw-YC098GiQtQ6qbjtx2rQ.png" /></figure><p><strong>The line plot exposes a trend where the majority of monthly revenues in 2021 have decreased in comparison to the previous year, except for April and June. Notably, June 2021 stands out with the highest sales among all months.</strong></p><h4>How about the distribution of price?</h4><p>Using a histogram plot, we can gain insights into the distribution of prices across the entire transaction dataset.</p><pre>fig, ax = plt.subplots()<br>ax = sns.histplot( x = &#39;Price&#39;, data = df, color = &#39;green&#39;)<br>ax.set_xlabel(&#39;Price&#39;, fontsize = 10)<br>ax.set_ylabel(&#39;Total Transaksi&#39;, fontsize = 10)<br>ax.set_title(&#39;Distribution of Prices&#39;, fontsize = 15, fontweight = &#39;bold&#39;)</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/663/1*XPV3GxGSnCAg7n-dIjxy1w.png" /></figure><p>The histogram displays a positively skewed pattern. Based on this visualization, <strong>we can infer that our customers are inclined to make purchases of products priced around $37.99, while showing less interest in products priced above $189</strong>.</p><h4>How about the number of customers? Have there been any decreases or increases?</h4><p>Considering the decline of 7.84% in revenue for 2021 in contrast to the preceding year, we investigate its effect on the count of distinct customers. To determine the total count of unique customers, we execute a groupby operation based on the year, aggregating the count of distinct CustomerID.</p><pre>df_uniq = df_1.groupby(&#39;Year&#39;).agg({&#39;CustomerID&#39; : &#39;nunique&#39;}).reset_index()<br>df_uniq</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/219/1*NzGU7Ces1AaIQ8PAtHeSAQ.png" /></figure><p><strong>The resulting barplot indicates a loss of 4 customers.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/657/1*BdjP5DsyZY8KAHvLaeZXqA.png" /></figure><p>Let’s explore into the specifics of our customers. <strong>Are there any customers who did not make a repurchase in 2021? If so, how many such customers are there?</strong></p><pre>df_2021null = pivot_cust[(pivot_cust[&#39;OrderID_2021&#39;].isnull()) &amp; (pivot_cust[&#39;OrderID_2020&#39;] &gt; 0)]<br>df_2021null.describe()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/334/1*6dzpZmO-PP2qyT4p4GMwNA.png" /></figure><p>Using the .describe() function, we will specifically focus on the counts, maximum, and minimum values. Through this, we conclude that a total of <strong>524 customers did not make any repurchases in 2021</strong>. The highest transaction count for an individual customer is 5, attributed to customerID number 422.</p><p><strong>Did we acquire any new customers in 2021? If yes, how many?</strong></p><pre>df_2020null = pivot_cust[(pivot_cust[&#39;OrderID_2020&#39;].isnull()) &amp; (pivot_cust[&#39;OrderID_2021&#39;] &gt; 0)]<br>df_2020null</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/396/1*AMGVID-oz15U5kIRZJE4rA.png" /></figure><p>By sorting the pivot table using the criteria of CustomerID, Year, and OrderID, we can deduce that the total count of<strong> new customers in the year 2021 amounts to 520 customers</strong>.</p><p><strong>Furthermore, are there any loyal customers who have consistently made purchases? How many such loyal customers exist and how frequently have they made purchases?</strong></p><pre>df_royalcust = pivot_cust[(pivot_cust[&#39;OrderID_2021&#39;] &gt; 0) &amp; (pivot_cust[&#39;OrderID_2020&#39;] &gt; 0)]<br>df_royalcust.sort_values(&#39;OrderID_2021&#39;, ascending = False)</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/410/1*vzyPdtfz_Qq6307jC5hHlw.png" /></figure><p><strong>There exist 637 loyal customers who have continuously made purchases of our products from 2020 to 2021</strong>. The transaction counts for these loyal customers are as follows. Notably, CustomerID number 496 holds the highest transaction count, indicating a significant increase in transactions compared to the previous year.</p><h4>What about the comparison of product annual sales between 2020 and 2021? Have there been any decreases or increases in sales?</h4><p>Let’s examine the sales for each product category. Are there any products that have seen an increase in sales?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/922/1*m0AMuPnWBqszEQOffga4ow.png" /></figure><p>From the barplot we have created, we can deduce that, <strong>on average, product sales year over year have decreased</strong>. However, exceptions can be observed in the case of Drones and E-Books, which have shown increased sales.</p><p><strong>Let’s further explore the cities of our customers.</strong></p><h4><strong>Have there been any cities that have experienced a significant loss of our customers? On the other hand, have there been any cities that have gained new customers for our products?</strong></h4><p>To explore the geographical distribution, initiate a groupby operation on the ‘CustomerCity’ and ‘Year’ columns. Aggregate the count of unique CustomerID values to assess the distribution within each city and year.</p><pre>df_city = df_1.groupby([&#39;CustomerCity&#39;, &#39;Year&#39;]).agg({&#39;CustomerID&#39; :&#39;nunique&#39;}).reset_index()<br>df_city.head()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/335/1*o-_IQZjnHnLh0HtI3wbWXQ.png" /></figure><p>Set ‘CustomerCity’ as the index, ‘Year’ as the columns, and utilize ‘CustomerID’ as the values within the pivot table. This arrangement will provide a clearer overview of the data.</p><pre>pivot_city = pd.pivot_table(df_city,<br>                            index = [&#39;CustomerCity&#39;],<br>                            columns = [&#39;Year&#39;],<br>                            values = [&#39;CustomerID&#39;])<br>pivot_city.columns = [&#39;_&#39;.join(str(s).strip() for s in col if s) for col in pivot_city.columns]<br>pivot_city.reset_index()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/485/1*uNRo3qWw1jug9pgGDrqBLg.png" /></figure><p>Our customers between the years 2020 and 2021 originate from a total of 361 cities. It’s worth highlighting that i<strong>n 2021, there are 38 newly added cities where our customers are located</strong>.</p><pre>df_2020nullcity = pivot_city[(pivot_city[&#39;CustomerID_2020&#39;].isnull()) &amp; (pivot_city[&#39;CustomerID_2021&#39;] &gt; 0)]<br>df_2020nullcity.head()</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/462/1*IxD3lc-gAe4sKLbS3Qhgzg.png" /></figure><p>Regrettably, we experienced a <strong>reduction of 47 cities compared to the previous year</strong>.</p><pre>df_2021nullcity = pivot_city[(pivot_city[&#39;CustomerID_2021&#39;].isnull()) &amp; (pivot_city[&#39;CustomerID_2020&#39;] &gt; 0)]<br>df_2021nullcity.head(10)</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/487/1*MsI29tZQzGke1VkKdoyc6g.png" /></figure><p>Among these,<strong> there are 276 cities with loyal customer bases</strong>, and interestingly, <strong>88 cities within this group witnessed an increase in customers year over year</strong>.</p><pre>df_citygrowthpositif = pivot_city[(pivot_city[&#39;CustomerID_2021&#39;]) &gt; (pivot_city[&#39;CustomerID_2020&#39;])]<br>df_citygrowthpositif.sort_values(&#39;CustomerID_2021&#39;, ascending = False)</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/499/1*Ipb9WqcOy-x5UWxQFoagnw.png" /></figure><p>Conversely, it’s important to acknowledge that <strong>82 cities within this loyal customer group experienced a decrease in customer count</strong>.</p><pre>df_citygrowthnegatif = pivot_city[(pivot_city[&#39;CustomerID_2021&#39;]) &lt; (pivot_city[&#39;CustomerID_2020&#39;])]<br>df_citygrowthnegatif.sort_values(&#39;CustomerID_2021&#39;, ascending = False)</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/466/1*fFeQPh95uebSSCENom8Y2A.png" /></figure><blockquote><strong><em>The link to the GitHub repository for this project is provided </em></strong><a href="https://github.com/vickyclarissa/project-01.git"><strong><em>here</em></strong></a><strong><em>.</em></strong></blockquote><h3>Step 3: What strategies we can take to increase our sales performance?</h3><p>After conducting an exploratory data analysis, we have identified several strategies to enhance our sales performance:</p><ol><li><strong>Price Optimization</strong>: There are noticeable differences in the distribution of prices across products. By adjusting prices strategically, we can tap into a wider customer base.</li><li><strong>Expand Distribution</strong>: The loss of customers from 47 cities highlights the need to ensure a more balanced distribution of our products. We should focus on expanding our presence both offline and online to ensure convenient access for customers across various locations.</li><li><strong>Product Optimization</strong>: Certain products, such as Training Videos and E-Books, have proven to be favorites among customers. Moreover, Drones have shown potential with a slight sales increase in 2021. On the other hand, products like Blueprints and Drone Kits witnessed a decline in sales. By targeting promotions, enhanced features, and introduce improvements we can capture more attention. Products with lower demand can be offered discounts or cashback with clear terms and conditions.</li><li><strong>Customer Engagement and Retention</strong>: With our existing customer database, we can take different approaches to engage and retain customers: <strong>Offer discounts or cashback</strong> through email or messages, particularly to customers at risk of churning and those who have already churned; <strong>Reward loyalty</strong> to our loyal customers as a sign of appreciation for their continued support.</li></ol><h3><strong>Step 4: Creating Sales Dashboard with Google Looker Studio</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4_Wb42Or8-hZbDyYlsyA7w.png" /></figure><p>Here is the final part of my first project: Data Analysis and Visualization Using Python and Google Looker Studio.</p><blockquote><strong><em>I have created a sales dashboard based on the dataset we previously analyzed and I have included the link </em></strong><a href="https://lookerstudio.google.com/s/pbfbEIMepHo"><strong><em>here</em></strong></a><strong><em>.</em></strong></blockquote><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=73e48a425bbc" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>