<?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 Blessing Ekwere on Medium]]></title>
        <description><![CDATA[Stories by Blessing Ekwere on Medium]]></description>
        <link>https://medium.com/@blessingekwere?source=rss-c1a243eecad1------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*P6oQo5TMknm-wWn1T-PZPQ.jpeg</url>
            <title>Stories by Blessing Ekwere on Medium</title>
            <link>https://medium.com/@blessingekwere?source=rss-c1a243eecad1------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Fri, 20 Nov 2026 10:18:59 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@blessingekwere/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[Understanding Docker: A Tool Every Data Engineer Needs]]></title>
            <link>https://medium.com/@blessingekwere/understanding-docker-a-tool-every-data-engineer-needs-afba7862cfcd?source=rss-c1a243eecad1------2</link>
            <guid isPermaLink="false">https://medium.com/p/afba7862cfcd</guid>
            <dc:creator><![CDATA[Blessing Ekwere]]></dc:creator>
            <pubDate>Mon, 29 Sep 2025 17:47:25 GMT</pubDate>
            <atom:updated>2025-10-01T11:40:30.753Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/686/1*XtgvXc9RFo80pnQpausbtQ.jpeg" /><figcaption>Source: Google</figcaption></figure><p>Consider this article an introductory article on docker with the goal of helping professionals understand Docker and why it has become such an essential tool in the world of data engineering.</p><p>When I worked as a data analyst, my world revolved around spreadsheets, relational databases, business intelligence tools and the ever-versatile Python.<br>I was comfortable there, cleaning and transforming data, building reports, telling stories with numbers, and then repeating the process all over again.</p><p>But everything changed when I transitioned into data engineering.<br>The data engineering landscape suddenly looked so different. Conversations were filled with new terms like pipelines, orchestration, containers, deployment amongst others.</p><p>And when it came to tools, things became even more overwhelming and interesting at the same time. The ones I knew were just SQL and Python, but now everyone was talking about Cloud, Airflow, dbt, Airbyte, Docker, Kubernetes, Apache Spark, and a few others I had never even heard of.</p><p>Then one day, I stumbled upon a YouTube video about Docker and why it was a must-know skill for data engineers.<br>At first, I didn’t take it seriously, until I noticed something interesting:<br>Almost every data engineering job post I saw mentioned Docker as a key requirement.</p><p>That was my wake-up call. If I wanted to build real systems that help businesses use data to solve problems, I needed to understand Docker and how it works and not just the commands, but the <strong><em>why</em></strong> behind it.<br>So I decided to start again from scratch, this time determined to truly understand how Docker fits into data engineering.</p><h3><strong>What exactly is docker?</strong></h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/651/1*LqfzQvJ2_y0L7_l3IojjBA.jpeg" /><figcaption>Source: Google</figcaption></figure><p>It is often said that “<em>when the purpose of a thing is not known, abuse is inevitable”.</em></p><p>As data professionals, understanding the purpose behind every action matters. A single mistake in a data pipeline can break an entire system. That is why I have made it a rule to question everything I do until I understand it clearly. Learning Docker was no different. I didn’t want to just memorize commands, I wanted to know <strong><em>why</em></strong> I was learning it.</p><p>When I first searched for a definition, I found the interesting line:</p><blockquote>“Docker is an open-source platform that lets you build, run, and deploy applications inside containers.”</blockquote><p>It sounded fancy, but it didn’t make much sense to me at the time. The first thing that confused me was the word <em>applications</em>. I remember thinking: <em>“Wait, are we building applications as data engineers? What does that even mean?”. </em>Funny how<em> </em>I stayed stuck on that question for a while (now I want to laugh <em>at myself</em>).</p><p>Then it finally clicked. Yes, we build applications, data pipelines are applications. They are systems that process, transform, and move data, just like software applications process user requests or display web pages.</p><p>And the “open-source” part? That simply means Docker is free for anyone to use, study and improve. That’s part of what makes it so powerful and widely adopted.</p><h3>What exactly is docker?- Let’s break it down further</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eKgfoww-wrC8YYeJ1Z4xaA.png" /><figcaption>Source: Google</figcaption></figure><p>Because my goal is to make Docker easy to understand, let’s break it down with real-life examples.</p><p><strong>Case 1: Architecture Analogy</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/612/1*qXtYsPAdmWaYXnGL8gtNAA.jpeg" /><figcaption>Source: Google</figcaption></figure><p>Imagine an architect designing a building. Every room serves a specific purpose one might be a meeting room, another for storage, another for relaxation. Each space is carefully designed with the right materials, furniture, and fittings to function properly.</p><p><em>Docker works in a similar way. It holds everything an application needs to run the code, dependencies, libraries, and configurations just like each room in a building contains all the elements required for its purpose.</em></p><p>Now picture this: a client wants the exact same design in a new location. Without Docker, the architect would have to start from scratch, sourcing new materials, rebuilding each room, and hoping everything fits together as it did before. Or imagine the current owner moves and wants their setup replicated in a new apartment. Again, without Docker, everything has to be rebuilt from the ground up.</p><p><em>With Docker, it’s like the architect has designed modular rooms that can be packed up neatly, moved anywhere, and reassembled exactly as they were. Nothing breaks, nothing changes and it just works.</em></p><p><strong>Case 2: Interior Decoration Analogy</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/650/1*A2zzRKZXGGuQ47M7tSc9IA.jpeg" /><figcaption>Source: Google</figcaption></figure><p>For a simpler example, think about interior decoration. Imagine you’ve decorated your living room, you have chosen paint colors, furniture, curtains, and lighting to create a specific mood. Now, you want to replicate that exact setup in a new apartment. Without Docker, you’d have to start from scratch, hoping the furniture matches and the paint shade looks the same.</p><p>With Docker, it’s like packaging your entire room with every single item and arrangement and recreating it perfectly in a new space with just a single click.</p><p><em>That is what Docker does for applications. It packages your entire environment, not just your code, but everything it depends on, so you can move it, share it, and run it anywhere without worrying about compatibility or setup issues</em>.</p><p>When I finally saw Docker this way, it clicked. It wasn’t some developer tool anymore; it was simply a way to organize and move systems efficiently, just like how architects design structures or decorators plan repeatable designs.</p><p><strong>Case 3: The Kitchen and Cooking Analogy</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ALcCW_Jrf07amUpw89b3QQ.jpeg" /><figcaption>Source: Google</figcaption></figure><p>Here’s another way to think about Docker, using something almost everyone has experienced which is cooking. Imagine you are a chef in a busy kitchen. To prepare a dish perfectly every time, you need the right ingredients, utensils, and appliances, all arranged in a way that makes cooking efficient.</p><p>Without Docker, cooking in a new kitchen can be stressful. Maybe the stove is different, the pots aren’t the ones you’re used to, or the pantry doesn’t have the ingredients you need. Every time you move to a new kitchen, you spend extra time figuring out where everything is, adjusting recipes, and hoping your dish turns out the same as before.</p><p>Docker solves this problem by giving you a portable kitchen. Every container is like a fully stocked, organized kitchen with all the exact ingredients, utensils, and appliances you need. You can move this kitchen anywhere, another house, another city, or a colleague’s workspace and start cooking immediately, knowing your dishes will taste the same every time.</p><p>Just like with our previous analogies of architecture and interior decoration, this kitchen analogy shows the power of Docker: it allows you to package everything needed to run your “application” (or dish) in a way that is portable, repeatable, and reliable.</p><h3>Docker Components</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tsS0AwA5TLbfAk0XXb5kyw.png" /><figcaption>Source: Google</figcaption></figure><p>Docker is a platform, which means it has different components that make it work properly. Let’s look at the main components of Docker using the three analogies we discussed earlier — architecture, interior decoration, and kitchen. These building blocks help you see how Docker works behind the scenes.</p><ol><li><strong>Docker Image:</strong></li></ol><blockquote>A Docker image is like a blueprint or recipe. It contains everything your application needs to run be it your code, libraries or dependencies.</blockquote><p>. In the architecture analogy, a Docker image is the building plan for a room.</p><p>. In the interior decoration analogy, it’s the design plan for a room layout.</p><p>. In the kitchen analogy, it is the recipe showing exactly what ingredients and tools are needed.</p><p><em>Note that a Docker image is static, it doesn’t run by itself, but you can use it over and over to create identical environments wherever you need them.</em></p><p><strong>2. Dockerfile:</strong></p><blockquote>A Dockerfile is the instruction manual for creating a Docker image. It’s a simple text file with step-by-step instructions.</blockquote><p>. In the architecture analogy, a Dockerfile is the construction guide, showing how to assemble the room from the blueprint.</p><p>. In the interior decoration analogy, it’s the installation guide explaining where each piece of furniture goes.</p><p>. In the kitchen analogy, it’s the step-by-step instruction for preparing the ingredients and cooking the dish exactly as intended.</p><p>With a Dockerfile, anyone can build the same image reliably, no matter where they are.</p><p><strong>3. Docker Hub</strong></p><blockquote>Docker Hub is like a library or marketplace of pre-built images. You can find images created by others, share your own, or use them as a starting point.</blockquote><p>. In the architecture analogy, Docker Hub is a store of building plans you can buy, modify, or use directly.</p><p>. In the interior decoration analogy, it’s a catalog of room designs ready to implement.</p><p>. In the kitchen analogy, it’s a collection of recipes you can follow or adapt.</p><p>Docker Hub makes sharing and collaborating much easier and saves time. You can explore it here: <a href="https://hub.docker.com/">Docker Hub</a>.</p><p><strong>4. Docker Container</strong></p><blockquote>A Docker container is a running instance of a Docker image. It’s where the application actually executes.</blockquote><p>. In the architecture analogy, it’s the fully constructed room you can walk into and use.</p><p>. In the interior decoration analogy, it’s the completed, furnished room ready for living.</p><p>. In the kitchen analogy, it’s the kitchen fully stocked and ready to cook.</p><p>You can create, start, stop, and delete containers, giving you a flexible and isolated environment to work in.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/670/1*5uu8My3CX5gt0YgeiGXvvg.png" /><figcaption>Source: Google</figcaption></figure><p><strong>5.</strong> <strong>Docker Volume:</strong></p><blockquote>A Docker volume is persistent storage for containers. It allows data to survive even if the container is removed or rebuilt.</blockquote><p>. In the architecture analogy, it’s a storage closet that keeps supplies safe even if the room is rebuilt.</p><p>. In the interior decoration analogy, it’s a closet or storage box with furniture or décor items that can be reused in a new setup.</p><p>. In the kitchen analogy, it’s a pantry that keeps ingredients safe no matter where you move your kitchen.</p><p>Volumes ensure your data isn’t lost when containers are recreated or moved.</p><h3>Why Use Docker? (The Real-World Pros and Cons)</h3><p>Now that we know that docker isn’t just hype, that it solves real problems that data engineers and developers face every day. But like every tool, it comes with tradeoffs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*bk1_fGHmT49WAyWUfmQuJw.png" /><figcaption>Source: Google</figcaption></figure><h3>Why you should always consider Docker</h3><p><strong>1. Consistency Everywhere: </strong>Ever had a script that runs perfectly on your laptop but fails on someone else’s machine? Docker fixes that. It packages your code, libraries, and dependencies into a single environment that behaves exactly the same anywhere whether it’s a laptop, server or cloud.</p><p><strong>2. Easy Collaboration: </strong>Teams can share images and spin up identical setups with one command. No more “it works on my system” drama.</p><p><strong>3. Portability: </strong>Move your containers across environments like you’re moving boxes from development to production, from local to cloud without breaking anything.</p><p><strong>4. Scalability: </strong>Running multiple containers side by side lets you scale systems fast. Perfect for big data pipelines or distributed workloads.</p><p><strong>5. Saves Time: </strong>Docker helps you save time because when you setup once, you can reuse forever. No need to reinstall dependencies or reconfigure environments every time.</p><h3>What You Should be Concerned about</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yPVbDkfXjeQewm2rFmxrVQ.png" /><figcaption>Source: Google</figcaption></figure><p>Now we know how amazing a problem-solving tool docker is,below are some areas that we should be worried about with docker.</p><p><strong>1. Learning Curve: </strong>Docker can be confusing at first . Understanding images, containers, Dockerfiles and networks takes a little time.</p><p><strong>2. Storage Usage: </strong>Containers and images can eat up disk space quickly if you don’t clean them up regularly.</p><p><strong>3. Performance Overhead: </strong>For heavy workloads, containers can be slightly slower than running directly on hardware though this might rarely be a dealbreaker.</p><p><strong>4. Debugging Can Be Tricky: </strong>When something breaks inside a container, debugging it isn’t always straightforward.</p><p><em>In my opinion docker actually makes your projects cleaner, faster, and more reliable especially when you’re building data systems that move from one environment to another. Once you get the hang of it, you’ll wonder how you ever built without it.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Qm9PGCZWyS6aFGfk_ioSOQ.png" /><figcaption>Source: Google</figcaption></figure><h3>Now let’s wrap it up</h3><p>When I finally understood Docker, it stopped feeling like a developer-only tool. It became something I could totally relate<em> </em>to, a system that keeps my work consistent, portable and reliable.</p><p>As data engineers, we build systems that need to work everywhere whether it is in development, testing or production environment. Docker makes that possible. It gives us the confidence that what runs on our local machine will run the same way in the cloud, without the usual “it works on my laptop” drama.</p><p>So the next time you hear the word <em>“Docker container”</em> or any other Docker concept, don’t treat it like another industry buzzword. Think of it as your modular building, your perfectly styled living room or your portable kitchen, designed to perform exactly the same way no matter where you take it.</p><p>Of course, there is so much more to Docker than what I have shared here, from orchestration, networking, multi-container setups and even tools like Docker Compose and Kubernetes that build on top of it. But this foundation is where it all starts.</p><p>If you found this article useful or you are also on your data engineering journey, let’s connect on <a href="https://www.linkedin.com/in/blessing-ekwere-857326216"><strong>LinkedIn</strong></a> and <a href="https://twitter.com/Eddie_Gregs?t=dF3996shVxvPJTePTtxDdw&amp;s=09"><strong>Twitter (X)</strong></a>…. I always do my best to share more around the things I am learning in this career path.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=afba7862cfcd" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Learning Data Engineering: What I Got Right, What I Got Wrong and What No One Talks About]]></title>
            <link>https://medium.com/@blessingekwere/learning-data-engineering-what-i-got-right-what-i-got-wrong-and-what-no-one-talks-about-8eb9a2632a7a?source=rss-c1a243eecad1------2</link>
            <guid isPermaLink="false">https://medium.com/p/8eb9a2632a7a</guid>
            <dc:creator><![CDATA[Blessing Ekwere]]></dc:creator>
            <pubDate>Sat, 12 Jul 2025 22:58:36 GMT</pubDate>
            <atom:updated>2025-07-13T13:17:42.399Z</atom:updated>
            <content:encoded><![CDATA[<h3>Learning Data Engineering: What I Got Right &amp; What I Got Wrong</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*UORux1EAVNf87qxs" /><figcaption>Photo by <a href="https://unsplash.com/@jeffgry?utm_source=medium&amp;utm_medium=referral">Oğuzhan Akdoğan</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><blockquote>“I AM A DATA ENGINEER!” — The Man, The Myth, The Motivation</blockquote><p>Thinking of writing this article randomly reminded me of a guy I used to work with. He would start his day loudly and proudly screaming:</p><blockquote><strong><em>“I AM A DATA ENGINEER!”</em></strong></blockquote><p>He would say it in meetings, drop it in Slack, and sprinkle it into conversations at different intervals. I half expected him to wear it as a badge on his chest. I was no hater, I was only curious why he was always acting that way.</p><p>At the time, I had just gotten to know about data and how data can be utilized. I was not even a data analyst yet. His screaming was part of what sparked my curiosity and even though not top on the list, I would say it was one of those contributory factors that eventually helped me land in this career path.</p><p>Honestly, I used to roll my eyes whenever he would scream that and think:</p><blockquote>What is the hype? Other people work with data. What exactly is he screaming about?</blockquote><p>I did not get it then.</p><p>But today, after spending months learning data engineering while working full-time, building projects without real infrastructure, doubting myself, restarting, crashing, Googling, retrying, and crashing again , I can now say this boldly:</p><blockquote>If I ever see that guy again, I will buy him a microphone and a speaker. I will say, “Please. Scream louder. You have earned it.” <em>Because now… I get it.</em></blockquote><h3>From Cleaning Data to Asking “Who Built the systems to enable me access the data seamlessly?”</h3><p>I have worked as a data analyst across multiple industries and I know how to clean messy data, access data stored in different locations, build performance dashboards and extract business insights using tools like Python, SQL, Excel, Power BI and more...</p><p>But somewhere along the line, I started asking a question I could no longer ignore:</p><blockquote>Where does this data even come from?</blockquote><p>Not just the “sales table” or the “customer database”. I mean the <strong>pipeline where data flows or moves from one location to another</strong>. Who built it? Who moves and cleans the data? Who ensures that it is accurate and clean to an extent before it lands where I can pick it and use it for my analysis?</p><p>That was the moment I realized I did not just want to analyze data anymore. I wanted more, I wanted to build the systems that deliver it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*_d38dl5DJ1n1rew6" /><figcaption>Photo by <a href="https://unsplash.com/@markuswinkler?utm_source=medium&amp;utm_medium=referral">Markus Winkler</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>Who Does What? Understanding the Roles</h3><p>Before moving further, I needed to clearly understand the differences between the key players in the data space because this was going to help me decide for the one hundredth time if I needed to pursue a career in data engineering.</p><h3>Data Analyst</h3><p>This is usually where some people start, including me. Data analysts work directly with data to find insights, build reports and dashboards, and support decision-making. They clean data, analyze trends, and communicate stories with visualizations and metrics to stakeholders.</p><h3>Data Engineer</h3><p>These are the builders behind the scenes. Data engineers create, maintain, and optimize the pipelines and infrastructure that collect, store, and process data. They focus on reliability, scalability, and automation. Their goal is to make sure data gets from source to destination in a clean, usable, desired shape and form and efficiently.</p><h3>Analytics Engineer</h3><p>This is a bridge role that sits between data engineering and analytics. Analytics engineers focus on modeling raw data into well-structured datasets that analysts and business users can easily understand and use. They work primarily with SQL and transformation tools, and they clean and organize data just before it lands in dashboards.</p><p>Understanding these distinctions helped me figure out where I was, where I wanted to go, and what I needed to learn.</p><h3>The Mental Shift I Was Not Ready For</h3><p>You know what nobody tells you? Or maybe someone did, but you did not take it as seriously as you should have :</p><blockquote><strong><em>Data engineering is not just about tools.</em></strong></blockquote><p>As an analyst, you ask questions like:</p><ul><li>What happened last month?</li><li>Why are customers dropping off in Region B?</li></ul><p>As a data engineer, you ask questions like:</p><ul><li>Where is this data coming from?</li><li>How do I automate this to run daily without human interference?</li><li>What happens if it fails at 3:27 AM on a Sunday?</li><li>What if the pipeline fails on my wedding day when I am about to say my vows?</li></ul><p>Analytics is more concerned about the insights and what the data is saying. <strong>Engineering is about structure, availability, scalability and stability.</strong></p><p>It took me weeks to stop thinking like an analyst and start thinking like a builder. And eventually when I started thinking like an engineer, I often thought of how to incorporate the data engineering skills I was acquiring in my day-to-day job as a data analyst because I did not just want to learn I needed to flex my muscles too.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*bpTEXoQwfS7iBnxM" /><figcaption>Photo by <a href="https://unsplash.com/@sebastiancoffee?utm_source=medium&amp;utm_medium=referral">Sebastian Willius</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3>Let Us Talk About Tools (And Cost)</h3><p>Here is another thing I never got warned about before leaving the safe zone of data analytics to transition in data engineering:</p><blockquote><strong><em>Learning data engineering requires more tools and most of them are not free.</em></strong></blockquote><p>In analytics, I could complete end-to-end projects with just Excel and Power BI. But in data engineering, I needed:</p><ul><li>A database/data warehouse (PostgreSQL, MySQL, or Snowflake)</li><li>Orchestration (Airflow, Prefect)</li><li>Containers (Docker)</li><li>Git and version control</li><li>Cloud storage (AWS S3, GCP, Azure Blob)</li><li>Possibly dbt, Terraform, CI/CD tools, monitoring, and logging systems</li></ul><p>Most of these tools come with:</p><ul><li>Complex setups</li><li>Usage limits</li><li>Some level of cost (eventually)</li></ul><p>You do not just “install and run.” You configure, containerize, connect, deploy… then debug for hours.</p><p>There were days I spent four hours trying to get Airflow to recognize a simple Python script just to move one column from one file to another.</p><p>And that is <strong>before</strong> my laptop starts begging for mercy under all the load.</p><h3>Learning Without a Real Environment = Fake It Till You Do Not Break It</h3><p>Unlike data analysts, who can practice on almost any open dataset, data engineers need <strong>infrastructure</strong>.</p><p>I did not have a dev team. I did not work in a cloud-heavy company. I had no access to enterprise pipelines. I was even worried about opening account with platforms that had free trials because I was worried that I may not utilize the free trial within the window period given.</p><p>So I had to find a way to get things done bit-by-bit. Eventually I…..</p><ul><li>Ingested open CSVs files into a local PostgreSQL database</li><li>Wrote Python scripts to mimic ETL pipelines</li><li>Scheduled tasks using Airflow on my laptop</li><li>Googled <em>“How to make it feel like you are a data engineer without being one?” </em><strong><em>Yes, I once googled this.</em></strong></li></ul><p>They were not big projects like I expected but they taught me:</p><ul><li>How data actually flows</li><li>What orchestration really means</li><li>That projects can be small, scrappy and still count</li></ul><p>Each mini project became a proof-of-concept and a measure of my growth and how far I have come, not for recruiters, but for myself.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*tZ3uvvA9ZZws93Qz2rbdUw.jpeg" /></figure><h3>The Days I Almost Gave Up (And Nearly Settled for Analytics Engineering)</h3><p>There were nights when:</p><ul><li>Airflow would not start</li><li>Docker would crash</li><li>PostgreSQL would not connect</li><li>And logs looked like gibberish from another planet or even a language that I didn’t understand.</li><li>Learning data engineering on a windows operating system stressed me and still stresses me especially during installations.</li><li>Days when I would start a project and then imposter syndrome would creep in.</li></ul><p>I would sit back and say:</p><blockquote><em>“Maybe this is not for me.”<br> “Maybe analytics engineering is enough.”</em></blockquote><p>And honestly? It was tempting.</p><p>Analytics engineering felt familiar because it is :</p><ul><li>Still SQL-based</li><li>Still transforming data</li><li>Less infrastructure, fewer crashes</li></ul><p>For a moment, I almost settled.</p><p>Until I realized… I was not quitting because I did not like data engineering or because I loved analytics engineering more.<br> I was quitting because I was scared I thought that I was not smart enough to do it.</p><p>But fear is never a good reason to stay small.</p><p>So I continued slowly, but forward and I am still moving.</p><h3>What I Got Right (By Guidance, Accident or Instinct)</h3><ul><li><strong>I focused on concepts before tools</strong><br> I did not chase “how to use Airflow” blindly. I wanted to understand <em>why orchestration matters</em>. On this I read a lot of books and articles.</li><li><strong>I accepted being a beginner</strong><br> This was tough. Going from “an experienced data analyst” to “confused newbie” is a hit to the ego but I did not care because ego isn’t going to pay the bills.</li><li><strong>I built tiny, ugly projects</strong><br> They were not portfolio-ready. But they worked and they taught me more than valuable lessons that I will not forget in a long time (that’s what debugging does sometimes).</li><li><strong>I stopped learning everything at once</strong><br> I initially wanted to learn Docker, dbt, Airflow, Spark, Terraform, and CI/CD all in six months but I took a break and decided to take it a step at a time.<br> <em>Pro tip: Trying to learn or do everything at once is the fastest way to burn out.</em></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/612/1*__gWh5-bRDa3-aelwXslTg.jpeg" /></figure><h3>What I Got Wrong (That You Can Avoid)</h3><ul><li><strong>Measuring progress by tools, not understanding.</strong> Just because I had not built a massive DAG did not mean I was not growing.</li><li><strong>Waiting for the “perfect” project</strong><br> Perfection is the enemy of progress. Start where you are. Build that small tiny project.</li><li><strong>Comparing myself to other data engineers on LinkedIn forgetting that everyone’s journey is different</strong>. You are not behind. You are just early.</li></ul><h3>The Small Wins That Kept Me Going</h3><ul><li>Explaining orchestration to someone and realizing I actually understood it.</li><li>Watching a DAG run successfully after 4 days of trial and error</li><li>Writing a reusable Python function that felt “clean”</li><li>Understanding basic concepts and principles</li><li>Completing that mini-project on my own with little or no guidance</li></ul><p>Not big wins. But real ones that are worth celebrating.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/306/1*-3U2mhKFysK-CybbKWvN-Q.jpeg" /></figure><h3>And Now… I Understand the Screaming</h3><p>So yes , that guy who used to scream “I AM A DATA ENGINEER”?</p><p>I finally get it.</p><p>Because when your work seems invisible, high-stakes, and deeply complex sometimes you <strong>need</strong> to scream.</p><p>There have been times that I had to clap for myself alone in my room just because my code ran and delivered the desired result after a few hours or days of debugging.</p><blockquote>If I ever meet him again, I will say:<em><br> “You were right all along. Scream louder. I brought the mic, let’s scream together”</em></blockquote><h3>Final Thoughts: Becoming, Not Just Learning</h3><ul><li>I am not an expert just yet, I am still a beginner.</li><li>I still search error messages word for word.</li><li>I still get nervous when a DAG fails or even before I start a new project.</li><li>I still feel like a beginner because I am.</li><li>Imposter syndrome never left but I would not be tired of fighting it.</li></ul><p>But I also know this:</p><blockquote><em>I am not just learning tools. I am building a mindset.<br> I am not just copying code. I am learning to solve problems.<br> I am not behind. I am just becoming.</em></blockquote><p>So if you are also learning data engineering or any other tech skill while juggling a full-time job and other aspects of life, without access to real infrastructure or a feel of what the production environment looks like:</p><ul><li>You are not slow.</li><li>You are not alone.</li><li>You are not failing.</li><li>You are just early.</li></ul><p>Keep building. Keep asking questions. Keep showing up.</p><blockquote><em>Because even the loudest data engineers were once quietly debugging a small ETL pipeline at midnight.</em></blockquote><blockquote><strong><em>“I AM A DATA ENGINEER.” </em></strong><em><br> Go ahead. Say it and scream it. It feels good.</em></blockquote><p>Are you a data engineer or you’re interested in data &amp; analytics engineering then let’s connect on <a href="https://www.linkedin.com/in/blessing-ekwere-857326216/">LinkedIn</a> and <a href="https://x.com/Eddie_Gregs?t=4e9GFqxoxl8cc_PpydIq7Q&amp;s=08">Twitter/X</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8eb9a2632a7a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Data in Building & Construction Industry]]></title>
            <link>https://medium.com/@blessingekwere/data-in-building-construction-industry-6f2f33991c62?source=rss-c1a243eecad1------2</link>
            <guid isPermaLink="false">https://medium.com/p/6f2f33991c62</guid>
            <dc:creator><![CDATA[Blessing Ekwere]]></dc:creator>
            <pubDate>Tue, 16 Apr 2024 21:00:04 GMT</pubDate>
            <atom:updated>2024-04-16T21:09:49.156Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*G7gGSpquBBIhGC5Z" /><figcaption>Photo by <a href="https://unsplash.com/@austindistel?utm_source=medium&amp;utm_medium=referral">Austin Distel</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3><strong>Introduction</strong></h3><p>If you’ve been around the tech ecosystem for a while, you would have probably heard the popular saying “data is everywhere” or if you’re just starting out as a data professional and seeking advice on how to apply the skills you have learnt in real life, people would often advise that you start from the industry you are already in.</p><p>As a beginner in the data career established in industries like construction, it can be hard to see how this applies in the real world. How can you utilize data in an industry where most of the work revolves around building design and construction? Can data really be as valuable on a construction site as it is in a trendy startup?</p><p>The answer is a definite and loud yes. Data is transforming the building industry, providing insights to improve efficiency, reduce costs, and make better decisions at every project stage. Whether you’re an experienced construction professional or someone interested in a data career, this article will explore the exciting ways data is impacting the built environment.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*EudHjfyzJEgd_MYl" /><figcaption>Photo by <a href="https://unsplash.com/@sunburned_surveyor?utm_source=medium&amp;utm_medium=referral">Scott Blake</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3><strong>About the Building Industry</strong></h3><p>Before going further let’s understand what the building industry is about and it’s components.</p><p>The building industry is responsible for creating the physical structures we live, work, and play in. It involves a wide range of activities, from planning and design to construction and maintenance. Here’s a quick look at each of them:</p><p>. Projects: From houses and skyscrapers to roads and bridges, the building industry caters to diverse needs.</p><p>. Key Players: Architects, engineers, project managers, construction workers, surveyors, and various trade specialists all contribute to building projects.</p><p>. Skills Needed: Technical skills like understand building codes, materials, mathematics, construction methods, and potential use of CAD/BIM software are crucial. However, communication, teamwork, and problem-solving are equally important.</p><h3><strong>Building a Career in Construction: It’s not just hammers and nails!</strong></h3><p>The construction industry offers a wide range of career paths, which might seem interesting to both technical and non-technical minds:</p><p>. <strong>Skilled Trades:</strong> Electricians, plumbers, carpenters, and welders are the backbone of construction, specializing in specific installations and tasks.</p><p>. <strong>Site Management:</strong> Project managers, site supervisors, and safety officers oversee projects, ensuring they are completed on time, within budget, and safely.</p><p><strong>. Design and Engineering:</strong> Architects, structural engineers, and civil engineers design, plan and construct buildings, ensuring they are functional, safe and aesthetically pleasing.</p><p><strong>. Quantity Surveyors:</strong> These professionals estimate project costs, manage materials, and ensure projects are financially viable.</p><p><strong>. Emerging Roles: </strong>With the rise of Building Information Modeling (BIM) and other technologies, data analysts and BIM specialists are beginning to be increasingly sought after.</p><p>The building industry offers a dynamic and rewarding career path for individuals who have passion for it or are keen on solving problems in the building industry.</p><p><em>Notice that we have mentioned that problem solving skills is one of the skills needed in the building industry but then in the tech system you need your problem solving skills too. Which means that as a data professional working in this field you’ll excel greatly because of your problem solving skills.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*nBKNBt7Z5WcNbyDa" /><figcaption>Photo by <a href="https://unsplash.com/@kaleidico?utm_source=medium&amp;utm_medium=referral">Kaleidico</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3><strong>How can data be utilized in this Industry?</strong></h3><p>In the building industry, data plays a crucial role in various stages of the project lifecycle, enabling building professionals to make informed decisions and optimize processes. One prominent example is the utilization of the popular Architects’ Data also simply known as the Neufert by Ernst Neufert, which is a reference book for spatial requirements in building design and site planning, used to extract data for spatial analysis of spaces and site analysis.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/735/1*wOVdGzX7cmMIb3W5Ci7lPw.jpeg" /><figcaption>A page screenshot of the data in Neufert Archidata showing space requirement for humans gotten from Google</figcaption></figure><p>Neufert Archidata provides building professionals with access to a lot of data about architectural standards and regulations, allowing them to extract relevant data such as dimensions, building codes, and material specifications. This data can then be analyzed spatially to optimize the layout and design of buildings, ensuring compliance with regulations and maximizing efficiency in space utilization.</p><p>For instance, architects and urban planners can use spatial analysis tools to assess the site characteristics, including topography, sun exposure, and environmental factors, to inform site planning and design decisions. By integrating data from Neufert Archidata into spatial analysis software or architectural design software such as Revit, AutoCAD, SketchUp etc. Professionals can visualize and analyze site / building data spatially, identifying potential challenges and opportunities for design optimization.</p><p>Furthermore, building professionals use data for various other analysis beyond spatial analysis. For example, structural engineers use structural analysis software to simulate the behavior of building structures under different load conditions, ensuring structural integrity and safety.</p><p>The integration of architectural software like Revit Architecture into Power BI, a business intelligence tool, enhances the capabilities of data analysis and visualization in the building industry. Building professionals can extract data from Revit models, such as building components, materials, and spatial relationships, and integrate it into Power BI for comprehensive analysis and reporting.</p><p>For instance, project managers can use Power BI to create dashboards and reports that provide insights into project progress, resource allocation, and budget tracking based on data extracted from Revit models. Architects can analyze design variations and compare performance metrics across different design iterations using Power BI’s interactive visualizations.</p><p>Overall, the integration of architectural software like Revit Architecture into Power BI enables building professionals to harness the power of data for informed decision-making, collaboration, and optimization throughout the building lifecycle. It facilitates seamless data exchange and analysis, ultimately leading to more efficient and sustainable building projects.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*ylKUX4w66n9iJ7K3" /><figcaption>Photo by <a href="https://unsplash.com/@homajob?utm_source=medium&amp;utm_medium=referral">Scott Graham</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><h3><strong>Possible Data Careers in this industry</strong></h3><p>In the building industry, data analytics plays a crucial role in optimizing processes, enhancing decision-making, and improving overall project outcomes. Data analysts can take on various roles within the building industry, utilizing their analytical skills to extract insights from data and drive value across different areas. Here are some possible data analytical roles that data analysts can take on in the building industry:</p><p><strong>1. Construction Project Performance Analyst:</strong> Data analysts can analyze project performance metrics such as cost, schedule adherence, quality, and safety to identify trends, inefficiencies, and areas for improvement. They can develop predictive models to forecast project outcomes and optimize resource allocation for better project management.</p><p><strong>2. Building Energy Analyst:</strong> Data analysts can specialize in analyzing energy consumption data from buildings to optimize energy efficiency and sustainability. They can perform energy audits, analyze building energy usage patterns, and recommend strategies for reducing energy consumption and carbon emissions.</p><p><strong>3. Building Performance Analyst:</strong> Data analysts can analyze building performance data to assess the effectiveness of building systems and components in meeting performance goals. They can conduct building simulations, analyze sensor data, and identify opportunities for optimizing indoor environmental quality, comfort, and occupant satisfaction.</p><p><strong>4. Facility Management Analyst:</strong> Data analysts can support facility managers in optimizing the operation and maintenance of building systems and equipment. They can analyze maintenance records, equipment performance data, and occupant feedback to identify maintenance trends, predict equipment failures, and optimize maintenance schedules.</p><p><strong>5. Real Estate Market Analyst:</strong> Data analysts can analyze real estate market data to provide insights into market trends, property valuations, and investment opportunities. They can analyze property sales data, rental market trends, demographic data, and economic indicators to inform real estate development decisions and investment strategies.</p><p><strong>6. Building Information Modeling (BIM) Analyst:</strong> Data analysts can specialize in analyzing data from Building Information Models (BIM) to support design, construction, and facility management activities. They can extract data from BIM models, perform spatial analysis, clash detection, and model validation to ensure design integrity and optimize building performance.</p><p><strong>7. Urban Planning Analyst:</strong> Data analysts can support urban planners in analyzing spatial data, demographic data, transportation data, and land use data to inform urban planning decisions. They can perform spatial analysis, develop predictive models, and simulate urban development scenarios to optimize land use, transportation infrastructure, and urban sustainability.</p><p>These are just a few examples of the different roles that data analysts can take on in the building industry. With their analytical skills and domain knowledge, data analysts can make significant contributions to improving efficiency, sustainability, and innovation in the built environment.</p><h3>Conclusion</h3><p>In conclusion, the intersection of data analytics and the construction industry presents an exciting opportunity for tech enthusiasts and professionals to leverage their data skills and drive innovation in this sector. By utilizing the power of data analytics, traditional practices can be transformed and processes streamlined to enhance efficiency, sustainability, and overall project outcomes.</p><p>To all tech enthusiasts and professionals, I encourage you to explore the possibilities of applying your data skills in the construction industry. Whether you’re a data scientist, analyst, engineer, or architect, your expertise in data analytics can make a significant impact by optimizing project performance, improving resource management, and driving informed decision-making.</p><p>Furthermore, I urge key stakeholders in the construction industry to embrace a data-centric approach and open their doors to the integration of data analytics into their organizations. By adopting data-driven practices, stakeholders can unlock valuable insights, mitigate risks, and capitalize on opportunities for growth and innovation. Let’s shift the paradigm from opinion-driven decision-making to data-driven decision-making, paving the way for a more efficient, sustainable, and resilient construction industry. Together, we can build a future where data is not just a tool, but a cornerstone of success in construction projects worldwide.</p><p>Thank you for reading. I hope you found this article insightful. Feel free to connect with me on <a href="http://linkedin.com/in/blessing-ekwere-857326216/">LinkedIn</a> and <a href="https://twitter.com/Eddie_Gregs">Twitter</a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6f2f33991c62" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[International Breweries Sales Analysis]]></title>
            <link>https://medium.com/@blessingekwere/international-breweries-sales-analysis-38cac9d9892c?source=rss-c1a243eecad1------2</link>
            <guid isPermaLink="false">https://medium.com/p/38cac9d9892c</guid>
            <dc:creator><![CDATA[Blessing Ekwere]]></dc:creator>
            <pubDate>Sat, 07 Oct 2023 10:18:14 GMT</pubDate>
            <atom:updated>2023-10-07T10:18:14.577Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Introduction</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/1*uyOQ3nx2SmZOc5e79GJexg.jpeg" /><figcaption>Image Source: Google</figcaption></figure><p>Brewery products have a rich history and a global presence, ranking among the world’s most cherished beverages, with billions of hectoliters consumed annually. The global brewery market boasts a substantial worth, and iconic brands play a pivotal role in shaping this industry landscape. These renowned labels have found homes in countries spanning the globe, with their popularity often influenced by regional preferences.</p><p>Beyond mere beverages, brewery products hold a profound cultural significance. They serve as means of celebration, socialization, and relaxation, transcending age and background to unite people worldwide. Brewery products have ingrained themselves deeply within various cultures, becoming integral to traditions and customs.</p><p>For instance, in West Africa, brewery products are synonymous with festivals and celebratory occasions. In Nigeria, these products grace the tables at weddings, funerals, and other significant life events. In Ghana, they are the beverages of choice at social gatherings, including lively parties and outdoor barbecues.</p><p>Furthermore, the brewery industry stands as a formidable economic force. Employing millions globally and generating billions in revenue each year, it has a profound impact. The top brewery brands are typically under the ownership of multinational corporations, resulting in a highly concentrated industry landscape.</p><p>Now, armed with an international brewery dataset spanning the years 2017 to 2019, join us on a captivating journey as we embark on an exploration of this data trove. Our mission: to uncover insights and empower informed decision-making in this dynamic realm.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*59L6bQFFPO9yZrm5" /><figcaption>Photo by <a href="https://unsplash.com/@merittthomas?utm_source=medium&amp;utm_medium=referral">Meritt Thomas</a> on <a href="https://unsplash.com/?utm_source=medium&amp;utm_medium=referral">Unsplash</a></figcaption></figure><p><strong>Aim and Objectives</strong></p><p>This data analysis project will examine sales data for three years (2017–2019) across seven international beer brands in five West African countries: Benin, Ghana, Nigeria, Senegal, and Togo. The goal of this project is to gain a deeper understanding of the international beer market and to identify the factors that drive beer sales in different countries.</p><p>The data set that will be used for this project includes a variety of information, such as the brands of beer that were purchased, the prices at which the beer was sold, and the countries in which the sales took place. This dataset holds the potential to unravel a multitude of intriguing inquiries regarding the global brewery market. Some of the compelling questions it can address include:</p><p>- Which brands command the highest profitability?</p><p>- What is the cumulative revenue attributed to each brand?</p><p>- How do sales patterns evolve for each brand across different time frames?</p><p>- What constitutes the total revenue stemming from brewery product sales in each country?</p><p>The answers to these questions can be used by breweries to make informed decisions about their marketing and sales strategies. For example, if a brewery knows that a particular brand is more popular in a certain country, they can focus their marketing efforts on that country.</p><p>This data analysis project will also be of interest to researchers and policymakers who are interested in the international beer market.</p><p><strong>Skills Demonstrated</strong></p><p>Data cleaning and preparation</p><p>Data visualization</p><p>Data analysis</p><p>Storytelling</p><p><strong>Data Sourcing</strong></p><p>The dataset was gotten from the mail as shared, and it had a total of 1050 rows and 13 columns. The columns in the dataset are described below:</p><p>1. SALES_ID: This is a unique identifier for each sales transaction or record. It is typically a numeric or alphanumeric code assigned to each sale.</p><p>2. SALES_REP: This column represents the name or identifier of the sales representative responsible for the sale. It identifies the individual handling the customer relationship and sale.</p><p>3. EMAILS: The Emails column likely contains the email addresses of either the sales representatives or customers associated with each sale. It can be used for communication and contact purposes.</p><p>4. BRANDS: Brands refers to the specific brands of beverages or products sold in each transaction. The dataset likely includes various beverage brands such as Beta Malt, Budweiser, Castle Lite, Eagle Lager, Grand Malt, Hero, and Trophy.</p><p>5. PLANT_COST: Plant Cost represents the cost incurred by the brewery or manufacturing plant to produce the products being sold. It includes production-related expenses.</p><p>6. UNIT_PRICE: This column indicates the price at which each unit of the product is sold. It represents the selling price per unit.</p><p>7. QUANTITY: Quantity specifies the number of units of the product sold in each transaction. It quantifies the volume of products sold.</p><p>8. COST: Cost represents the total cost associated with the sale, which may include production costs, shipping costs, and other expenses related to the transaction.</p><p>9. PROFIT: The profit column likely calculates the profit generated from each sale. It is typically calculated as the difference between the total revenue (sales) and the total cost.</p><p>10. COUNTRIES: Countries indicates the countries where the sales transactions occurred. The dataset may include sales data from multiple countries, such as Benin, Ghana, Nigeria, Senegal, and Togo.</p><p>11. REGION: Region specifies the geographic regions or areas within the mentioned countries where the sales took place. Common regions mentioned include North Central, North East, North West, South East, South South, and West.</p><p>12. MONTHS: The months column represents the specific month in which each sales transaction occurred. It is used to track sales over time and analyze monthly trends.</p><p>13. YEARS: This indicates the calendar year associated with each sales transaction. It helps in analyzing sales performance and trends on an annual basis.</p><p><strong>Data Transformation</strong></p><p>As mentioned in the email, I had the freedom to select the data analysis tool of my choice, and I opted for Power BI as my tool of choice. The entire analysis and visualization process took place within the Power BI environment.</p><p>Upon downloading the dataset to my local computer, I initiated Power BI Desktop, which is installed on my system. I proceeded by utilizing the ‘Get Data’ command, recognizing that my dataset was stored in an Excel file. Consequently, I selected the ‘from Excel’ option, located the file, and imported it into Power BI Desktop.</p><p>Once the data was loaded into Power BI Desktop, I adhered to the principle that downloaded data should be thoroughly cleansed and prepared for analysis. To accomplish this, I clicked on the ‘Transform’ button within Power BI to access the Power Query Editor for precise data cleaning and transformation.</p><p>Within the Power Query Editor, I carefully reviewed each column in the dataset and observed several discrepancies that required attention. Here are the issues I identified and the respective corrective actions taken:</p><p>1. Duplicated Rows: I identified the presence of duplicate rows within the dataset, which I promptly removed. After this operation, the dataset was refined to 1047 rows and 13 columns.</p><p>2. Formatting Errors: Certain columns exhibited formatting errors that needed correction to ensure smooth analysis.</p><p>3. Capitalization: In specific columns, such as the ‘Brand’ column, the data was in lowercase. I rectified this by standardizing the capitalization.</p><p>4. Incorrect Delimiter: The ‘Region’ column contained concatenated values without proper spacing. To address this, I first split the column based on character count from the far left and subsequently merged the values using a space as the delimiter, ensuring the accurate representation of regions in the column. Additionally, the ‘Year’ column contained erroneous entries, such as ‘2107,’ ‘2108,’ and ‘2109,’ instead of the correct ‘2017,’ ‘2018,’ and ‘2019,’ respectively. These errors, potentially stemming from data entry oversights, were corrected accordingly.</p><p>5. Handling Null Values: Notably, the ‘Month’ column contained null values. Instead of outright removal, I replaced these null values with ‘-99,’ serving as a placeholder. This decision was made to preserve the data’s integrity, especially considering its relevance to business activities. Furthermore, the ‘Year’ column contained incorrect entries, as previously noted. These errors, possibly arising from data entry oversights, were rectified.</p><p>6. Inconsistent Data: I observed that the ‘Region’ column contained values specific to regions in Nigeria, such as ‘North Central,’ ‘North East,’ ‘North West,’ ‘South East,’ ‘South South,’ and ‘West.’ However, these regional labels are not applicable to other African countries, including Benin, Ghana, Senegal, and Togo. Given the international scope of this analysis, focusing on countries, I decided to omit the ‘Region’ column, as it would conflict with the ‘Country’ column, with the latter taking precedence in the context of our analysis.</p><p>After applying all these steps, I was now left with a cleaned and transformed the data and was ready to be used for further analysis and visualization in Power BI.</p><p><strong>Data Analysis and Visualization</strong></p><p>After completing the data cleaning and transformation phase, I reloaded the refined dataset into Power BI Desktop. As my objective was to derive essential Key Performance Indicators (KPIs), I turned to DAX (Data Analysis Expressions) for the task.</p><p>To initiate this process, I established a dedicated table named ‘_Measures’ to serve as the repository for all my calculated measures. Subsequently, I delved into the KPI calculations, focusing on four pivotal metrics: Total Revenue, Total Profit, Total Quantity of Goods Sold, and Gross Profit Margin.</p><p>Following the computation of these KPIs, I transitioned into the analysis of the questions I had initially outlined.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/420/1*6EDCTmiwLEqrnv02cCKCCg.png" /><figcaption>Profitability by Brands</figcaption></figure><p>In terms of brand profitability, Castle Lite emerged as the top performer, generating an impressive $34,606,900. It was closely followed by Budweiser, with a profit of $31,568,500, and Eagle Lager, with $10,224,240. Beta Malt, Grand Malt, Hero, and Trophy also contributed to profitability, yielding $8,754,410, $7,577,340, $6,453,000, and $6,402,950 respectively.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/355/1*p4P0nsK17y_Ig0-D9ojMIQ.png" /><figcaption>Total Revenue generated by Brands</figcaption></figure><p>When examining the total revenue generated by brands, Budweiser took the lead with a total revenue of $9,470,550,000. It was followed closely by Castle Lite, with $8,651,745,000, and Eagle Lager with $4,792,612,500. Hero also made a substantial contribution with $3,845,988,000 in revenue. Trophy, Grand Malt, and Beta Malt followed, generating revenue of $3,841,770,000, $2,822,559,150, and $2,795,158,050 respectively.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/485/1*yuP7wejFH86JzPEma8sgnA.png" /><figcaption>Brand sales by Year</figcaption></figure><p>Shifting our focus to the sales trend by brand over the years, in 2017, Eagle Lager emerged as the top-selling product with a total of 48,459 units sold. It was closely trailed by Beta Malt with 48,042 units sold, and Castle Lite with 47,307 units sold. Budweiser, Grand Malt, Hero, and Trophy each recorded sales of 46,720, 45,902, 41,314, and 39,819 units, respectively. In 2018, Trophy claimed the top spot with 49,390 units sold, followed by Castle Lite with 46,698 units, and Budweiser with 45,526 units sold. Hero, Beta Malt, Eagle Lager, and Grand Malt all achieved sales figures of 45,417, 43,266, 39,910, and 35,202 units, respectively. In 2019, Grand Malt led the pack with 45,185 units sold, followed by Hero with 42,329 units, and Eagle Lager with 39,434 units sold. Trophy followed with 38,850 units sold. Castle Lite, Budweiser, and Beta Malt recorded sales figures of 34,169, 34,028, and 33,755 units, respectively.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/318/1*oPe4shzEYzmyjeN7VJ07TA.png" /><figcaption>Profit generated by Countries</figcaption></figure><p>Shifting our attention to profit generated from sales by country, Senegal secured the top position with a profit of $21,485,190, closely followed by Nigeria, contributing a profit of $21,376,210. Ghana, Benin, and Togo generated profits of $21,013,050, 20,961,970, and $20,751,000 respectively.</p><p><strong>Dashboard</strong></p><p>Below is the dashboard that answers all the questions this analysis intended to answer. Kindly click <a href="https://app.powerbi.com/view?r=eyJrIjoiMmI4ZThiYjMtYjE4MC00MmNkLWIzOTUtZjMyZDI3YjU3ODJkIiwidCI6IjMwMDM2MmQ4LTZkNmItNGY5ZC04YjQwLTUwZjhkNjkxNmVjZCJ9">here</a> to access the interactive dashboard.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*X4I0DlTCxH-XvADKFoD1qw.jpeg" /><figcaption>International Breweries Sales Analysis Dashboard</figcaption></figure><p><strong>Recommendation</strong></p><p>1. Strategic Brand Focus: Based on the profitability analysis, it’s evident that Castle Lite and Budweiser are strong performers. Breweries should consider allocating more resources and marketing efforts to these brands, taking into account their track record of generating higher profits.</p><p>2. Market Expansion: Given the substantial revenue generated by Budweiser, breweries should explore opportunities to expand the presence of this brand in international markets. Conduct market research and develop targeted marketing strategies to tap into new regions and demographics.</p><p>3. Seasonal Marketing: Recognizing the cultural significance of brewery products in West Africa, breweries can benefit from aligning marketing efforts with festivals and celebrations. Developing seasonal marketing campaigns tailored to specific regions and events can boost sales during these periods.</p><p>4. Data Entry Quality Control: To manage errors stemming from data entry, implement rigorous quality control measures. This includes validation checks, data entry training, and double-checking entries for accuracy. Additionally, consider implementing automated data entry systems to reduce human error.</p><p>5. Regular Data Audits: Schedule periodic data audits to identify and rectify discrepancies. This ensures data accuracy and reliability, enhancing the overall quality of analysis and decision-making.</p><p>6. Customer Engagement: Leverage the email data to establish better customer engagement. Implement email marketing campaigns to keep customers informed about new products, promotions, and events. Personalize communication to build stronger customer relationships.</p><p>7. Supply Chain Optimization: Explore ways to optimize the supply chain to reduce production costs, which can positively impact profitability. Streamline processes, negotiate better deals with suppliers, and minimize waste in the production cycle.</p><p>8. Data-Driven Decision-Making: Encourage a data-driven culture within the organization. Provide training to staff on data analysis tools and techniques to empower them to make informed decisions based on data insights.</p><p>9. Periodic Review: Continuously monitor and review sales trends, profitability, and customer feedback. Regularly revisit these recommendations to adapt to changing market dynamics.</p><p>By implementing these recommendations and maintaining a focus on data quality, breweries can enhance their market presence, improve profitability, and establish themselves as leaders in the global brewery industry.</p><p><strong>Conclusion</strong></p><p>In conclusion, the international brewery market is a dynamic and culturally significant industry with a rich history. Our analysis of sales data spanning three years and seven beer brands across five West African countries has provided valuable insights.</p><p>We’ve identified top-performing brands, analyzed revenue, and observed sales trends. Additionally, we’ve highlighted the importance of data quality and error management in data analysis.</p><p>This analysis equips breweries with actionable recommendations to drive profitability, expand their market reach, and improve customer engagement.</p><p>As the global brewery industry continues to evolve, data-driven decision-making remains essential for success. By leveraging insights from this analysis and embracing a data-centric approach, breweries can thrive in this ever-changing landscape.</p><p>Cheers to informed decision-making and a prosperous brewery industry!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=38cac9d9892c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Customer Spending pattern of Adventure Works]]></title>
            <link>https://medium.com/@blessingekwere/customer-spending-pattern-of-adventure-works-65338a2d0e2?source=rss-c1a243eecad1------2</link>
            <guid isPermaLink="false">https://medium.com/p/65338a2d0e2</guid>
            <dc:creator><![CDATA[Blessing Ekwere]]></dc:creator>
            <pubDate>Sun, 16 Apr 2023 16:28:47 GMT</pubDate>
            <atom:updated>2023-04-27T16:52:55.212Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Introduction</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/1*ixVnK6ew-uFEkSNXpWBFsw.jpeg" /><figcaption>Photo Source: Google</figcaption></figure><p><strong>About Adventure Works: </strong>The Adventure Works database is a fictitious company that sells bicycles, bicycle components, and accessories to consumers, retailers, and wholesalers. It contains tables for sales orders, customers, products, employees, and other business entities.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/594/1*-KYfwoZ9l2Wod4ZOXWJspg.jpeg" /><figcaption>Photo Source: Google</figcaption></figure><p><strong>About the Project: </strong>Customer spending pattern analysis is the process of analyzing and understanding how customers spend their money on products and services offered by a business. This analysis can provide valuable insights into consumer behavior, preferences, and purchasing habits, which can help businesses optimize their marketing strategies, improve customer engagement, and increase revenue</p><p><strong>Problem Statement: </strong>Adventure Works is a fictional company that sells bicycles, bicycle components, and accessories to consumers, retailers, and wholesalers. The company is looking to analyze customer spending patterns to better understand customer behavior, preferences, and purchasing habits, in order to improve customer engagement and increase revenue. The company has a vast customer base, and it is becoming increasingly challenging to understand customer behavior and preferences using traditional methods. Adventure Works has a point-of-sale system and an e-commerce platform that captures transactional data on customer purchases.</p><p>The company would like to leverage this data to analyze customer spending patterns and identify trends and patterns in customer behavior. Adventure Works would like to understand the following:</p><p>What products and services are most popular among customers?</p><p>What are the typical spending patterns of customers?</p><p>What factors influence customer spending behavior?</p><p>Can customer spending behavior be used to predict future purchases?</p><p>How can Adventure Works improve customer engagement and increase revenue based on customer spending patterns?</p><p>To address these questions, Adventure Works needs to conduct a comprehensive analysis of customer spending patterns using data analytics techniques. The analysis will involve collecting and analyzing data on customer transactions, including the products or services purchased, the frequency of purchases, the amount spent, and the payment method used. The data will be used to segment customers based on their spending habits, identify high-value customers, predict future purchasing behavior, and develop targeted marketing campaigns to improve customer engagement and loyalty.</p><p><strong>Data Sourcing:</strong> The data for this project was gotten from the Adventure works 2019 database. First and foremost, I downloaded the Adventure Works 2019 backup file <a href="https://www.youtube.com/redirect?event=video_description&amp;redir_token=QUFFLUhqbTVtbGl1UkFUTHVlaEx2cE1mNmhxa2xTXy1fQXxBQ3Jtc0trUkM3OWNUQ19WcS1Tc1I1QzByV20xeWhuWWxoMnZLZHlQMkVjRkRaSno4cGtudF9iVGl3Z1RMTEk2VmxMVHRqLWlYa25qMjJ3QWtuQkhWV25RZ19Md1lxT2JvRm1kMnI5UDRiRTBDNEh6MUlPOGJUNA&amp;q=https%3A%2F%2Fgithub.com%2Fmicrosoft%2Fsql-server-samples%2Freleases%2Fdownload%2Fadventureworks%2FAdventureWorks2019.bak&amp;v=VpY0Q_kwtIw">here</a> . After that I restored database into my Microsoft SQL server, the database is made up of 71 tables so I had to study the database schema diagram and database dictionary to know which tables I needed to use for my analysis. You can download the database dictionary <a href="https://www.youtube.com/redirect?event=video_description&amp;redir_token=QUFFLUhqbGxlcHpOY1ZhMEZSdDV5Q3ZIekdoR0tfR3lkQXxBQ3Jtc0tuVWlPc1pQU2JvUnhMdl9NdVhIWVZRYnRiZHZMb0FiN3BFZnczZk03U2tIa0xod2dxSTRIRTl3Vm9zdkJaQTctSVJJRHJXLThOdFZsc19CZ2dCS18yZkQyTDJfN1J2M3hjM2hBWkp4ZzBXUkF3dDA0WQ&amp;q=https%3A%2F%2F1drv.ms%2Fb%2Fs%21AvIjHUSDuSKA-k7NrVlgzRQbynuu&amp;v=AnfWWapSncY">here</a> and also download the schema diagram <a href="https://www.youtube.com/redirect?event=video_description&amp;redir_token=QUFFLUhqbHRQUDdUTTIwWXZXSW1jUXdaQUtoNldvOVBCUXxBQ3Jtc0ttVWNVYkdCcWMwWGYxQXllM1pGVlNxNm5SVHQ0Ui1IRzdXZWhIVzRaMllTMWtfd0t4ckVFZzZKbEhRQWdxaERNRDMxV2NkX01NRUkwVTc2S294bjNCOTZxYWMzaHc0bXBzWldSdWc4NFZ1RUhia0NBZw&amp;q=https%3A%2F%2F1drv.ms%2Fu%2Fs%21AvIjHUSDuSKA-k2UHQDDxbxvI5io&amp;v=AnfWWapSncY">here</a> .</p><p><strong>Skills Demonstrated</strong></p><p>I demonstrated a few skills in the execution of this project and they are:</p><p>1. Data Retrieval and Extraction.</p><p>2. Data Querying, Cleaning and Transformation</p><p>3. DAX</p><p>4. Data Visualization:</p><p>5. Predictive Analytics</p><p>6. Business Intelligence</p><p><strong>Tools Used</strong></p><p>The tools used for this project are Microsoft SQL Server and Power BI</p><p><strong>Data Transformation/Modelling</strong></p><p>After restoring the Adventure Works database into Microsoft SQL Server, it was time for me to write queries which will help answer the questions for my analysis. Out of the 71 tables in the database, I used only 7 tables for my analysis and they are</p><p>1. Sales.SalesOrderHeader</p><p>2. Sales.Customer</p><p>3. Sales.SalesTerriitory</p><p>4. Sales.SalesorderDetails</p><p>5. Production.Product</p><p>6. Production.SubCategory</p><p>7. Production.ProductCategory</p><p>After writing the queries on Microsoft SQL server, I created views for all the queries I written.</p><p>Then I launched my Power BI, clicked on the get data button and clicked on SQL database, then I connected my SQL server to Power BI by imputing my Microsoft SQL server name and the database I needed to extract information from and connected.</p><p>After the connection was established, I selected the views which I had earlier created in Microsoft SQL server.</p><p>After importing the views and previewing them, I loaded it into Power query editor and performed some cleaning on the data like renaming, dropping unwanted columns, replacing values, checking for duplicates and null values.</p><p>After cleaning all the views which I imported as tables into Power BI using the power query editor, I loaded the tables into Power BI and closed Power Query editor. I Then proceeded to the modelling tab to ensure that ensure that the data was modelled properly and if not model it properly to create connection between tables. The schema for this project is a Snow-flake schema.</p><p><strong>Analysis</strong></p><p>The analysis was done based on the views that were created in Microsoft SQL server and imported to Power BI.</p><p><strong>Insight and Visualization</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/919/1*1aiz4vrMaXt_Adgj1qAUxw.png" /></figure><p>· In the product category, sales of bike generated the highest revenue of $94.65M followed by sales of components which generated $11.48M</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/948/1*8XDCT5KsiGXiExcucj2noA.png" /></figure><p>· The average processing time for all products is 7 days</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/1*6SG5PfTmnpK08M7FjWzSTQ.png" /></figure><p>· Within the four years covered in this analysis, 2013 was the year that recorded the highest number of orders of 14200, followed by 2014 which recorded 11800 orders.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/879/1*_YKjC9hjhmkFhn0FejuB4g.png" /></figure><p>· France had the highest Average order value and Australia recorded the highest customer count.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/626/1*gyuFs66gxJ83JEs6N8fkng.png" /></figure><p>· The number of orders definitely affected the revenue, hence, 2013 generated the highest revenue of $49M</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/631/1*UokkcMH8eKGfq-j2rJjz4w.png" /></figure><p>· USA is the country that generated the highest revenue</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/878/1*rAYAoU566IK8xJZOEHrpvA.png" /></figure><p>· The customer with the ID 29818 generated the highest revenue of $989,180.00</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/523/1*duVxPJI3BA9SFbNHPOzysw.png" /></figure><p>· The customer with the ID 29818 had the highest purchase frequency of 28</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qS6rjGOd1yInPcbqblOsTg.jpeg" /><figcaption>Adventure Works Customer Spending Pattern Report</figcaption></figure><p><strong>Recommendation</strong></p><p>1. Adventure works should identify the most profitable customers and offer them personalized deals or loyalty programs to encourage repeat purchases.</p><p>2. Measures should be put in place to ensure that the products that are most popular among customers and are in high demand are always well-stocked and easily accessible.</p><p>3. Discounts should be applied to products that have low sales volumes.</p><p>4. Seasonal spending patterns should be monitored to ensure that inventory levels are aligned with demand during peak seasons.</p><p>5. Consider upselling or cross-selling opportunities to customers who have purchased a particular product or service.</p><p>6.Better ways to reduce costs should be considered without sacrificing quality, such as negotiating better prices with suppliers or optimizing inventory management.</p><p>7. Customer spending patterns should continuously be monitored and analyzed to identify new opportunities for growth or improvement.</p><p>Kindly find the link to the report <a href="https://app.powerbi.com/groups/me/reports/bba24acc-f378-4290-9fff-e52c434f4bf0/ReportSection">here</a></p><p>The End</p><p>Thank you for taking your time to go through this article/ project</p><p>Your feedback and recommendation will be greatly appreciated.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=65338a2d0e2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Air Quality in Beijing]]></title>
            <link>https://medium.com/@blessingekwere/air-quality-in-beijing-df72ca6bf4ae?source=rss-c1a243eecad1------2</link>
            <guid isPermaLink="false">https://medium.com/p/df72ca6bf4ae</guid>
            <dc:creator><![CDATA[Blessing Ekwere]]></dc:creator>
            <pubDate>Fri, 07 Apr 2023 15:51:38 GMT</pubDate>
            <atom:updated>2023-04-27T17:00:00.524Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Impact of weather Conditions on Air Quality</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ixV7D49NDqLEPTZAcR9lQQ.jpeg" /><figcaption>Photo Source: Google</figcaption></figure><p><strong>Introduction</strong></p><p>Beijing is the capital city of the People’s Republic of China and was formerly known as Peking. It is the world’s most populous city with an estimated population of over 21 million residents. The city is located in the northern part of China and covers an area of 16,500 square kilometers.</p><p>In the past few years, air pollution problem in China, especially Beijing, has been so severe that it has received widespread attention from all over the world. Cities are dense areas of economic activities, and therefore, populations, and Beijing is the political and economic center of China. After a stage of radical pursuit of economic growth, improving air quality and overall living environment is the current focus of China’s realization of green development. Therefore, it is important to study Beijing’s air quality issues to find ways to tackle air pollution problems and provide a reference for other cities.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CS_QTB75HgQjYvMthvyx1A.jpeg" /><figcaption>Photo Source: Google</figcaption></figure><p><strong>Air quality basics</strong></p><p>Air quality refers to the degree to which the air is suitable or clean enough for humans or the environment.</p><p>Good air quality means the air is free of harmful substances. Air quality is rated using air quality categories (AQC) which rate the amount of pollution in the air based on airborne concentrations of major pollutants — the higher the AQC rating, the poorer the air quality.</p><p>It has been reported that air quality categories (AQC) based on visibility and 6 major airborne pollutants:</p><p>· ground-level ozone</p><p>· carbon monoxide</p><p>· sulfur dioxide</p><p>· nitrogen dioxide</p><p>· particulate matter as PM10 and PM2.5.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*thTwEev1lYlAupbeVo9E3Q.jpeg" /><figcaption>Photo Source: Google</figcaption></figure><p><strong>Data Sourcing</strong></p><p>The data was downloaded in a comma separated value (csv) format from a website. It is made up of 13 columns and 43,824 rows. The link to the dataset can be found here. The columns are</p><p>. No — Row Number</p><p>. Year — Year of the data in the row</p><p>. Month — Month of the data in the row</p><p>. Day — Day of the data in the row</p><p>. Hour — Hour of the data in the row</p><p>. pm2.5 — pm2.5 concentration</p><p>. DEWP — Dew Point</p><p>. TEMP — Temperature</p><p>. PRES — Pressure (hpa)</p><p>. cbwd — Combined Wind Direction</p><p>. Iws — Cumulated wind speed (m/s)</p><p>. Is — Cumulated hours of snow (hours)</p><p>. Ir — Cumulated hours of rain (hours)</p><p>The link to the dataset can be found <a href="https://archive.ics.uci.edu/ml/datasets/Beijing+PM2.5+Data">here</a></p><p><strong>Problem Statement</strong></p><p>The air quality in Beijing is a major concern due to high levels of particulate matter 2.5 (PM 2.5) pollution, which poses serious health risks to the population. Despite efforts to reduce emissions and improve air quality, PM2.5 levels in Beijing remain consistently high, particularly during winter months when coal burning increases. To better understand the sources and drivers of PM2.5 pollution in Beijing, there is need for an analysis of the air quality data and identification of key factors contributing to the persistent high level of pollution</p><p>Having known all of these, let’s now take a look at the air quality data in Beijing to draw insights and help make informed decisions. Amonst the questions to be answered in this project are:</p><p>1. What is the total pm2.5 of combined wind direction (cbwd)</p><p>2. What the total hours of rainfall of combined wind direction (cbwd)</p><p>3. What is the prediction on pm2.5 value in air quality in the next five years?</p><p>4. What is likely to be the total hours of rainfall in the next five years?</p><p>5. Which season, month and year recorded the highest Pm2.5?</p><p><strong>Tools Used: </strong>Power BI and Power Query Editor</p><p><strong>Skills Demonstrated</strong></p><p>The skills recorded in the course of carrying out this project are</p><p>. Data Cleaning, Transformation, and manipulation</p><p>. Data Analysis Expression (DAX)</p><p>. Data Visualization</p><p><strong>Data Transformation</strong></p><p>. To begin the data transformation, I launched the Power BI interface and used the get data command to import the data as a Text/CSV file.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TL7n_YbtMPkYAoSyxvUdXg.png" /><figcaption>Getting Data as Text/ CSV</figcaption></figure><p>. I then clicked the transform button to load the data into power query editor.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hi1t0BQ0BxPV8efF_ih_fg.png" /><figcaption>Loading the data into power query editor</figcaption></figure><p>. I used the first rows as headers.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ctHRgEDMIXf5DH8_EokN5A.png" /><figcaption>Before setting first rows as headers</figcaption></figure><p>. I then selected the year, month, and day columns, clicked on merge columns, used hyphen as a delimiter to separate the values and then I renamed the column to full date and formatted it to date type.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KcEEtlcwYQXsidPlRVoIEA.png" /><figcaption>Merging Year, Month and Day column to full date column</figcaption></figure><p>. I noticed that the pm2.5 column had “N/A” as values, which I replaced with Zero(0) using the replace value command</p><p>. The NE in cbwd column was replaced with North-East, while the NW and SE were replaced with North-West and South-East respectively. I then replaced the “cv” in cbwd column to South-West, since it might have been a data entry error.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*W5wdZ17FzCdGMtZBgAeDHg.png" /><figcaption>Replacing the values in cbwd column</figcaption></figure><p>. I then proceeded to convert the pressure column from hectopascal (hpa) to atmospheric pressure (atm) using the custom column command</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*voVBhDYa1atXDFuyW7JXMQ.png" /><figcaption>Converting the values of pressure from hpa to atm</figcaption></figure><p>. I also used the conditional column command to add a column named “season”.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bDyiq1WMPD6xBedmgF-SSg.png" /><figcaption>Using conditional column command to add a new column called season</figcaption></figure><p>The season column was grouped as follows:</p><p>December to February was called winter.</p><p>March to May was called Spring.</p><p>June to August was called Summer.</p><p>September to November was called Autumn.</p><p>After this, the data was then loaded into Power BI for further analysis and Power query was closed.</p><p><strong>Analysis and Visualization</strong></p><p>.Looking at all combined wind direction and the total amount of pm2.5, it is observed that South East had the highest average of PM2.5 with a total of 120.20 µg/m3, followed by South West with a total of 105.623 µg/m3, North West and North East had a total of 85.83 µg/m3 and 63.83 µg/m3 respectively.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/511/1*__KjUNiwC7TwUOkW5Sm2vA.png" /><figcaption>Total Hours of Rain by Combined Wind Direction</figcaption></figure><p>.Comparing the four combined wind directions, it can be seen that North West had the highest number of hours of rainfall with a total of 3755 hours, followed by South East with a total of 1792 and North East and South West had a total of 1642 and 1353 hours of rainfall respectively.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/478/1*oryv7uJJUuVxoQ7OxJDUvg.png" /><figcaption>Average PM2.5 by Combined Wind Direction</figcaption></figure><p>. Comparing the total amount of PM2.5 across the years, January 2013 had the highest amount of PM2.5 with about 191.97 µg/m3 and January 2011 recorded the lowest amount of pm2.5 with 40.55 µg/m3.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/892/1*W5ZWi84ZXkxDyFf18O8h7Q.png" /><figcaption>Average PM2.5 by Month and Year</figcaption></figure><p>. Considering pm2.5 across all seasons, the analysis showed that PM2.5 was highest in winter season with 28.44% followed by Autumn with 26.35%</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/357/1*9WaMNX9v27paTG6BTHcpRw.png" /><figcaption>Percentage of PM2.5 across all seasons</figcaption></figure><p>. I carried out a prediction of pm2.5 for the next five years which is from 2015 to 2019, and the analysis 2015 will have a pm2.5 of 96.10 µg/m3, 2016 will have a pm2.5 of 90.84 µg/m3, while 2017, 2018 and 2019 will have a pm2.5 value of 85.51 µg/m3, 100.76 µg/m3 and 96.63 µg/m3 respectively.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/924/1*4-iqu9j_q32wbjhzGAGfbQ.png" /><figcaption>Prediction of Average PM2.5 for the next five years</figcaption></figure><p>. I also carried out a prediction of the total hours of rainfall for the five years and it showed that in 2015 the total hours will be 2347, in 2016 the total hours will be 1231, in 2017 the total hours will be 2366 and in 2018 and 2019 the hours of rainfall will be 1410 and 1188 hours respectively.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/569/1*qZoifHwqyVS7PjVTWm-NGg.png" /><figcaption>Prediction of the total hours of rainfall for the next five years</figcaption></figure><p>. From the analysis, it showed that as the hours of snow increased, there was also an increase in the average amount PM2.5 though not a steady increase, the number of hours it rained remained zero when the number of hours it snowed increased except for the increase in hours that it rained when it only snowed for two hours. This means that there is likely to be an increase in PM2.5 after it rains than during the rain. This is because rain can help wash out PM2.5 particles from the atmosphere effectively removing them from the air we breathe because rain captures the particles and pulls them down to the ground.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/449/1*D8EiGz0uk5KTA94XttpB3g.png" /><figcaption>Analysis of how snow hours and average rainfall hours affects PM2.5</figcaption></figure><p>. The analysis also showed that pm2.5 kept increasing with temperature though not on a steady increase, while dew point kept reducing until the temperature got to 10 degrees Celsius then the dew point began to increase. This is likely because higher temperatures can accelerate chemical reactions that produce PM2.5 particles. Also, when dew point is high, it can help trap PM2.5 particles in the air rather than being dispersed and this is because water droplets can act as a surface for particles to adhere to.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*MZMTd45hXyu_qh7OXr1Sog.png" /><figcaption>Analysis of how Dew Point and Temperature affects Pm2.5</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*220YbLJ-qsilomQkIOoFDA.jpeg" /><figcaption>Air Quality In Beijing Dashboard</figcaption></figure><p><strong>Conclusion and Recommendation</strong></p><p>In conclusion, Beijing has been known to face severe air quality problems due to various activities such as industrial activities, transportation and coal-based energy production. To address these problems a few recommendations I will make are:</p><ol><li>Enforcing stricter regulations on industrial emissions, and incentivizing companies to adopt cleaner production methods.</li><li>Raising Public awareness of the health risks associated with air pollution and encouraging individuals to take steps to reduce their own emissions.</li><li>collaborating with other countries and international organizations to share best practices and technologies for improving air quality.</li><li>Promoting renewable energy sources and phasing out coal-based energy production.</li></ol><p>Finally, while there have been some positive developments in improving air quality in Beijing, there is still much work to be done to address the issue. It will require the commitment and cooperation of stakeholders, from government and industry to individuals, to make a meaningful difference in reducing air pollution and improving the health and well being of the city’s residents.</p><p>Kindly find the link to the Power BI report <a href="https://app.powerbi.com/groups/me/reports/1f3a7b1d-1963-4a54-9540-3ba28936f6b6/ReportSection">here</a></p><p>Thank you for reading this article. I sincerely appreciate</p><p>Kindly share your thoughts, recommendations and comments.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=df72ca6bf4ae" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Nashville Housing Dataset]]></title>
            <link>https://medium.com/@blessingekwere/nashville-housing-dataset-440574440c82?source=rss-c1a243eecad1------2</link>
            <guid isPermaLink="false">https://medium.com/p/440574440c82</guid>
            <dc:creator><![CDATA[Blessing Ekwere]]></dc:creator>
            <pubDate>Wed, 05 Apr 2023 12:55:55 GMT</pubDate>
            <atom:updated>2023-04-05T15:12:57.893Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/902/1*bxWqmqRY4Sh-WygSJVS_cQ.jpeg" /><figcaption>Photo Source: Google</figcaption></figure><p><strong>Introduction</strong></p><p><strong>Nashville</strong> is the state capital of and most populous city in Tennessee and the seat of Davidson County. With a population of 689,447 at the 2020 U.S census. Nashville is the 21st most-populous city in the U.S, and the fourth most populous city in the southeastern U.S located on the Cumberland River the city is the center of the Nashville metropolitan area, which is one of the fastest growing in the nation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/500/1*axBWRIPJM-t0HwxN1P_wdg.jpeg" /><figcaption>Photo Source: Google</figcaption></figure><p><strong>About the Dataset: </strong>This dataset is about Nashville housing. The data contains 19 columns and 56,477 rows. The columns which are:</p><p>. UniqueID — Id number attributed to buyer</p><p>. ParcelID — Code attributed to a land</p><p>. LandUse — Shows the different use of land</p><p>. SalesPrice — Cost of land</p><p>. LegalReference — Citation is the practice of crediting and referring to a authorized documents and sources.</p><p>. OwnerName: Name of land owner</p><p>. Acreage: The size of an area of land in acres</p><p>. LandValue: The worth of the land</p><p>. BuildingValue: Worth of the building</p><p>. YearBuilt: Year the building was built</p><p>. FullBath: A bathroom that includes shower, a bathtub, sink and toilet</p><p>. SaleDate: Date when the land was sold</p><p>. SaleAddress: Address of the land sold</p><p>. City: Location of the land</p><p>. OwnerAddress: Owner’s house address</p><p>. OwnerCity: City where the owner lives</p><p>. OwnerState: State where the owner lives</p><p>The link to the dataset can be found <a href="https://github.com/blessingekwere/Nashvielle-Housing-Dataset/blob/main/Nashville%20Housing%20Data%20for%20Data%20Cleaning.xlsx">here</a></p><p><strong>About the Project: </strong>This project is a data cleaning of the Nashville Dataset with the aim of ensuring that it is fit to be used for further analysis.</p><p><strong>Tool Used: </strong>Microsoft SQL Server</p><p><strong>Observation in the data</strong></p><p>There were few observation in the data which were</p><p>. Wrong formatting of date column</p><p>. Some rows in the property address had null values</p><p>. The property address column has both the city and house address in the same column</p><p>. The owner address has the state, city, and address in the same column</p><p>. Some rows in the SoldAsVacant has “Y” and “N” instead of “Yes” and “No”</p><p>. The data had some duplicate rows</p><p>. The data had some columns that would not be needed for analysis</p><p><strong>Data Cleaning</strong></p><p>The data was imported in a comma separated value (csv) format into Microsoft SQL server using the import and export wizard and after that the data cleaning process commenced. Kindly follow through to see the steps I took to clean the data</p><p>The first thing I had to do was to standardize the date format</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2ekLoLjEDP-q0wnS-9SwCQ.png" /><figcaption>Standardizing date format in SaleDate column</figcaption></figure><p>I then proceeded to populate the property address column</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bmLNTE8bb8U6nM00Rgbbdg.png" /><figcaption>Populating PropertyAddress column</figcaption></figure><p>After that I went on to break out the PropertyAddress column into Individual Columns</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wojjxwhYFGLVFiG5GGCgOw.png" /><figcaption>Breaking out the PropertyAddress column into individual column</figcaption></figure><p>I also broke out the OwnerAddress column into Individual Columns</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*w7W2wNVKQXKnxoFVxQ5cxQ.png" /><figcaption>Breaking out the OwnerAddress column into individual column</figcaption></figure><p>I updated the SoldAsVacant column with a “Yes” for “Y” and “No” for “N”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*FxjQvmJqIXf3HtgB9tQMxQ.png" /><figcaption>Updating the SoldAsVacant column</figcaption></figure><p>Finally I decided to delete the unused columns or columns that will not be needed for further analysis of this project</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6MAsCZPuCQ5HPHJU767mBQ.png" /><figcaption>Deleting Unused columns</figcaption></figure><p><strong>Conclusion</strong></p><p>So far, I have been able to clean the data by transforming the data, formatting the columns correctly, removing columns that will not be useful for my analysis.</p><p>Kindly drop your review in the comment. Your suggestions and recommendation will be highly appreciated. Thank you</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=440574440c82" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Amazon Sales Dataset]]></title>
            <link>https://medium.com/@blessingekwere/amazon-sales-dataset-71b0eb90c866?source=rss-c1a243eecad1------2</link>
            <guid isPermaLink="false">https://medium.com/p/71b0eb90c866</guid>
            <dc:creator><![CDATA[Blessing Ekwere]]></dc:creator>
            <pubDate>Sat, 01 Apr 2023 14:36:50 GMT</pubDate>
            <atom:updated>2023-04-02T02:19:49.523Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/512/1*4kk2kkKilaKYrlsxl4DKHA.png" /><figcaption>Photo Source: Google</figcaption></figure><h3>About Amazon</h3><p>Amazon is an American multinational technology company focusing on e-commerce, cloud computing, online advertising, digital streaming, and artificial intelligence . It has been referred to as “one of the most influential economic and cultural forces in the world”, and is one of the world’s most valuable brands. Amazon takes care of everything from shipping and pricing to customer service and returns.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/540/1*fWmi5RLZzyVD0-QpN4UPfA.jpeg" /><figcaption>Photo Source: Google</figcaption></figure><h3>About the Data</h3><p>This dataset has the data of 1K+ Amazon product’s ratings and reviews as per their details listed on the official website of Amazon. The data was downloaded from Kaggle in a comma separated value format (csv) and it has 16 columns and 1465 rows. The columns which are:</p><ul><li>product_id — Product ID</li><li>product_name — Name of the Product</li><li>category — Category of the Product</li><li>discounted_price — Discounted Price of the Product</li><li>actual_price — Actual Price of the Product</li><li>discount_percentage — Percentage of Discount for the Product</li><li>rating — Rating of the Product</li><li>rating_count — Number of people who voted for the Amazon rating</li><li>about_product — Description about the Product</li><li>user_id — ID of the user who wrote review for the Product</li><li>user_name — Name of the user who wrote review for the Product</li><li>review_id — ID of the user review</li><li>review_title — Short review</li><li>review_content — Long review</li><li>img_link — Image Link of the Product</li><li>product_link — Official Website Link of the Product</li></ul><p>The link to the dataset can be found <a href="https://www.kaggle.com/datasets/karkavelrajaj/amazon-sales-dataset">here</a></p><h3>Objective of the Project</h3><p>The Objective of this project is to perform an exploratory data analysis, discover insights and trends in the data that will be used to make informed decisions and make recommendations based on the data if necessary.</p><h3>Tools Used</h3><p>The tool used for this project are Power Query Editor and Power BI</p><h3>Observation with the Data</h3><p>After downloading the data, there were few observations I made about the data. Some of them were:</p><p>. Incorrect formatting of columns</p><p>. Inconsistent spacing</p><p>. The dataset had no date, which means it covers an unspecified period of time</p><p>. The dataset had some special and unprintable characters</p><p>. The dataset contained some irrelevant columns which will not be useful for analysis</p><h3>Data Cleaning</h3><p>As mentioned earlier, I used Power Query editor and Power Query for this project.</p><p>Before starting out any project, the purpose of cleaning data cannot be over-emphasized. Knowing that from my observations which I listed earlier, there were quite a number of things that had to be done on the data to make it fit for use for further actions or analysis, I had to clean the data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*m37_vPiFlWWOdw4_PTsZNg.png" /><figcaption>Power BI launch page</figcaption></figure><p>To start, I launched Power BI, proceeded to the ‘get data’ command in Power BI. Then I chose the Text/CSV format to import the data into Power Query Editor.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4JBt1MaQ3AHaXtliPXBjRw.png" /><figcaption>Importing the data as Text/CSV</figcaption></figure><p>The data was loaded to power query editor, and then I proceeded to the cleaning by clicking on the transform button.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*mARK5wwD5moBlTDlKamN3Q.png" /><figcaption>Loaded data in Power BI</figcaption></figure><p>Since I will be needing the category column for my analysis and it needed to be cleaned and formatted properly. I had to split the category column into two new columns namely primary and secondary category using a delimiter.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*owHWcEA2MXWotuBMBvs8gg.png" /><figcaption>Splitting the category column</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bYzXvx-0nZaf9KYhFraOCg.png" /><figcaption>The category column(s) after splitting</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Lgi2Pc-sl5DkmHb6FYqgug.png" /><figcaption>The category column(s) after splitting which some were later removed</figcaption></figure><p>After splitting the columns and renaming them, I discovered that the ‘primary category’ and ‘secondary category’ were without proper spacing and had special characters. The ‘&amp;’ symbol was replaced with ‘and’ using the replace value command. And also the columns were cleaned using the trim command and spaced properly.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZfMHy3fDH6e_E2l4iIGjwA.png" /><figcaption>Cleaning the primary and secondary columns using the replace value command</figcaption></figure><p>Proceeding to the discount price and actual price columns , I noticed that the value was in Indian Rupee (₹) which I need to change to US dollars ($) using the current exchange rate since the date is for an unspecified period of time. But before then, I had to remove the Indian Rupee (₹) using the replace value command and formatted it to fixed decimal data type.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EhzgdlAVvZZaEiHjP-SaCw.png" /><figcaption>Discounted_Price and Actual_Price columns with Indian Rupee(₹) which were later removed</figcaption></figure><p>I proceeded to carrying out the conversion by using the custom column command in the add column tab. As earlier mentioned, I used the current exchange rate which is ₹1 is equal to $ 0.012, I simply multiplied each column by 0.012 and renamed them respectively, repositioned the new added columns and dropped the other ones.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TguZ5Il-c3ongzCgL5tWYA.png" /><figcaption>Converting Discounted_Price column</figcaption></figure><p>After dropping the other discounted_price and actual_price columns which were replaced with the newly converted columns, I went further to the rating column.</p><p>The rating column had two null values which I replaced them with zero. I then decided to add an additional column and name it rating_scale of 1 to 5 since I will be needing that for my analysis. To do that, I used the conditional column. For my rating scale, <strong>ratings &lt;=1</strong> were classified as <strong>Very Dissatisfied</strong>, <strong>ratings &lt;=2</strong> were classified as <strong>Dissatisfied, ratings &lt;=3 </strong>were classified as <strong>Neither Satisfied nor Dissatisfied, ratings &lt;=4</strong> were classified as <strong>Satisfied </strong>else other ratings were classified as <strong>Very Satisfied.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5Vivg5EYrBj0-_B9RFJ6dQ.png" /><figcaption>Using conditional column to add the rating_scale column</figcaption></figure><p>After the data cleaning, I decided to drop some columns that will not be useful in my analysis. The columns I dropped were ‘User_id’, ‘User_name’, ‘review_id’, review_title, ‘img_link’ and ‘product_link’. I also dropped row 1280 because it had an errorneous data that could not be resolved. So finally I was left with 11 rows and 1464 columns to be used for my exploration and analysis.</p><h3>Analysis</h3><p>In order to be able to analyze my cleaned data property and get meaningful insights from it. I decided to use the new measures to command to get new measures which which either serve as a metric or be combined with other functions using DAX to perform some calculations and give insight to the data. Some of the measures I incorporated into this analysis were:</p><p>. Loss Margin: This is the ratio that represents a loss percentage of a selling price. It is calculated by dividing the loss by the selling price and then multiplying it by 100.</p><p>. Loss: This is the difference between the actual price and discounted price.</p><p>. Review: This is the sum of rating count</p><p>. Products: This is the count of products that were sold within the unspecified period that the data covers.</p><h3>Visualization</h3><p>After creating the measures and going through the data to ensure that all was fine. I proceeded to the report tab of Power BI to visualize the data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*g9IhPCEmsAbsy9rLAhfR3g.jpeg" /></figure><h3>Insight</h3><p>. For the secondary category, amazon incurred more losses from sales of home theatre and TV video because of it’s discounts. The home theatre and TV video which also falls under electronics in primary category were mostly sold, hence making electronics to be product that the company incurred losses within this unspecified period.</p><p>. Considering the discount price by percentage in the primary category, <strong>electronics products</strong> enjoyed <strong>68.79%</strong> discount, followed by <strong>home and kitchen</strong> products which enjoyed <strong>22.84% </strong>discount and then <strong>computer and accessories</strong> products which enjoyed <strong>8.37%</strong> discount.</p><p>. In the secondary category products, <strong>Home theatre and TV video</strong> had a total discounted price of <strong>252,890</strong>, followed by <strong>mobile accessories</strong> with <strong>172, 290.</strong></p><p>. According to the rating scale, <strong>78.43%</strong> of customers were <strong>very satisfied</strong> with their purchases, while<strong> 21.57%</strong> of customers were <strong>satisfied</strong> and <strong>0.01%</strong> of customers were <strong>neither satisfied nor dissatisfied</strong> with their purchases. Which means no customer was dissatisfied. This also implies that amazon’s of giving its customers satisfaction and value for their money has been achieved.</p><h3>Recommendation</h3><p>Having gone through the amazon sales data for the unspecified period of time and analyzed it, it is amazing to say that customer satisfaction has been attained but then it is of great necessity that the cost and pricing options must be reviewed by the company to ensure that it achieves profitability while continuing to satisfy it’s customers and giving them value for their money.</p><p>Thank you so much for reading this article.</p><p>Your comments, suggestions and recommendations will be highly appreciated.</p><p>Please kindly click <a href="https://twitter.com/Eddie_Gregs?t=dF3996shVxvPJTePTtxDdw&amp;s=09">here</a> to connect with me on twitter</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=71b0eb90c866" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Chemicals in Cosmetics]]></title>
            <link>https://medium.com/@blessingekwere/chemicals-in-cosmetics-4322019e58c7?source=rss-c1a243eecad1------2</link>
            <guid isPermaLink="false">https://medium.com/p/4322019e58c7</guid>
            <dc:creator><![CDATA[Blessing Ekwere]]></dc:creator>
            <pubDate>Wed, 29 Mar 2023 11:49:51 GMT</pubDate>
            <atom:updated>2023-03-29T21:11:58.123Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/508/1*RRUDJRnTqsty90-2BSl3FQ.jpeg" /><figcaption>Photo Source: Google</figcaption></figure><p>Do you ever wonder what ingredients are in your favorite cosmetics? Did you know that there are over 12,000 chemicals used in cosmetics and personal care products?</p><p>A chemical is any substance that has a defined composition. In other words, a chemical is always made up of the same “stuff.” Some chemicals occur in nature, such as water. Other chemicals are manufactured, such as chlorine (used for bleaching fabrics or in swimming pools). Chemicals are all around you: the food you eat, the clothes you wear. You, in fact, are made up of a wide variety of chemicals. But in this context and for the purpose of this analysis, we will be focusing on the chemicals in your cosmetics and personal care products.</p><p><strong>Cosmetics</strong> are constituted mixtures of chemical compounds derived from either natural resources, or synthetically created ones. Cosmetics have various purposes. Those designed for personal care and skin care can be used to cleanse or protect the body or skin. Cosmetics designed to enhance or alter one’s appearance (makeup) can be used to conceal blemishes, enhance one’s natural features (such as the eyebrows and eyelashes), add color to a person’s face, or change the appearance of the face entirely to resemble a different person, creature or object. Cosmetics can also be designed to add fragrance to the body</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/820/1*vN5Scou_Bq_adsqmnbOAPA.png" /><figcaption>Photo Source: Google</figcaption></figure><h3>About the Dataset</h3><p>This data reflect information that has been reported to the California Safe Cosmetics Program (CSCP) in the California Department of Public Health (CDPH). The primary purpose of the CSCP is to collect information on hazardous and potentially hazardous ingredients in cosmetic products sold in California and to make this information available to the public. For all cosmetic products sold in California, the California Safe Cosmetics Act (“the Act”) requires the manufacturer, packer, and/or distributor named on the product label to provide to the CSCP a list of all cosmetic products that contain any ingredients known or suspected to cause cancer, birth defects, or other developmental or reproductive harm.</p><p>To assist companies with reporting, CDPH has compiled a list of reportable ingredients based on lists and reports available from the authoritative scientific bodies cited in the Act to cause cancer or reproductive harm; it is meant to serve as guidance and is not all-inclusive. Companies with reportable ingredients in their products must submit information to the California Safe Cosmetics Program.</p><h3>Objective of the project</h3><p>The objective of this project is to get to know more about the chemicals used in cosmetics and personal care products and to answer a few questions about it. I decided to use Microsoft SQL server for the exploration and analysis of this data and then use Tableau for the Visualization.</p><p>So what’s really in your makeup bag? Scroll through and take a look at some of the most popular cosmetics ingredients.</p><h3>Data Collection</h3><p>This data was downloaded from Kaggle. It contains 114,298 rows and 23 columns. The columns contained information such as</p><p>. Index — Serial Number</p><p>. CDPHId -The California Department of Public Health (CDPH) internal identification number for product.</p><p>. ProductName — Product name as entered by manufacturer, packer and/or distributor.</p><p>. CSFId- CDPH internal identification number for a color/scent/flavor.</p><p>. CSF — Color, scent and/or flavor as entered by manufacturer, packer and/or distributor.</p><p>. CompanyId — CDPH internal identification number for company.</p><p>. Company Name — Company name as entered by manufacturer, packer and/or distributor.</p><p>. BrandName — Brand name as entered by manufacturer, packer and/or distributor. .</p><p>. PrimaryCategoryId — CDPH internal identification number for category.</p><p>. PrimaryCategory — Type of product (13 primary categories: Baby Products, Bath Products, Fragrances, Hair Care Products (non-coloring), Hair Coloring Products, Makeup Products (non-permanent), Nail Products, Oral Hygiene Products, Personal Care Products, Shaving Products, Skin Care Products, Sun-Related Products, Tattoos and Permanent Makeup).</p><p>. SubCategoryId — CDPH internal identification number for subcategory.</p><p>. SubCategory — Type of product within one of the 13 primary categories.</p><p>. CasId — CDPH internal identification number for chemical.</p><p>. CasNumber — Chemical identification number</p><p>. ChemicalId — CDPH internal identification number for this chemical’s record specific to this product.</p><p>. ChemicalName — Name of chemical substance in cosmetic product. Note that chemical substances may have more than one Chemical Name.</p><p>. InitialDateReported — Date the product profile was created by manufacturer, packer and/or distributor, that is, the date that the product was first reported to CDPH.</p><p>. MostRecentDateReported — Date the product profile was last modified by manufacturer, packer and/or distributor.</p><p>. DiscontinuedDate — If applicable, date the product was discontinued.</p><p>. ChemicalCreatedAt — The date that this chemical was first reported to CDPH for this product.</p><p>. ChemicalUpdated At — Date this chemical report was last modified by manufacturer, packer and/or distributor. This field is unique from the product profile.</p><p>. ChemicalDateRemoved — If applicable, date the chemical was removed from product, i.e., reformulation.</p><p>. ChemicalCount — Total number of current chemicals reported for this product. This number does not include chemicals that have been removed from a product. This number is a calculated field based on current reporting.</p><p>The link to the dataset can be found <a href="https://www.kaggle.com/datasets/thedevastator/chemicals-in-cosmetics-what-s-really-in-your">here</a>.</p><h3>Data Cleaning</h3><p>I created a database called “Chemicals_In_Cosmetics” and then imported the data in a comma separated value (csv) format into Microsoft SQL server using the import and export wizard.</p><p>It was quite a messy one, so definitely before proceeding to anything else, the data had to be cleaned to make it suitable to be used for further analysis and decision making. Some of the steps I took to clean the data was removing excess spaces using the trim command, formatting all columns correctly, checking for duplicates, null values, unprintable characters and misspellings.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/924/1*L5Mp-zpafuw7w-R6IinLdA.png" /><figcaption>Using Trim Command to remove extra spaces from the Brand Name column</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/992/1*LepBB1lSkR14wNPEmNxqJQ.png" /><figcaption>Formatting the ChemicalDateRemoved column to Datetime</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WToK2qw1HgdQypD1u8Y4bg.png" /><figcaption>Correcting the date in the ChemicalDateRemoved Column</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nNEzg79cCCXt-D82sQB1Gw.png" /></figure><h3>Analysis</h3><p>After cleaning the data, I had to write queries to answer few questions about the data. Some of which are:</p><p>1: Chemicals used most in Cosmetic and Personal Care Products and showing the result in descending order.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*b8rvEwHvhqvWu-FGDK-ZyQ.png" /></figure><p>The result of this query showed that Titanium Oxide is the most used chemical in cosmetic and personal care product with a count of <strong>93,252</strong>. It was followed by Silica, crystalline with a count of <strong>2742</strong> and then retinol/retilyn esters with a count of <strong>2154</strong>.</p><p>Also, Arsenic (inorganic oxides) Cocamide, Spironolactone, Phenacemide, Extract of coffee bean, Diethanolamides of the fatty acids of coconut oil, Sodium Bromate, Acetylsalicylic acid, N-Nitroso-diethanolamine, Cocamide diethanolamine Vinyl acetate, Retinol palmitate, Distillates (coal tar), 2-Propyleneacrolein were the least used chemicals with a count of <strong>1 </strong>respectively.</p><p>2. Companies that used most reported chemicals in their cosmetics and personal care products.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ks3IiQ6ktZ8ybP8hFLT1tg.png" /></figure><p>The result of this query showed that L&#39;Oréal USA is the company that used most reported chemicals in their cosmetics and personal care product with a count of <strong>5,738</strong> followed by Revlon Consumer Product Corporation with a count of <strong>4,202</strong> and Bare Escentuals Beauty, Inc. with a count <strong>3,679</strong>.</p><p>The companies with the least reported chemicals were Zalan Products, Inc. with a count of 1, University Medical Pharmaceuticals Corp with a count of 1 and Cosmetic Dermatology Inc. with a count of 1 respectively.</p><p>3. Identify the brands that had chemicals which were mostly reported in 2018.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DU2FBJAMAN6nYkLw_PXXqA.png" /></figure><p>The result of this query showed Brands that had chemicals that were mostly reported in 2018</p><p>4. Which brands had chemicals discontinued and removed?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0Yop2JgYlIwk9rURLAcOJg.png" /></figure><p>This result of this query showed chemicals that were discontinued and removed</p><p>5. Identify the period between the creation of the removed chemicals and when they were actually removed</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*yWKCN8FZ9n2fyBUjgbKHtg.png" /></figure><p>This query showed the difference in the days between the day of creation and the date of removal of a chemical.</p><p>6. How long were removed chemicals in baby products used?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rCNKPa8usnlymG4Z7J0glw.png" /></figure><p>The result of this query showed that it took 2097 days for Titanium dioxide to be removed from baby products and it took Cocamide diethanolamine Trade Secret 545 days and 13 days respectively.</p><h3>Visualization</h3><p>As earlier stated, I used Tableau for the visualization. I used my visual to answer few questions about the data (not all the query results were Visualized). Below is the outcome of my visualization.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OxLzpjX62DdSJatvPG-9xw.png" /><figcaption>Chemicals In Cosmetics Dashboard</figcaption></figure><h3>Insights</h3><p>.From my analysis, it shows that Titanium Dioxide is the most used chemical in cosmetics and personal care products.</p><p>.I also found out that the brand that had most reported chemicals in 2018 was Two faced cosmetics.</p><p>. The analysis also showed that the company that used most reported chemicals for production was L&#39;Oréal USA.</p><p>. Also the analysis showed that Titanium Dioxide, Cocamide diethanolamine and Trade Secret were used for 2,097, 545 and 13 days respectively in baby products.</p><h3>Conclusion</h3><p>Now we have information on the chemicals in our cosmetics and personal care products which will help in making further informed decisions.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4322019e58c7" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[New York City Airbnb Market]]></title>
            <link>https://medium.com/@blessingekwere/new-york-city-airbnb-market-4c13b7ee9d55?source=rss-c1a243eecad1------2</link>
            <guid isPermaLink="false">https://medium.com/p/4c13b7ee9d55</guid>
            <dc:creator><![CDATA[Blessing Ekwere]]></dc:creator>
            <pubDate>Tue, 28 Mar 2023 16:58:40 GMT</pubDate>
            <atom:updated>2023-09-02T14:07:45.435Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/750/1*1Q7bE9k9ZbTQ9BFaYPz3LA.jpeg" /><figcaption>Photo Source: Google</figcaption></figure><h3>About New York Airbnb</h3><p>New York City (NYC) is the most populous city in the United States. It is one of the most-visited cities in the world. New York City comprises 5 boroughs sitting where the Hudson River meets the Atlantic Ocean. At its core is Manhattan, a densely populated borough that’s among the world’s major commercial, financial and cultural centers. Its iconic sites include skyscrapers such as the Empire State Building and sprawling Central Park. And because it is one of the most populous cities in the United States, it runs an Airbnb system.</p><p>New York Airbnb is an online market place that connects people who want to rent out their homes with people who are looking for accommodations in New York City. Since 2008 till date there has been many Airbnb listings to meet the high demand for temporary lodging for anywhere between a few nights to many months. Over time, guests and hosts have used Airbnb to expand on traveling possibilities and present more unique, personalized way of experiencing the world.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/661/1*POX5VVzz3OmBL9umtabF4g.jpeg" /><figcaption>Photo Source: Google</figcaption></figure><h3>About Dataset</h3><p>This data was downloaded from Kaggle, and it is the review data for January to July 2019. This data contains three tables namely the airbnb_last_review, airbnb_price and airbnb_room_type. All the tables contain 3 columns and 25,209 rows respectively and it covers the five cities of New York called Boroughs.</p><p>— <strong>The airbnb_last_review table comprises of</strong></p><p>. Listing_Id — Id numberof the listing</p><p>. Host name — Name of listing host(s)</p><p>. Last review columns — Date of review</p><p>— <strong>The airbnb_price table comprises of</strong></p><p>. Listing_Id — Id number of the listing</p><p>. Nbhood_full — Location of the listing</p><p>. price columns — Price of the listing</p><p>—<strong> The airbnb_room_type table comprises of</strong></p><p>. Listing_id — Id number of the listing</p><p>. Description — Description of the room type</p><p>. Room_type — Type of room available in the listing</p><p>The original dataset can be found <a href="https://www.kaggle.com/datasets/ebrahimelgazar/new-york-city-airbnb-market">here</a>.</p><h3>Objective of the Project</h3><p>The goal of this analysis is to use the dataset to perform and apply various data cleaning techniques. The analyzed data will be used to provide insights, make informed decisions.</p><p>This will be achieved by answering questions such as price and count of listing by borough and neighborhood, average price of listing per month, average price and revenue of listing by room type, and revenue of listing by month.</p><h3>Observation on the data</h3><p>The data was not a very messy one, yet there were a few things that had to be done to make the data suitable for use for further actions or analysis. Amongst the few things noticed were:</p><p>. Incorrect Formatting</p><p>. Inconsistent spacing</p><p>. Wrong column names</p><h3>Collecting the Data</h3><p>As earlier stated, the data was downloaded from Kaggle in a comma separated value format (.csv) and saved in a folder.</p><h3>Data Cleaning</h3><p>Power BI is the tool that I decided to use for this project. To start, I launched Power BI desktop and then went on to import the tables individually as Text/CSV through the ‘Get Data’ Command into Power BI. Let’s start with the last review table. After importing the data, the power query editor was open.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C3Lt2grL2TjALe2h1O0nwQ.png" /><figcaption>Importing the last review table as a CSV file into Power BI</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7LZdQJwas3afoQS6j74JZA.png" /><figcaption>Preview of the last review table before transforming</figcaption></figure><p>. In power query editor, I formatted all the columns in the table to their right data types. I then proceeded to clean the host name column by using the trim command to remove extra spaces.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-2Lc6mtGpUIlh5NedPa7_Q.png" /><figcaption>Duplicating the last review column and renaming the new column</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dyNmRyEVmTboNFVneot9hA.png" /><figcaption>Transforming the new month column</figcaption></figure><p>. I then duplicated the last review column, renamed it to ‘Month’ and transformed it based on the month because I will be needing the month name for my analysis.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*S-4aAu5TW50pRgy4rNkRxQ.png" /><figcaption>Duplicating the last review column and naming the new column ‘Month Number’</figcaption></figure><p>. I duplicated the last review column again, renamed it to ‘Month Number’ and transformed it based on the month and then sorted the column by month number because I will be needing the month number too for my analysis. The ‘&amp;’ symbol in the host name column was replaced with ‘and’ using the replace values command. And then the airbnb_last_review table was then loaded and closed into Power BI.</p><p>. Using the same steps of getting data, I imported the price table into power query editor. The “dollars” was eliminated using the replace values command and the right formatting applied to all columns. The “nhbood-full’ column was replaced with “neighborhood”.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_jVL7OEGOsplneIJ2t54Jg.png" /><figcaption>Splitting the neighborhood column</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9gnKYCHJ-zIHdjdV5BJrwg.png" /><figcaption>Splitting the neighborhood column using a delimiter</figcaption></figure><p>I noticed that the neighborhood column had the major neighborhood/ cities in New York City and other details, so I decided to split that column into “neighbouhood_1” and “neighbourhood_2” using comma(,) as a delimiter. Neighborhood_1 column contained the major cities in New York City called Borough and neighborhood_2 contained other details about the neighborhood. And then airbnb_price table was loaded and closed into Power BI.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gvhHx3UsaLrsdm1BUDoZDg.png" /><figcaption>The neighborhood column after splitting</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GygbcFYrXA0pB4nudpfH5Q.png" /></figure><p>. Using the same steps of getting data, all columns were formatted properly. I imported the room type into power query editor. I noticed the inconsistencies in the room type column, some were in upper case and some were in lower case which I had to clean using by capitalizing every value in that column. I replaced the “Entire Home/Apt” to “Entire Home” using the replace values command. And then airbnb_roomtype table was loaded and closed into Power BI.</p><h3>The Analysis</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XTTEViU8baArfNPlyFcgtg.png" /></figure><p>. Knowing that all the details I needed for analysis were available in the tables except the Average price per month. I created a new measure by right clicking on the price airbnb_price and choosing new measure, which then opened the formular bar in Power BI. I then used the formular ‘Avg_Price_Per_month = Average (airbnb_price[Price]) * 365/12’ to get the average price per month.</p><h3>The Visualization</h3><p>For the visualization, there were few questions I needed to answer with my visuals and they are</p><ol><li>What is the price of listing by neighborhood?</li><li>What is the average price of listing by neighborhood?</li><li>What is the average price and total revenue of listing?</li><li>What is the revenue of listing by month?</li><li>What is the average price of listing by room type?</li><li>What is the count of listing by borough?</li><li>What is the percentage of revenue by room type?</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*kyUVbV63sXf_MhQUPBcJoA.jpeg" /><figcaption>New York Airbnb Dashboard</figcaption></figure><h3>Insights</h3><p>. Manhattan generated the most revenue among the five boroughs with a total of $1.9M followed by Brooklyn with a total of $1.28M. The Queens, Bronx and Staten Island generated $320,765, $55,231 and $22,974 respectively.</p><p>. For the five boroughs, The average price of listing for Manhattan was $184.00, followed by Brooklyn with an average of $121.90. The Queens, Bronx and Staten Island generated $92.81, $86.04 and $79.12 respectively.</p><p>. June recorded the highest revenue of $1.78M followed by July with $0.68 which means that there was an increase in revenue from January to June and then 61.8% decrease in revenue in July.</p><p>. For the three room types, the average price of listing for Entire Home was $194.17, followed by Private Room with an average of $81.64 and Shared Room $79.12.</p><p>. For the three room types, Entire Home generated 73.1% of the revenue, Private Room generated 25.94% of the revenue and Shared Room generated 0.88%.</p><p>. Of the five boroughs, Brooklyn had the highest number of listings with 10,466 listings, followed by Manhattan with 10,322. Queens, Bronx and Staten Island had 3456, 698 and 267 listings respectively.</p><p>Kindly feel free to find and interact with my Power BI dashboard <a href="https://app.powerbi.com/groups/me/reports/20f9614d-d4b3-4810-98a7-b13d09351b7f/ReportSection">here</a></p><p>Your feedback and reviews will be highly appreciated</p><p>Thank you for reading</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=4c13b7ee9d55" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>