<?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 Modu Daniel on Medium]]></title>
        <description><![CDATA[Stories by Modu Daniel on Medium]]></description>
        <link>https://medium.com/@modudharniel16?source=rss-1c3429dbd153------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*mj39ypJd1NjzfjwvrV_JBA.jpeg</url>
            <title>Stories by Modu Daniel on Medium</title>
            <link>https://medium.com/@modudharniel16?source=rss-1c3429dbd153------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 01:59:11 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@modudharniel16/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[KPMG Virtual Data Analytics Internship (Part 2)]]></title>
            <link>https://modudharniel16.medium.com/kpmg-virtual-data-analytics-internship-experience-ii-f9094a2de325?source=rss-1c3429dbd153------2</link>
            <guid isPermaLink="false">https://medium.com/p/f9094a2de325</guid>
            <category><![CDATA[data]]></category>
            <category><![CDATA[kpmg]]></category>
            <category><![CDATA[jobs]]></category>
            <category><![CDATA[science]]></category>
            <category><![CDATA[internships]]></category>
            <dc:creator><![CDATA[Modu Daniel]]></dc:creator>
            <pubDate>Wed, 17 May 2023 16:21:06 GMT</pubDate>
            <atom:updated>2023-05-17T16:39:45.226Z</atom:updated>
            <content:encoded><![CDATA[<h3>KPMG virtual data analytics internship (Part 2)</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/528/1*akwXCvqbTo_ztvNyed20-w.png" /></figure><p>In this article, I will be discussing modules 2 and 3 of the internship. If you haven’t seen the first article that explains what the internship is about and the details of module 1, I advise you to check it <a href="https://medium.com/@modudharniel16/kpmg-virtual-data-analytics-internship-a28f9f8009b">HERE.</a><br> Follow me in the step-by-step process of how I tackled modules 2 and 3.</p><p><strong>Module 2 — Data insights</strong></p><p>The objective of this module was dependent on my results from module 1. I was asked to analyze the now cleaned datasets (from module 1) by discovering patterns and insights and in turn, recommend which of the 1000 new customers (in the new_customer dataset) should be targeted to drive the most value for the organization. (SPROCKET LTD.)</p><p>The three phases involved in this module were;</p><p>· Data Exploration</p><p>· Model development</p><p>· Interpretation</p><p><strong>Data exploration</strong></p><p>To begin data exploration, I found the transactions dataset to be very useful as most of its fields could be analyzed to derive insights.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/1*h-CYnCiZNUUePdhPtehLwQ.png" /><figcaption>Transactions dataset</figcaption></figure><p>In order to make the transactions dataset more useful, I decided to join some fields from other datasets that would be helpful in my analysis. I did so in order to have all the data I am working with, in a single sheet and not multiple locations. To do so, I had to use the VLOOKUP function to insert other relevant fields from <strong>customer-demographic</strong> dataset like Name, Age category, Job industry, etc.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/1*wpVh-W213y7NKIV3qj5ESg.png" /><figcaption>Image showing how I converted date of birth to age category</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/1*ouGFTt0v3d5o3baAN4VXmQ.png" /></figure><p>Image showing some fields already entered into the transactions worksheet with the VLOOKUP function in the formula bar. Such that, columns N, O, P, Q …….. were imported from the <strong>customer_demographic</strong> dataset. The mode of use of the function was by using the <strong>customer_demographic</strong> dataset as the LOOKUP table and using a specific reference that was unique and linking both datasets (transactions and customer_demographic) together — -&gt; ‘Customer_id’ (just the way primary and foreign key work).<br> It could have been done faster using SQL or Power Query.<br>Having completed the transactions dataset, I could now begin my data exploration to uncover trends in my data.<br> I created an extra column called ‘Profit’.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/557/1*_3fZb4pD1Vqt-1B_XYL9dA.png" /></figure><p>It might be baffling why the formula in the formula bar for profit looked like that. The F column was for order status (approved or canceled order) and I wanted to calculate the profit for just approved orders. <br> The formula would first check if the entry in F2 was approved, if it was, it would subtract the <strong>standard_cost </strong>from the <strong>list_price</strong> and if the entry was not approved (i.e. canceled), it would return 0.</p><p>The next step was creating a pivot table as it was the fastest way to filter the data and discover trends. Most insights were uncovered from this process.<br> To begin, I highlighted the entire data and created a pivot table in several sheets. <br> From the results I got, I had them represented in pivot charts which would be collated to make my dashboard later on. Some of the pivot charts I was able to come up with are shown below;</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/601/1*Dpu3NLIjenKjGNktkQKKLw.png" /><figcaption>A line chart showing profits for different months.</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/601/1*sktmbvpAldyAKRrQsCmWLQ.png" /><figcaption>A doughnut chart</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/1*uyA7UXQTC4XcB1K_bPkSMg.png" /><figcaption>A column chart showing profits by various age categories.</figcaption></figure><p>The charts were able to show me the customer segment, the age category, the state, and the job industry that contributed significantly to the company’s profit.</p><p>I also had to account for other factors by involving statistical approaches. For instance, I wanted to find out if there was any relationship between profit and the number of previous bike-related purchases of customers in the past 3 years. To do that, I employed linear regression.<br> <strong>why Regression?<br> </strong>The Y-variable here was profit and the X-variable was the number of bike-related purchases. Both were numerical values and hence the need for linear regression.<br> <strong>The Result:</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/1*5SJEI_fUanr0bkP8zkM5lg.png" /><figcaption>A regression analysis.</figcaption></figure><p>The P-value was below 0.05 (0.00018) Hence, there was a relationship between both variables but the R-square value was below 80% (0.46%) showing that the X-variable was not a major factor affecting the Y-variable (profit).</p><p>Furthermore, I decided to check the relationship between car ownership and profit. The statistical approach I used was ANOVA (Analysis of Variance). <br> <strong>WHY ANOVA?</strong><br> ANOVA is used when the Y-variable is numerical (profit) and the X-variable is categorical / non-numerical (responses to whether the customers owned cars or not was a categorical form of data — — <strong>Yes / No</strong> responses<br>THE RESULT:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/602/1*Wl1kWu1eCUxqLjLjnS0OHg.png" /><figcaption>ANOVA table.</figcaption></figure><p>The P-value was below 0.05 (0.04) Hence, there was a slight relationship between both variables but the R-square value was below 80% almost 0 (0.02%) showing that the effect of car ownership was not a major factor affecting the profit and so concluding that customers without cars will yield more profit by purchasing bikes and customers with cars will not yield profit by not purchasing bikes might not be all that true.</p><p><strong>MODEL DEVELOPMENT:<br> </strong>From the trends and insights I discovered, I designed an algorithm sketch that determines easily who our target customers should be from the 1000 new customers.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/584/1*dFqAARLBuX-yizX__1H3Bg.png" /><figcaption>Model.</figcaption></figure><p><strong>INTERPRETATION:<br> </strong>The analyzed data shows that there was no relationship between profit and whether the customers have cars or not. Such that, a customer with a car can buy as many bikes as a customer without a car. Also, the age category that yielded the most profit was the age group (41 -50) and they were mainly the mass customer segment. The top 5 industries contributing to high profit were financial services, manufacturing, n/a, Health, and Retail. <br> Additional data from the Australian Bureau of Statistics showed that there was an increased population in NSW and VIC majorly implying these regions will contribute more to profit as increased population results in congestion, hence, the need for bikes rather than the use of vehicles.<br> After applying these identified trends to the new customer dataset, we identified <strong>87</strong> potential customers that would definitely contribute significantly to the profit of the company in the coming years. [NB: The numbers can be further increased by widening the criteria for the selection.]</p><p>All the above were the results I arrived at after the completion of module 2.</p><p><strong>Module 3: DATA PRESENTATION</strong></p><p>This aspect focused on presenting my findings and insights on an interactive dashboard.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NC9H5p6YZa1FVVLxN9-hfQ.jpeg" /><figcaption>Dashboard.</figcaption></figure><p>This marks the end of the narration of my KPMG Virtual Internship experience, I hope it was helpful.</p><p>Thank you to everyone that kept reading. I have seen your reviews, corrections, and commendations, Thank you for all the support.</p><p>You can subscribe to get notified about my stories as I would be releasing an important article soon.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f9094a2de325" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[KPMG Virtual Data analytics Internship Experience]]></title>
            <link>https://modudharniel16.medium.com/kpmg-virtual-data-analytics-internship-a28f9f8009b?source=rss-1c3429dbd153------2</link>
            <guid isPermaLink="false">https://medium.com/p/a28f9f8009b</guid>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[internships]]></category>
            <category><![CDATA[microsoft-excel]]></category>
            <category><![CDATA[kpmg]]></category>
            <category><![CDATA[data]]></category>
            <dc:creator><![CDATA[Modu Daniel]]></dc:creator>
            <pubDate>Tue, 02 May 2023 06:02:39 GMT</pubDate>
            <atom:updated>2023-05-21T16:08:16.200Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/528/1*akwXCvqbTo_ztvNyed20-w.png" /></figure><p>I just recently completed the KPMG virtual data analytics internship and it was an amazing experience. I had applied my knowledge and experience in Microsoft Excel and statistics from the just concluded bootcamp hosted by <strong>NHIT and Octave Incorporations </strong>for Medical students to tackle the challenge.</p><p><strong>A brief information about the internship<br></strong> KPMG virtual data analytics internship is one of the biggest, free, virtual and educative internship programs which should be a must for every entry data analyst. It gives you the real-life experience as an employed data analyst and tests you on how well you can apply your skills to solve real life problems. <br> It is divided into three modules, which are;</p><p>Module 1 — Data quality assessment<br> Module 2 — Data insights<br> Module 3 — Data insights and Presentation</p><p>In this article and the successive ones, I will walk you through my journey and how I came about solution to the various challenges. This article will focus on<strong><em> Module 1 — Data quality assessment.</em></strong></p><p>For this part of the internship, we were presented with three different datasets (<strong>customer_demographic, customer_address, and Transactions)</strong> and as the title of the module implied, we were to assess the quality of the data based on specified criteria (data quality dimensions) which were;</p><p>· Accuracy</p><p>· Completeness</p><p>· Consistency</p><p>· Current (present-day)</p><p>· Relevancy</p><p>· Validity</p><p>· Uniqueness</p><p>The main objective was to assess the data quality and make recommendations on ways to clean the data and mitigate these data quality issues.</p><p>In simple terms, it was a test of my data cleaning skills. How well I could identify outliers, duplicates, errors and filter the meaningful data from those that were unimportant.</p><p>Cleaning the first dataset, I checked for duplicates (you can refer to my data cleaning article with Excel article <a href="https://medium.com/@modudharniel16/data-cleaning-with-microsoft-excel-6d18a2109409">here</a>) to ensure uniqueness of the data. The data were unique but had several flaws when reviewed wit the data quality dimensions in mind.<br> The dataset lacked completeness as only 125 out of 4000 customers had last name. Gender and job title weren’t specified by some customers (though they could be worked with as they didn’t occur as duplicates). Brands, product line, product class had a lot of blanks as well.<br> Speaking about consistency, the gender specifications in the field were not consistent. As F, Femal and Female were made to represent Female. Just F could have been used for Female and M for Male to ensure consistency. Even when representing states in the <strong>customer_address</strong> dataset, NSW / VIC could have been used throughout the data entry than mixing with New South Wales and Victoria respectively. These inconsistencies had to be replaced using the Find and Replace tool.<br> Based on relevance, the date of birth field was not very useful in its form so I had it converted to age. [ = 2023 — year (F2) ]<br> The default column also had to be deleted as it contained non-printable letters.</p><p>The field, product_first _sold_date, was in number format and had to be converted to date format.</p><p>Below is a copy of my email to the company highlighting the data quality issues and ways to mitigate them.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/640/1*FWtDpY-vZxBhxVj8odqM8g.png" /><figcaption>1</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/592/1*-_6wXq1UTSXnWNp1KI3YnA.png" /><figcaption>2</figcaption></figure><p>Other articles will be describing Module 2 and 3. Thank you for reading.</p><p>For those interested in the internship, you can sign up at <a href="https://www.theforage.com/">theforage.com</a></p><p>The article explaining module 2 and 3 is <a href="https://medium.com/@modudharniel16/kpmg-virtual-data-analytics-internship-experience-ii-f9094a2de325">HERE</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a28f9f8009b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[My 10 days of Excel Capstone Project.]]></title>
            <link>https://modudharniel16.medium.com/my-10-days-of-excel-capstone-project-d76a4f9d3e7d?source=rss-1c3429dbd153------2</link>
            <guid isPermaLink="false">https://medium.com/p/d76a4f9d3e7d</guid>
            <category><![CDATA[capstone-project]]></category>
            <category><![CDATA[excel]]></category>
            <category><![CDATA[challenge]]></category>
            <dc:creator><![CDATA[Modu Daniel]]></dc:creator>
            <pubDate>Fri, 30 Sep 2022 04:30:58 GMT</pubDate>
            <atom:updated>2022-09-30T06:03:16.744Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1023/1*xclJPj99AscbX7UGEUGGfw.png" /></figure><p>This article will be explaining my capstone project which I did at the end of the 10-day Excel challenge hosted by <em>Joyce Nwanochi</em>. I am very grateful for such opportunity as we were able to learn more about Excel as a tool for data analysis and connect with other learners.<br> Before I dive into the capstone project, I want to briefly write out some of the amazing topics we covered during the 10-day challenge.<br> We started about the basics of Excel and understanding the layouts of the software then we looked at cell referencing, functions and formulas, formatting, data cleaning, pivot tables and using Excel for simple visualizations. <br> The activities were interactive and helpful as we had datasets to practice with, articles and YouTube videos to learn more, live session and a capstone project to wrap up the entire challenge.</p><p>The capstone project was for application of all what we learnt since the beginning of the challenge. I was able to complete the project and build my first dashboard.<br>The dataset for the capstone project was a small dataset for a Cosmetics company.</p><p>The tasks had already been given so we were to provide answers to them by analyzing the available data but before the analysis, I decided to clean the data; remove duplicates, check for blanks, ensure data type uniformity in each column to make sure I was working with accurate data. <br> No duplicates were found and after doing the entire cleaning and I was ready to start providing answers.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*l9_-jWuE8ZvpYlJy0ktL4w.png" /></figure><ul><li>The tasks to be completed shown above</li></ul><p>I added four more columns (which have the green fill) to put my answers and make the dataset look more understandable.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/985/1*jitadHIwy-ij_UecgCFCbw.png" /></figure><p>The products column had the product category and the product code, the price column showed the price of each cosmetic product they purchased in dollars, the client column contained the names of the clients, the client code showed the unique code for each client’s purchase, the order showed the number of orders of a particular product by a client.</p><p>1.) To get the product code from the products column, I used the LEFT function. Since the product code was a consistent five digit for each row, I saw that to be an effective way of separating them. Writing the function into Product_code column, H2 specifically as =LEFT(A2,5). The preset command means that the function should go to the content of cell A2 and write out the five numbers/characters counting from the left. After doing that, I filled the formula down to the rest of the row.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/809/1*duW2TOFSWSk4XogyG9HDAA.png" /></figure><p>I did a similar thing in the product category column so as to return the category for a particular product. The products category were; Masc for Mascara, Exfo for Exfoliator, Eyel for Eyeliner, Foun for foundation and so on. To write out the product category, in cell G2, I wrote out the function, =RIGHT(A2,4). Using the same explanation above, the product category were four letters at the end of the product and the function would return the four characters from the right of the value in A2.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/687/1*Gb_T5LNbPseBeJbVhOeF5g.png" /></figure><p>Having done those, I successfully split the product into the product code and the product category column.</p><p>2.)Using VLOOKUP, I retrieved the exact matches to product from sheet 2 to get the product name. It required bringing the product names from sheet 2 to sheet 1. The VLOOKUP was an excellent choice to match the product column values in sheet 1 and return the product name from sheet 2 as a suitable array was made in sheet 2 consisting of just products and product name.<br>In the column for product name, I entered the VLOOKUP function as shown below with the syntax =VLOOKUP(lookup_value,table_array,col_index_num,range_lookup)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/996/1*mAoA0TD4hmmGyeZ2bJ_taA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/302/1*xqFG6i5Zk8WqLVoRbHCf2w.png" /></figure><p>The second image is sheet 2</p><p>In cell I2, under the product name column, I wrote the VLOOKUP function as shown in the formula bar above (image 1). Following the aforementioned VLOOKUP syntax, the lookup_value was the value in A2(written as products), the table_array was the range in sheet 2(as shown in the second image) which is where we want to retrieve information from (the range has a fixed reference so that when I fill the function downwards, the table_array from sheet 2 doesn’t change as the lookup_value changes), the column index number is the column whose value we want to be returned and that is column number 2(product name column in sheet 2 as seen in image 2), the range_lookup was to choose an exact match or approximate match(false signifies an exact match).<br> The function correctly returns the product name in column I and the function is filled downwards.</p><p>3.) The total amount is gotten by just simple multiplication of the values in column B(price) and E(order). E.g =(B2 * E2) will return the total amount for the first product sold in F2.</p><p>4.) To create a pivot table, we select a cell in with data in the table &gt; insert tab &gt; tables ribbon &gt; pivot tables &gt; select new sheet &gt; input desired cell location &gt; ok.</p><p>i) To find the top 5 products by order, from the field list, we drag the product name into row label and the orders into values. The value automatically returns the sum of the orders for each product. Sort the sum of orders in descending order and use filter to show only the names of the top 5 products.</p><p>The column chart depicts the information.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jyUMWdNHp54eMyI5jYb-JA.png" /></figure><p>ii) After creating a new pivot table in another sheet, I dragged product category from the field list into row labels and the orders into the values. Applying sort filter returns the top 3 product categories based on the sum of orders. The column chart depicts the information.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CKPE9KSFrn54Sfy0RUZULg.png" /></figure><p>iii) Creating a new pivot table on a new sheet, dragging the product name and the total from the filed list to the row label and values respectively, then sort and filter produces the top 5 revenue generating products. The bar chart below depicts the information.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*g2TgLoaHcjgHNZg3ItU5Gw.png" /></figure><p>iv) For client with the highest order, the client is put in row labels and the orders in the value area. Applying sort function in descending order arranges the clients with the top order to the client with the least. The column chart below depicts the information.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vNBOkxhMWb3g_IL_IqcHDQ.png" /></figure><p>Generally, adding the charts, I clicked on the pivot table &gt; pivot table tools options ribbon &gt; tools &gt; pivot chart &gt; select chart of my choice.<br> After editing the chart by removing field lists, gridlines, headers, the chart looks like what we had above.</p><p>To create my dashboard, I resized the charts and gathered them individually to the dashboard sheet. After the resizing and alignments, I arrived at this;</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/720/1*vx87ragBtopogzdNSEsUDA.png" /></figure><p>From the Analysis, we could make some data-driven conclusions like</p><p>· The So-soft Exfoliator generated more revenue for the company.</p><p>· Rudiger Pharmacy was the client with the most orders.</p><p>· The most ordered product was the So-soft Exfoliator.</p><p>· The product category with the most order was Foundation then Exfoliator.</p><p>These conclusions can guide the business decisions and enable them develop a smart business strategy.</p><p>Thank you for reading.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d76a4f9d3e7d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Data Cleaning With Microsoft Excel]]></title>
            <link>https://modudharniel16.medium.com/data-cleaning-with-microsoft-excel-6d18a2109409?source=rss-1c3429dbd153------2</link>
            <guid isPermaLink="false">https://medium.com/p/6d18a2109409</guid>
            <dc:creator><![CDATA[Modu Daniel]]></dc:creator>
            <pubDate>Wed, 28 Sep 2022 14:40:14 GMT</pubDate>
            <atom:updated>2022-09-28T15:07:38.683Z</atom:updated>
            <content:encoded><![CDATA[<h3>Data Cleaning With Microsoft Excel.</h3><p>This article is meant to explain how i used Microsoft Excel to convert dirty structured data into clean data so insights can be generated easily.<br>The project is guided by my scope of work which i would share below. The scope of work is an outline of all the activities i did in my data cleaning project including the deliverables, milestones, timeline. I intend to use this article as the final aspect of the scope of work, Report.</p><p>Following the scope of work, the first thing i had to do was collect the data I wanted to clean and being specific, the data i needed was an healthcare dataset which i got from ForesightBI ( name of dataset: Hospital data with mixed numbers and characters).<br>The scope of work showed the specificity of the project and what the project did not include.</p><p>After the data collection, i could then start cleaning the data. This was quite a simple dataset to clean as i didn’t use more of complex functions but i had to do more of observations to identify some mistakes in the data entry.<br>Before i started the cleaning, I saved the unclean data as two different documents with different names so incase i made a mistake in my cleaning and I needed the dataset in its default form, i would just refer to my backup.</p><p>To start my cleaning, I added filter buttons to all the column labels.<br>Select all the column headers &gt; home tab &gt; Editing ribbon &gt; sort and filter &gt; click filter.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DxDpEFjWKU1X5FbdtzNsQg.png" /></figure><p>Next, I selected all the data and removed duplicates. To do that, you select all the data &gt; click data tab &gt; click Remove duplicates on the data tools ribbon &gt; select all columns &gt; ok.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*o5G2_Xr9olwxNug7WixbVg.png" /></figure><p>After removing duplicates, i started checking each columns individually for errors using the filter button at the top.<br>The first column (Description) took the most time as the mistake were in the wordings and measurement values and required careful observation. For example, Under the description;<br>- 5% Dextrose Saline 500mls was a mistake during data entry. the intended description was 5% Dextrose Saline 500ml. (same thing for the 5% Dextrose Water 500ml).<br>- Another mistake was the use of abbreviations and then full words when representing some drugs. Like Inj Vit. K and Inj. Vitamin K meant the same thing so the error just allowed for unnecessary multiple entries.<br>- Under the description, some outliers were also found like <em>name of pharmacist, signature and phone number </em>which were out of context as the column was meant for names of drugs.<br>- other mistakes were in spellings. like the Syr Brocholite (correct: Broncholyte), hand glove and hand gloves e.t.c.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/557/1*ualjHHYSoImMVG3r5u56zw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/345/1*iV-L_ibPTXI_sIC-vP4brg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/455/1*nYrBFFt7k69xdbOvZTW42A.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/455/1*cY7r74q_nSu0fF8bLDF8eA.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/455/1*LsDYgyDJwG9tBNbIa5jVqQ.png" /></figure><p>For other columns that contained numerical data, the mistakes were just interchanging numbers with letters. for instance, 1 was replaced with i, 0 was replaced with O, S was replaced with 5. I corrected those minor errors and the data was clean and ready for use.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DxDpEFjWKU1X5FbdtzNsQg.png" /></figure><p>Recall, the project just focused on data cleaning not using the cleaned data for analysis.</p><p>Thank you for reading.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6d18a2109409" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Technological Solutions for the Omega Energia Website to Improve Customer Purchasing Experience.]]></title>
            <link>https://modudharniel16.medium.com/technological-solutions-for-the-omega-energia-website-to-improve-customer-purchasing-experience-13516c113fad?source=rss-1c3429dbd153------2</link>
            <guid isPermaLink="false">https://medium.com/p/13516c113fad</guid>
            <dc:creator><![CDATA[Modu Daniel]]></dc:creator>
            <pubDate>Sat, 24 Sep 2022 16:17:07 GMT</pubDate>
            <atom:updated>2023-05-26T01:15:52.185Z</atom:updated>
            <content:encoded><![CDATA[<p>This was the question faced to participants who chose the retail and digital commerce expertise track in the Tetrix challenge.</p><p>Omega Energia is a Brazilian Energy producing company established in 2008 for the sole purpose of providing clean, renewable energy for the people at a cheaper rate. It supplies several large organizations like Air Liquide, Heineken, Bayer etc with renewable energy.</p><p>The energy is purchased on the Website (omegaenergia.com.br).</p><p>In order to provide technological solutions to improve the customer purchasing experience, we need to first identify their problems with the website.</p><p>As a novice in the data analytics field, I began with the first phase of data analysis which was identifying the problem to be solved and to know these problems I decided to conduct a questionnaire and gather data from Brazilians to avoid arriving at a bias conclusion (because I could just simulate the purchasing experience on the website but that would lead to an unfair conclusion).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/627/1*phKgRjqg6OQxRgb-20uONw.jpeg" /></figure><ul><li>The above shows the questionnaire I organised to gather information from potential customers.</li></ul><p>After days of asking around on several social media platforms, it turned out that the Brazilians who had participated in the survey never heard of Omega Energia or were using a different Energy company. Like the picture below , he uses Celpe rather than Omega.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/720/1*DBVAxp8mmS3nYL-bfidOSA.jpeg" /></figure><p>Due to the short time I had for the challenge, I couldn’t continue the survey so I decided to simulate the purchasing process myself.</p><p>Simulating the purchasing experience as a customer on the website, I discovered that the website didn’t work quite well with my Google language translator tool. It would return to its default language (Portuguese) when I move to a new page which was quite inconveniencing as I am an English speaker. I believe that must have also been an issue for customers whose official language wasn’t Portuguese.</p><p>Since that was a problem, re-developing the e-commerce platform to suit diverse languages using a compatible language translator could help improve customer purchasing experience for those battling with language barrier.</p><p>Another problem was the requirement of numerous log in details. Portuguese speaking individuals that might have understood Portuguese would still have found energy purchase hard as not all customers might have access to their CNPJ of companies, making Business to Consumer (B2C) impossible. The companies stand as third parties. To overcome this, a website for direct customers could be built separately from that of companies requiring fewer details like just email and password for access to their customer information.</p><p>If Omega can address these issues, customer purchasing experience might improve considerably on the e-commerce platform.</p><p>*These recommendations and conclusions were done based on my individual simulation and not based on customer survey data gathered so it might be bias.</p><p>#TetrixChallenge</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=13516c113fad" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>