<?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 Ayoolaaishat on Medium]]></title>
        <description><![CDATA[Stories by Ayoolaaishat on Medium]]></description>
        <link>https://medium.com/@ayoolaaishat510?source=rss-a5ca5b1095b1------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*hG-Jqqc2lnTTY3Hs</url>
            <title>Stories by Ayoolaaishat on Medium</title>
            <link>https://medium.com/@ayoolaaishat510?source=rss-a5ca5b1095b1------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 04:07:10 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@ayoolaaishat510/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[Data Preparation in Power BI: Cleaning, Transforming, and Loading Data]]></title>
            <link>https://medium.com/@ayoolaaishat510/data-preparation-in-power-bi-cleaning-transforming-and-loading-data-1fb3cd92d6ae?source=rss-a5ca5b1095b1------2</link>
            <guid isPermaLink="false">https://medium.com/p/1fb3cd92d6ae</guid>
            <dc:creator><![CDATA[Ayoolaaishat]]></dc:creator>
            <pubDate>Fri, 08 May 2026 14:11:20 GMT</pubDate>
            <atom:updated>2026-05-08T14:11:20.600Z</atom:updated>
            <content:encoded><![CDATA[<h4>Introduction</h4><p>In data analysis, the quality of results depends greatly on the quality of the data being used. Most datasets collected from organizations, surveys, websites, or businesses usually contain errors such as duplicates, empty fields, incorrect formats, and inconsistent values. If these issues are not corrected before analysis, they can lead to inaccurate reports and poor decision-making.</p><p>Power BI provides a powerful feature called <strong>Power Query Editor</strong>, which helps analysts clean, organize, and transform raw data into a structured format suitable for reporting and visualization. Data preparation is therefore an important process because it improves data accuracy, consistency, and reliability.</p><p>This article explains different data preparation techniques in Power BI and how they can be used to transform messy data into meaningful insights.</p><h4>Understanding Power Query Editor</h4><p>Power Query Editor is the environment in Power BI where data transformation and cleaning take place before the data is loaded into the report view.</p><p>To open Power Query Editor:</p><ol><li>Launch Power BI Desktop</li><li>Import your dataset</li><li>Click on <strong>Transform Data</strong></li></ol><p>The Power Query Editor contains:</p><ul><li>Queries pane</li><li>Ribbon tools</li><li>Data preview section</li><li>Applied steps section</li></ul><p>One advantage of Power Query Editor is that every transformation step performed is automatically saved in the Applied Steps pane, making the cleaning process easier to track and edit.</p><p>Power Query follows the ETL process:</p><ul><li><strong>Extract</strong> data from different sources</li><li><strong>Transform</strong> the data into a useful format</li><li><strong>Load</strong> the cleaned data into Power BI</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lSonDAwbQdsB8ET6m3lCxw.png" /><figcaption>Power Query Editor</figcaption></figure><h4>Creating Index Columns</h4><p>An Index Column is used to assign sequential numbers to rows in a table. This helps create unique identifiers and improves data organization.</p><p>For example</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/645/1*xZ_hfLH0Wgvg4r9f-MFDFQ.png" /></figure><h4>Steps for Creating an Index Column</h4><ol><li>Open Power Query Editor</li><li>Click on <strong>Add Column</strong></li><li>Select <strong>Index Column</strong></li><li>Choose:</li></ol><ul><li>From 0</li><li>From 1</li><li>Custom</li></ul><p>Index columns are useful when:</p><ul><li>Ranking records</li><li>Tracking row positions</li><li>Creating IDs</li><li>Sorting data properly</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7as9kTPOvfch0j--bnagbw.png" /><figcaption>Index column from 0</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*t2UqqvgigCPmiqACcVUQsw.png" /><figcaption>Index column from 1</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rWUv-A0nPprAF-crqGVnXA.png" /><figcaption>custom</figcaption></figure><h4>Custom Column index</h4><p>A <strong>Custom Index Column</strong> allows you to create a numbering system that starts from any number you choose instead of the default 0 or 1.</p><p>This is useful when:</p><ul><li>You want your records to begin from a specific number</li><li>You are continuing an existing numbering sequence</li><li>You want customized IDs for business records</li></ul><h4>Conditional Columns and Logic</h4><p>Conditional Columns allow users to create new columns based on specified conditions. This works similarly to IF statements in Excel.</p><p>For example, a business may want to classify sales performance as:</p><ul><li>High Sales</li><li>Average Sales</li><li>Low Sales</li></ul><p>Based on sales amount.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/622/1*VjIFH5nCQmmJjFc-06bXMQ.png" /></figure><h4>Steps</h4><ol><li>Go to <strong>Add Column</strong></li><li>Select <strong>Conditional Column</strong></li><li>Set conditions and output values</li></ol><p>Conditional columns help automate business rules and simplify data categorization</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ipbfDZokg0MPpg3jeFnpJg.png" /><figcaption>conditional column</figcaption></figure><h4>Using Column from Examples</h4><p>The “Column from Examples” feature helps users generate new columns based on patterns detected from examples provided manually.</p><p>For instance:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/673/1*hukxVIEX7ZwMfsRNyPiOUg.png" /></figure><p>Power BI automatically recognizes the transformation pattern and applies it to other rows.</p><h4>Steps</h4><ol><li>Click <strong>Add Column</strong></li><li>Select <strong>Column from Examples</strong></li><li>Enter sample outputs</li></ol><p>This feature reduces manual work and improves efficiency during data transformation.</p><h4>Grouping and Aggregation</h4><p>Grouping helps summarize large datasets into meaningful information.</p><p>For example, a company may want to calculate total sales by region.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/550/1*w4HkR3vqvaCVDCMlHIQbFQ.png" /></figure><h4>Steps</h4><ol><li>Select the column to group</li><li>Click <strong>Group By</strong></li><li>Choose the aggregation type:</li></ol><ul><li>Sum</li><li>Count</li><li>Average</li><li>Maximum</li><li>Minimum</li></ul><p>Grouping and aggregation help analysts identify trends and summarize business performance quickly.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*15BBYhRmLay16gjBqg2Akw.png" /><figcaption>Grouping by count</figcaption></figure><h4>Pivoting and Unpivoting Columns</h4><p>Pivoting and unpivoting are important techniques for restructuring datasets.</p><h4>Pivoting Columns</h4><p>Pivoting converts row values into columns.</p><p>Example:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/612/1*2ZOFCbhWLWka3Jrxz7xsbw.png" /></figure><h4>Steps to Pivot Columns in Power BI</h4><ol><li>Open <strong>Power Query Editor</strong></li><li>Select the column you want to pivot</li><li>Go to the <strong>Transform</strong> tab</li><li>Click <strong>Pivot Column</strong></li><li>Choose the values column</li><li>Select the aggregation option such as:</li></ol><ul><li>Sum</li><li>Count</li><li>Average</li></ul><p>7. Click <strong>OK</strong></p><p>After pivoting, the selected row values become column headers.</p><h4>Unpivoting Columns</h4><p>Unpivoting converts columns into rows.</p><p>Example:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/694/1*CCVmJ8RWibHFkmUz1cnVOw.png" /></figure><p>Unpivoting is very useful when preparing datasets for dashboards and visualizations.</p><h4>Steps to Unpivot Columns in Power BI</h4><ol><li>Open <strong>Power Query Editor</strong></li><li>Select the columns you want to unpivot</li><li>Go to the <strong>Transform</strong> tab</li><li>Click <strong>Unpivot Columns</strong></li></ol><p>You can choose:</p><ul><li><strong>Unpivot Columns</strong></li><li><strong>Unpivot Other Columns</strong></li></ul><p>After unpivoting, Power BI automatically creates:</p><ul><li>Attribute column</li><li>Value column</li></ul><h4>Importance of Pivoting and Unpivoting</h4><p>These transformations help:</p><ul><li>Organize datasets properly</li><li>Improve report structure</li><li>Prepare data for visualization</li><li>Simplify analysis</li><li>Convert messy data into a standardized format</li></ul><p>Pivoting and unpivoting are widely used in business reporting and dashboard creation because they improve data flexibility and readability.</p><h4>Merging and Appending Queries</h4><p>Power BI allows users to combine multiple datasets using Merge and Append operations.</p><h4>Merging Queries</h4><p>Merging combines tables horizontally using a common field.</p><p>For example:</p><ul><li>Customer Table</li><li>Orders Table</li></ul><p>Both tables can be merged using Customer ID.</p><p>This works similarly to SQL joins.</p><h4>Steps to Merge Queries in Power BI</h4><ol><li>Open <strong>Power Query Editor</strong></li><li>Go to the <strong>Home</strong> tab</li><li>Click <strong>Merge Queries</strong></li><li>Select the first table</li><li>Select the second table</li><li>Choose the matching column from both tables</li><li>Select the join type:</li></ol><ul><li>Left Outer</li><li>Right Outer</li><li>Inner</li><li>Full Outer</li></ul><p>8. Click <strong>OK</strong></p><p>9. Expand the merged column to display the required fields</p><p>After merging, related information from both tables becomes available in a single dataset.</p><h4>Importance of Merging Queries</h4><p>Merging helps to:</p><ul><li>Combine related datasets</li><li>Reduce data duplication</li><li>Improve analysis accuracy</li><li>Create complete business reports</li></ul><h4>Appending Queries</h4><p>Appending queries combines datasets vertically by adding rows from one table below another table.</p><p>This is useful when datasets have the same column structure.</p><p>For example:</p><ul><li>January Sales Table</li><li>February Sales Table</li><li>March Sales Table</li></ul><p>These tables can be appended into one complete sales dataset.</p><h4>Steps to Append Queries in Power BI</h4><ol><li>Open <strong>Power Query Editor</strong></li><li>Go to the <strong>Home</strong> tab</li><li>Click <strong>Append Queries</strong></li><li>Choose:</li></ol><ul><li>Two tables</li><li>Three or more tables</li></ul><p>5. Select the tables to combine</p><p>6. Click <strong>OK</strong></p><p>Power BI automatically stacks the rows together into a single table</p><h4>Importance of Appending Queries</h4><p>Appending queries helps to:</p><ul><li>Combine monthly or yearly reports</li><li>Consolidate datasets from different branches</li><li>Simplify reporting</li><li>Save time during analysis</li></ul><h4>Date and Time Transformations</h4><p>Date transformation is important for time-based analysis in Power BI.</p><p>Power BI allows users to extract:</p><ul><li>Year</li><li>Month</li><li>Quarter</li><li>Day</li></ul><p>For example:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/751/1*20De9PY7ESdz4yR0N6dd6w.png" /></figure><p>Date transformations are useful for:</p><ul><li>Trend analysis</li><li>Monthly reports</li><li>Forecasting</li><li>Time intelligence</li></ul><h4>Steps for Date and Time Transformations in Power BI</h4><ol><li>Open <strong>Power Query Editor</strong></li><li>Select the column containing date values</li><li>Go to the <strong>Add Column</strong> tab or <strong>Transform</strong> tab</li><li>Click on <strong>Date</strong></li><li>Choose the transformation you want such as:</li></ol><ul><li>Year</li><li>Month</li><li>Quarter</li><li>Day</li><li>Week</li><li>Start of Month</li><li>End of Month</li></ul><p>For time transformation:</p><ol><li>Select the time column</li><li>Click Time</li><li>Choose:</li></ol><ul><li>Hour</li><li>Minute</li><li>Second</li></ul><p>Power BI automatically creates a new column containing the transformed values.</p><p>Example of Date Transformation</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/783/1*CwWCkLtkxBnz3TpFH2k2FA.png" /></figure><h4>Importance of Date and Time Transformations</h4><p>Date and time transformations help analysts to:</p><ul><li>Analyze sales trends over time</li><li>Compare monthly or yearly performance</li><li>Build time-based dashboards</li><li>Improve reporting accuracy</li><li>Perform forecasting and seasonal analysis</li></ul><p>These transformations are widely used in business intelligence because most organizational reports depend heavily on time-based data.</p><h4>Adding Prefix and Suffix in Power Query</h4><p>Power BI also allows users to add prefixes or suffixes to values in a column.</p><p>Example:</p><p>Before:</p><ul><li>001</li><li>002</li></ul><p>After:</p><ul><li>EMP-001</li><li>EMP-002</li></ul><p>This helps standardize data formatting.</p><h4>Steps</h4><ol><li>Select the column</li><li>Go to <strong>Transform</strong></li><li>Choose <strong>Format</strong></li><li>Add Prefix or Suffix</li></ol><h4>Data Profiling Techniques</h4><p>Data profiling helps users understand the quality and structure of datasets.</p><p>Power BI provides three major profiling tools:</p><h4>Column Quality</h4><p>Shows:</p><ul><li>Valid values</li><li>Error values</li><li>Empty values</li></ul><h4>Column Distribution</h4><p>Displays:</p><ul><li>Distinct values</li><li>Duplicate values</li><li>Unique values</li></ul><h4>Column Profile</h4><p>Provides detailed statistics for each column.</p><p>These tools help identify data issues before analysis begins.</p><h4>Handling Missing or Null Values</h4><p>Missing values can reduce the accuracy of reports and dashboards.</p><p>Power BI provides several ways to handle null values.</p><h4>Replacing Null Values</h4><p>Null values can be replaced with:</p><ul><li>Zero</li><li>Average values</li><li>Custom text</li></ul><h4>Removing Null Values</h4><p>Rows containing missing values can also be removed completely.</p><h4>Removing Duplicates</h4><p>Duplicate records may affect calculations and lead to incorrect insights.</p><p>To remove duplicates:</p><ol><li>Select the column</li><li>Click <strong>Remove Duplicates</strong></li></ol><p>Handling missing data improves the reliability and consistency of analysis.</p><h4>Practical Application of Data Preparation</h4><p>Imagine a retail business collecting sales data from different branches. The dataset may contain:</p><ul><li>Duplicate transactions</li><li>Missing customer information</li><li>Incorrect date formats</li><li>Inconsistent product names</li></ul><p>Using Power Query Editor, the business can clean and organize the dataset before creating dashboards. This ensures accurate reporting and better decision-making.</p><h4>Conclusion</h4><p>Data preparation is one of the most important stages in data analysis because clean data produces reliable insights. Power BI’s Power Query Editor provides effective tools for cleaning, transforming, and organizing data before visualization.</p><p>Techniques such as creating conditional columns, grouping data, merging queries, handling null values, and profiling datasets help analysts prepare high-quality data for reporting and decision-making.</p><p>By understanding and applying these techniques, data analysts can improve report accuracy, simplify workflows, and build more effective dashboards in Power BI.</p><h4>References</h4><ul><li>Microsoft. (2025). <em>Power Query documentation</em>. Retrieved from<br> <a href="https://learn.microsoft.com/en-us/power-query/?utm_source=chatgpt.com">Microsoft Learn Power Query Documentation</a></li><li>Microsoft. (2025). <em>Power BI documentation</em>. Retrieved from<br> <a href="https://learn.microsoft.com/en-us/power-bi/?utm_source=chatgpt.com">Microsoft Learn Power BI Documentation</a></li><li>Foster, M. (2020). <em>Power BI Complete Reference</em>. Packt Publishing.</li><li>Ferrari, A., &amp; Russo, M. (2021). <em>Analyzing Data with Power BI and Power Pivot for Excel</em>. Microsoft Press.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1fb3cd92d6ae" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Getting Data from Different Sources in Power BI: A Beginner-Friendly Guide]]></title>
            <link>https://medium.com/@ayoolaaishat510/getting-data-from-different-sources-in-power-bi-a-beginner-friendly-guide-5631bbf31d4d?source=rss-a5ca5b1095b1------2</link>
            <guid isPermaLink="false">https://medium.com/p/5631bbf31d4d</guid>
            <dc:creator><![CDATA[Ayoolaaishat]]></dc:creator>
            <pubDate>Fri, 03 Apr 2026 10:32:14 GMT</pubDate>
            <atom:updated>2026-04-03T10:32:14.247Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Introduction</strong></p><p>Power BI is a powerful business intelligence tool that allows data analysts to transform raw data into meaningful insights. One of its greatest strengths is its ability to connect to data from multiple sources, making it easy to analyze and visualize information in one place.</p><p>In this article, I will walk you through how to get data from different sources in Power BI, with practical examples and simple explanations.</p><p>Why Connecting Multiple Data Sources Matters In real-world scenarios, data is rarely stored in one place. As a data analyst, you may need to combine:</p><p>•	Excel files</p><p>•	SQL databases</p><p>•	Web data</p><p>•	Cloud platforms</p><p>Power BI makes this process seamless by allowing you to connect, transform, and merge data efficiently.</p><p><strong>Common Data Sources in Power BI</strong></p><p>Power BI supports a wide range of data sources, including:</p><ol><li><strong>Excel Files</strong></li></ol><p>Excel is one of the most commonly used data sources.</p><p>Steps:</p><p>1.	Open Power BI Desktop</p><p>2.	Click on Get Data</p><p>3.	Select Excel Workbook</p><p>4.	Load your file</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jgrKXKivy5BZjVpO_-icPg.png" /><figcaption><strong>Excel File</strong></figcaption></figure><p><strong>2. SQL Server Database</strong></p><p>For structured and large datasets, SQL databases are ideal.</p><p>Steps:</p><p>1.	Click Get Data</p><p>2.	Select SQL Server</p><p>3.	Enter server and database name</p><p>4.	Choose Import or DirectQuery</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*v3fgG4Q_jAGts9_0Pierqw.png" /><figcaption><strong>SQL File</strong></figcaption></figure><p><strong>3. Web Data</strong></p><p>You can extract data directly from websites.</p><p>Steps:</p><p>1.	Click Get Data</p><p>2.	Select Web</p><p>3.	Paste the URL</p><p>4.	Load the data</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*a22GTJGfbJgSADlN7rGiww.png" /><figcaption><strong>Web Data</strong></figcaption></figure><p><strong>4. CSV Files</strong></p><p>CSV files are lightweight and easy to use.</p><p>Steps:</p><p>1.	Click Get Data</p><p>2.	Select Text/CSV</p><p>3.	Load your file</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QjaCxioVGxvk4zq4_EW-rQ.png" /><figcaption><strong>CSV File</strong></figcaption></figure><p>5.<strong> Cloud Sources (e.g., SharePoint, OneDrive)</strong></p><p>Power BI can connect to cloud-based storage.</p><p><strong>Transforming Data Using Power Query</strong></p><p>After loading data, it often needs cleaning and transformation.</p><p>With Power Query Editor, you can:</p><p>•	Remove duplicates</p><p>•	Filter rows</p><p>•	Rename columns</p><p>•	Change data types</p><p>•	Merge multiple tables</p><p><strong>Combining Data from Multiple Sources</strong></p><p>One of Power BI’s strongest features is data integration.</p><p>You can:</p><p>•	Merge queries (like SQL JOIN)</p><p>•	Append queries (combine tables)</p><p>Example:</p><p>You can merge customer data from Excel with sales data from SQL.</p><p><strong>Conclusion</strong></p><p>Getting data from different sources in Power BI is a fundamental skill for any data analyst. Mastering this process allows you to work with real-world datasets and create powerful dashboards that drive decision-making.</p><p>As I continue my learning journey, I am excited to explore more advanced Power BI features and apply them to real-life projects.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5631bbf31d4d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Driving Retail Growth Through Customer & Sales Analytics]]></title>
            <link>https://medium.com/@ayoolaaishat510/driving-retail-growth-through-customer-sales-analytics-b3938f20cc54?source=rss-a5ca5b1095b1------2</link>
            <guid isPermaLink="false">https://medium.com/p/b3938f20cc54</guid>
            <dc:creator><![CDATA[Ayoolaaishat]]></dc:creator>
            <pubDate>Mon, 26 Jan 2026 07:25:46 GMT</pubDate>
            <atom:updated>2026-01-26T07:25:46.130Z</atom:updated>
            <content:encoded><![CDATA[<h3>Executive Overview</h3><p>This project was designed with a <strong>business audience</strong> in mind specifically retail managers, marketing teams, and decision-makers seeking practical insights rather than technical detail. The objective of this analysis is to uncover <strong>customer spending patterns and purchase behaviors</strong> that directly influence revenue performance and customer lifetime value at SellCheapy Retail. Specifically, the analysis focuses on understanding <strong>who buys, how often they buy, what they buy together, and where growth or churn risks exist</strong>.</p><p>By transforming transactional and customer-level data into actionable intelligence, this dashboard addresses key business questions such as:</p><ul><li>Which customer segments contribute the most to revenue and gross profit?</li><li>How strong is customer retention, and which RFM segments require immediate attention?</li><li>What products are most frequently purchased together, and how can this inform bundling and cross-sell strategies?</li><li>How do purchase behaviors differ across regions and customer types (one-time, regular, frequent)?</li><li>How do demographic factors such as education level and profession influence average order value?</li></ul><p>The insights are presented in a concise, decision-focused format to support <strong>sales optimization, targeted retention strategies, regional performance improvement, and data-driven strategic planning</strong>.</p><h3>Introduction</h3><p>In today’s competitive retail environment, having access to data is no longer enough organizations must be able to convert data into meaningful business insight. Many retail businesses struggle with understanding customer value, purchasing behavior, and product performance despite having large volumes of transactional data.</p><p>This project explores how SQL and Power BI can be used to bridge this gap by analyzing customer transactions and translating them into actionable insights that support business growth and performance improvement.</p><p><strong>Statement of the Problem</strong></p><p>Retail managers often face critical questions: Which customers are driving the most revenue? Which products should be promoted or bundled to increase sales? Are there regions underperforming despite potential demand?</p><p>Without a clear understanding of customer behavior and product performance, businesses risk missed opportunities, ineffective marketing strategies, and stagnant sales growth. This project addresses these challenges by leveraging SQL and Power BI to transform raw retail data into meaningful insights. By analyzing customer transactions, identifying spending patterns, and visualizing key trends, the goal is to empower retail decision-makers to act confidently, optimize revenue, and build lasting customer relationships.</p><p><strong>Dataset Used</strong></p><p>AdventureWorks2019 (OLTP version) A Microsoft sample database representing a fictional manufacturing and retail company. For this project, the dataset was used to simulate a real-world retail analytics scenario.</p><p><strong>Core Tables Analyzed</strong></p><p>•	Sales.SalesOrderHeader — Order-level information (order dates, customers, territories)</p><p>•	Sales.SalesOrderDetail — Line-item details (products, quantities, prices)</p><p>•	Sales.Customer — Customer demographic information</p><p>•	Person.Person — Personal details such as customer names</p><p>•	Production.Product — Product specifications</p><p>•	Production.ProductSubCategory — Product subcategory details</p><p>•	Production.ProductCategory — Product category details</p><p>•	Sales.SalesTerritory — Geographic sales regions</p><p><strong>Tools &amp; Technologies Used</strong></p><p>•	SQL Data exploration, table selection, joins, and logic building</p><p>•	Power BI Data modeling, DAX measures, analysis, and dashboard creation</p><p>•	ERD (Entity Relationship Diagram) Understanding data relationships</p><p><strong>The Data Analytics Process</strong></p><ol><li><strong>Understanding the Business Problem</strong></li></ol><p>I started by carefully reading and understanding the project brief to clearly identify the business questions to be answered. This step helped me understand what insights were required and guided my decisions throughout the project.</p><h4>2. Data Understanding &amp; Relevant Table Selection</h4><p>Since the database was already defined as <a href="https://drive.google.com/file/d/1YQh6lqLCXe8MnOT3W_6h5Af-BsUiiNM_/view?usp=drive_link"><strong>AdventureWorks2019 </strong>,</a> this step focused purely on understanding the data structure and selecting the most relevant tables for the analysis.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*cfpy57wziGx4nIIKSG2low.gif" /></figure><p>I reviewed the schemas to identify tables that captured customer demographics, sales transactions, product information, and geographic performance. Only business-critical tables were selected to keep the analysis focused and aligned with the project objectives.</p><p>The selected tables were later joined using SQL to create a clean, analysis-ready dataset.</p><p>3. <strong>Creating the ERD for Selected Tables</strong></p><p>I created an Entity Relationship Diagram (ERD) to visualize how the selected tables relate to each other. This step was important for understanding primary keys, foreign keys, and table relationships before writing SQL queries.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*XUBhgATNLVENimk5ClOLbQ.png" /></figure><h4>4. Writing Business Logic (SQL)</h4><p>Using SQL, I applied business logic by joining and transforming tables to create these views :</p><ul><li>Sales</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vKnbVjsIdHfzbI5gQ_LqQQ.png" /></figure><ul><li>Product</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tyzMR0LKzqdPwtT4dOzOAw.png" /></figure><ul><li><strong>Sales Territory</strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fdFJJFrYDRNtko2g6fMpOg.png" /></figure><ul><li><strong>Customer</strong></li></ul><p>This step ensured the dataset was structured correctly for analysis and reporting.</p><p><strong>5. Creating ERD for Views</strong></p><p>After creating SQL views, I built an ERD for the views to assist me while performing exploratory analysis.</p><h3>6. Exploratory Analysis</h3><p>I performed exploratory analysis with a focus on:</p><ul><li><strong>Customer Spend Analysis</strong></li><li><strong>Product Pairings (Market Basket Analysis —MBA)</strong></li></ul><p>This helped uncover patterns in customer behavior and product relationships.</p><h3>7. Product Pairing (MBA)</h3><p>I analyzed product combinations frequently purchased together to identify cross-selling and upselling opportunities.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*sy1W7AuKNHWKk7V_q-P_VA.png" /></figure><h3>8. RFM Analysis</h3><p>I conducted <strong>RFM Analysis</strong> to segment customers based on:</p><ul><li><strong>Recency</strong> — How recently a customer made a purchase</li><li><strong>Frequency</strong> — How often they purchase</li><li><strong>Monetary</strong> — How much they spend</li></ul><p>This allowed me to classify customers into meaningful segments such as high-value customers, loyal customers, and at-risk customers.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*L5I5ZHoITehSeK-VL-ObTA.png" /></figure><h3>9. Power BI Visualization &amp; Insights</h3><p>Finally, I connected the prepared data to <strong>Power BI</strong>, created DAX measures, and built interactive dashboards to visualize trends, customer segments, and product performance.</p><p><strong>Overview</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*giKPLK_3b-XqtSk2gFFMew.png" /></figure><p><strong>Analytics</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*48BLwPHyWMCd_fPvU4Ogpw.png" /></figure><p><strong>Customer Segment</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*MsOp4xIFB1tLtv-D1oFSew.png" /></figure><h3>Project Deliverables</h3><p>The project deliverables included:</p><ul><li><strong>SQL Queries</strong> for data extraction, transformation, and analysis</li><li><strong>Executive Summary Dashboard</strong> built in Power BI</li><li><strong>Customer Segmentation Report</strong> based on RFM analsis</li></ul><h3>Key Outcomes</h3><ul><li>Improved understanding of end-to-end data analysis workflow</li><li>Practical application of SQL joins, views, and business logic</li><li>Hands-on experience with customer analytics and RFM segmentation</li><li>Clear, interactive dashboards that support business decision-making</li></ul><h3>Conclusion</h3><p>This project enhanced my confidence in using <strong>SQL and Power BI together</strong> to solve real-world business problems. It reflects my ability to move from raw data to meaningful insights through structured analysis, making it a strong addition to my data analytics portfolio.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b3938f20cc54" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[MOBILE PHONE SALES DATA ANALYSIS]]></title>
            <link>https://medium.com/@ayoolaaishat510/mobile-phone-sales-data-analysis-316274ff1f81?source=rss-a5ca5b1095b1------2</link>
            <guid isPermaLink="false">https://medium.com/p/316274ff1f81</guid>
            <dc:creator><![CDATA[Ayoolaaishat]]></dc:creator>
            <pubDate>Wed, 24 Dec 2025 13:44:22 GMT</pubDate>
            <atom:updated>2025-12-24T13:44:22.344Z</atom:updated>
            <content:encoded><![CDATA[<h3>Introduction</h3><p>As a Data Analyst for a multinational mobile phone retailer, I was tasked with analyzing a comprehensive dataset containing detailed information on mobile phone transactions, customer demographics, product specifications, and geographical distribution.</p><p>To conduct this analysis, I utilized Microsoft Excel, employing features such as PivotTables, charts, and slicers to uncover patterns and derive actionable insights.</p><h3>Objectives</h3><p>The main objective of this project is to build a comprehensive report that:</p><ul><li>Tells the story of mobile phone sales across different regions.</li><li>Highlights best-selling products and emerging trends.</li><li>Provides a clear view of customer behavior based on demographics and purchase channels.</li><li>Supports the business in making informed, data-driven decisions to improve strategy and growth.</li></ul><p><strong>Data Structure</strong></p><p>The dataset contains four tables in an Excel workbook and is comprehensive enough to allow me explore patterns in sales performance, customer behavior, and regional trends.</p><ol><li><strong>Fact sales</strong> : It is the table, which captures individual sales transactions, including the date of sale, product sold, quantity, revenue, customer age group, gender, location, sales channel, and payment method. This fact table connects to three supporting dimension tables.</li><li><strong>Dim Products</strong>:This table provides detailed attributes about each mobile device sold such as brand, model, storage size, color, operating system, and retail price.</li><li><strong>Dim location</strong>: This table adds geographical context, indicating the country and city where each transaction occurred.</li><li><strong>Dictionary</strong>: This table serves as a lookup table, decoding numerical IDs in the fact table into human-readable sales channels (e.g., online, in-store) and payment types (e.g., cash, card).</li></ol><p><strong>Data Processing</strong></p><p>With the dataset already cleaned and all relevant attributes merged into a single unified Fact sales table, the data processing phase focused primarily on preparing the dataset for analysis and visualization in Excel. This comprehensive table included transactional records enriched with product details (such as brand, model, operating system, storage size, and color), customer demographics (age group and gender), geographic information (city and country), and sales logistics (channel, payment type, quantity sold, and revenue).</p><p>I began by loading the dataset into Excel and inspecting the columns to understand the data types and available dimensions. Since all necessary information was contained in one consolidated table, there was no need to create relationships or perform lookups. This simplified the workflow significantly, allowing me to focus on building <strong>PivotTables</strong> directly from the Fact sales sheet.</p><p>To enable flexible analysis, I created several PivotTables that sliced the data by various factors such as brand, storage size, location, customer age group, and sales channel. These PivotTables formed the foundation for the dashboard and allowed for quick aggregation of sales volume, revenue, and customer trends across different categories.</p><p><strong>Analysis and Insights</strong></p><p>With the data properly structured and loaded into Excel, I conducted a series of analyses to uncover trends in mobile phone sales across various dimensions such as product features, geography, customer demographics, and sales channels. The analysis was performed using functions, Pivot tables.</p><p>To provide meaningful insights and track the performance of mobile phone sales throughout 2024, I defined a set of Key Performance Indicators (KPIs). These KPIs helped summarize complex transactional data into digestible business metrics, allowing stakeholders to make informed decisions. All KPIs were derived using PivotTables and visualized in Excel charts as part of the dashboard.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1VPlmEW9VKGlUdixL95vcg.png" /></figure><h3>Starting with the Basics: Revenue and Units Sold</h3><p>The first KPIs I looked at were <strong>Total Revenue ($13,360,389)</strong> and <strong>Total Units Sold (17,173)</strong>.</p><p>When I saw these numbers, my first thought was: <em>“Okay, revenue looks strong, but how much are we making per unit?”</em> That naturally led me to the next calculation.</p><p>Calculating the Average Revenue per Unit By dividing revenue by total units, I got the <strong>Average Revenue Per Unit: $777.99</strong>.</p><p>This was an important moment because it told me about the company’s pricing position. An average selling price of nearly $778 suggested that these weren’t budget phones — this company was clearly competing in the <strong>mid-to-premium smartphone market</strong>.</p><p>At this point, I started thinking: <em>“We’re not just selling a lot, we’re selling value at a strong price point.”</em></p><h3>Exploring Customer Demographics</h3><p>Next, I moved on to customer profiles. The dataset had:</p><ul><li>Female customers: 170</li><li>Male customers: 134</li><li>Other gender: 31</li></ul><p>When I first compared this to the total units sold, something felt off. With over 17,000 units sold but only 335 customers recorded by gender, I realized this was either a sample or that many customers bought multiple phones.</p><p>That insight made me pause — it reminded me that numbers don’t always tell the whole story, and data quality matters just as much as quantity.</p><p>Still, the gender distribution was valuable. It showed a fairly balanced customer base, slightly higher for female buyers, with representation of “Other Gender,” which signals inclusivity in the dataset.</p><p>In addition to PivotTables, I utilized a variety of visual tools within Microsoft Excel such as bar charts, line charts, and pie charts to effectively communicate key findings and support data-driven decision-making. I also made extensive use of slicers to dynamically filter the data by specific months and cities, enabling a more interactive and targeted analysis. These visualizations were instrumental in highlighting trends, comparing performance across regions, and gaining deeper insights into customer behavior and sales dynamics.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/431/1*NYzzChcXlfiGD_ccWhRLzg.png" /><figcaption>Top Selling Brands By Unit</figcaption></figure><ul><li>Apple leads the pack with 76 units, showing its dominance in customer preference, OnePlus follows closely at 70 units, proving its growing popularity and ability to challenge more established players, Xiaomi (65), Samsung (63), and Google (61) round out the top five, showing relatively close competition in this range.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/306/1*hEXZG7lgxXS5N6mqpp-J8w.png" /><figcaption>Sales By Storage Size By Unit Sold</figcaption></figure><p>When I looked at the sales distribution by storage size, I wanted to see if customers preferred more affordable lower storage options or leaned towards higher storage premium models.</p><p>Here’s what the data revealed:</p><ul><li>64GB models led the sales with 118 units sold.</li><li>256GB models followed closely at 110 units sold.</li><li>128GB models came last with 107 units sold.</li></ul><p><strong>Insight:</strong> The differences between the three storage sizes are relatively small, but the preference for 64GB devices suggests many customers are still price-sensitive, choosing affordability over extra storage. However, the fact that 256GB models sold almost as much as 64GB shows that there’s also a strong demand for premium, high-storage devices. Interestingly, the 128GB option underperformed, possibly because it sits in an awkward middle ground — not as affordable as 64GB, and not as appealing as 256GB.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/360/1*_yHWUpuO9v0JJ2f7wxmQtQ.png" /><figcaption>Sales By Channels</figcaption></figure><p>To understand where most of the sales came from, I analyzed the distribution of sales channels. The pie chart revealed a clear story:</p><ul><li>Online sales dominate with 62% of total sales.</li><li>Retail stores account for 26%, showing that physical outlets still play a significant role.</li><li>Partner channels contribute 12%<strong>,</strong> the smallest share among the three.</li></ul><p><strong>Insight</strong>: The dominance of online sales highlights how much customers are shifting towards digital-first purchasing behavior. Convenience, wider variety, and competitive pricing could be driving this trend. At the same time, retail stores remain important with more than a quarter of sales still happening offline, they continue to serve customers who want a hands-on experience before buying. Partner channels, though smaller, may represent an opportunity to grow through collaborations and third-party vendors.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/688/1*7cmt5fQoZVmkByHrsV5GFA.png" /><figcaption>Mobile Brands Sales By Age And Gender</figcaption></figure><h3>Analysis</h3><p>When I broke down mobile brand sales by age groups, the insights became much clearer about which brands resonate with different generations:</p><ul><li>Ages 18–25: OnePlus dominates among the youngest buyers. This suggests that its pricing and performance appeal strongly to students and young professionals who want premium features at competitive prices.</li><li>Ages 26–33: Again, OnePlus takes the lead, proving it’s not just a youth brand but also popular with early-career adults.</li><li>Ages 34–41: Xiaomi rises to the top. This group may value affordability and functionality, and Xiaomi fits that profile well.</li><li>Ages 42–49: Samsung leads in this range, reflecting its long-standing reputation and brand trust, which resonates with more established professionals.</li><li>Ages 50–57: Apple dominates here, showing that older buyers with stronger purchasing power lean toward premium, brand-recognized products.</li><li>Ages 58–65: Surprisingly, OnePlus re-emerges as the leader, suggesting that even older buyers are open to trying newer brands that offer good value.</li></ul><p><strong>Insight</strong></p><p>This analysis shows that brand preference shifts significantly across age groups.</p><ul><li>OnePlus is clearly the breakout star, appealing strongly to younger and even older demographics.</li><li>Xiaomi and Samsung capture the middle-age market, focusing on balance and reliability.</li><li>Apple remains a status symbol, especially for mature customers with higher disposable income.</li></ul><p>This pattern suggests that companies should adopt age-targeted marketing strategies:</p><ul><li>Highlight affordability and features for younger buyers.</li><li>Emphasize reliability and trust for middle-aged buyers.</li><li>Position as a premium lifestyle brand for older, wealthier customers.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/350/1*2WgB_CC4Z0-ds_VL7rGeRg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/358/1*NIv7BXkNpD1zBUEOMhDUxg.png" /><figcaption>Total Revenue By Cities</figcaption></figure><h3>Analysis</h3><p>This chart breaks down total revenue across different cities, with a slicer to filter results dynamically. The <strong>Top 5 cities</strong> contribute the largest share of overall revenue, standing out as the company’s strongest markets.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/689/1*BOZhB7XFo_EjBgq1ho-HmQ.png" /><figcaption>Monthly Revenue Trends</figcaption></figure><p>When I first looked at this chart, the green line alone gave me a sense of ups and downs across the months. But what really brought the story to life was the horizontal benchmark line at 1 million.</p><p>This line acted as a reference point, showing me instantly whether revenue for a given month was above or below the 1M mark. I found that for most of the year, sales comfortably stayed above it especially in January, July, and December. These were the strong months where the business clearly performed well.</p><p>However, what caught my eye were the months that dipped below the 1M line particularly April and September. This wasn’t just a small decline; it was a red flag that revenue fell short of what I would consider a healthy baseline for the business. To me, this highlighted two things:</p><ol><li>The business is capable of performing well, but struggles with consistency.</li><li>There may be external or internal factors (like demand fluctuations, weak promotions, or competition) dragging performance below target in these months.</li><li>By placing that 1M benchmark, I wasn’t just looking at revenue trends — I was now able to separate strong months from underperforming ones. That insight is powerful because it gives the business a clear target: keep monthly sales above 1M, and investigate what’s causing the dips when it doesn’t happen.</li></ol><p>My insight: The horizontal line turned this from just a trend chart into a performance evaluation tool. It helped me frame the story of not just “what happened,” but also “when performance slipped below expectations.”</p><p><strong>THE DASHBOARD</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6TCfHvU3heCzmzfLnWR_KA.png" /></figure><p><strong>RECOMMENDATION</strong></p><p>Based on my analysis, here are the key recommendations I would make to the business:</p><ol><li><strong>Stabilize Monthly Revenue Around the 1M Benchmark</strong></li></ol><ul><li>The trend chart revealed that revenue fell below 1M in certain months (especially April and September). The business should strengthen promotional activities, run product launches, or introduce targeted discounts in these weaker months to maintain consistency.</li></ul><p><strong>2. Capitalize on High-Demand Periods</strong></p><ul><li>July and December stood out as strong revenue months. The business should double down on marketing campaigns, product bundling, and upselling strategies during these periods to maximize returns.</li></ul><p><strong>3. Leverage Online Sales Strength</strong></p><ul><li>Since 62% of sales came from online channels, investing in digital marketing, optimizing the online shopping experience, and offering exclusive online deals could further boost performance. Retail and partner sales can be improved by aligning promotions with the success of online strategies.</li></ul><p><strong>4. Tailor Products to Customer Preferences</strong></p><ul><li>Storage size analysis showed 64GB models as the most sold, closely followed by 256GB and 128GB. This indicates demand across both budget-conscious and premium buyers. Stocking a balanced inventory of entry-level and high-storage phones will help meet diverse customer needs.</li></ul><p><strong>5. Segment Marketing by Age &amp; Gender</strong></p><ul><li>Different age groups showed loyalty to different brands (e.g., OnePlus dominated among younger buyers, while Apple led among older customers). Targeted campaigns based on age and brand preference will likely increase conversion rates.</li></ul><p><strong>6. Focus on High-Revenue Cities</strong></p><p>Since the top cities generated the bulk of total revenue, these should be prioritized for localized campaigns, customer retention programs, and better after-sales services. At the same time, cities with lower revenue can be explored with growth strategies such as influencer partnerships or retail expansions.</p><p>— — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — — —</p><p>This is my first data analytics project. Thank you for taking your time to go through this! Kindly give a clap for wider reach and connect with me on my socials for more content on analytics.</p><p><a href="https://www.linkedin.com/in/aishat-ayoola-687244354/">Linkedin</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=316274ff1f81" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>