Speaker 1: Hey guys, Satyajit Patnaik here and welcome back to my channel. Couple of days back, I did a video on how to showcase your data science projects in an interview. And that video kind of did well and I received a lot of comments that, Sir, why don't you do the same for a data analyst role? As a data analyst or as a business analyst, how do we showcase our projects? What is the right way of explaining? And that is why we are here. I'm going to explain in this video how to portray yourself as a data analyst and how to explain your project in an interview. So let's get started. In case you haven't seen my other video, which is how to explain a data science project, you can probably check out my video which will be available here or in the description below as well. And then you can come back to this video because I have shared a lot of information, which I think will be beneficial if you go back to that video and watch it out because I'm not going to explain each and everything here. I'll directly jump into the topic that how a data analyst explains one of his project. Now just giving you a brief idea that data analytics is a part of data science and analytics domain, right? So many people, there are multiple things that a data analyst do. Now your role could, might not or might be a part of what I'm talking about, but majorly a data analyst does these things. So a data analyst works on reports, they work on creating stunning dashboards, they work on the data gathering part, data collection part and converting an unstructured data to structured data, cleaning the data and doing your initial data analysis, which is other wise called as exploratory data analysis. So these are the major work and in some cases, data analysts are also responsible for, you know, putting the data into SQL databases or MySQL databases or NoSQL databases, extracting data and all these things. So these kind of work a data analyst do. And why data analytics is a part of data science and analytics? Because it all starts from data analytics, right? And then once we analyze the data and then the next step happens, which is the model building, feature engineering and model iterations and all those things, which is done by the data science, machine learning and all those things. Now we'll quickly jump onto the topic that I am going to pretend myself as a data analyst. To be very honest, I haven't worked as a data analyst, but I know how a data analyst works. So I'll quickly jump onto this particular topic of what we are talking about, what this video is all about. So if you haven't seen my Power BI videos, you can go back to my Power BI videos. In part two, I have explained how to create a dashboard related to customer churn prediction. Now I took the same example in the previous video as well in the data scientist video. I'm going to take the same example in the data analytics video as well. So you can probably watch both of them and then you will be able to understand what data analysts do and what the data scientists do. OK, so let's say we have analyzed customer data and have created a dashboard out of it. So what I'll write in my resume is analyze the customer profile data for HGC or for XYZ and created a dashboard in Power BI explaining the customer churn reasons and showcasing the different customer attributes. This is what you can write down in your resumes and now let's say the interviewer asks me talk about yourself. Of course, I'll talk about myself and the immediate question will be explain one of your projects. So we'll quickly start and I will pretend myself as the student or the candidate and I will start answering this. So explain one of your projects which you did in your previous company or in last few months. So what I'll talk about is so in XYZ company in one of my previous companies what we did is there was one major problem which the higher management and the sales team identified that customers were churning. So the customer churn prediction sorry the customer churn percentage was going higher and higher and our major problem was to retain the customer. So that is where the data science team pitched in and they told that they will be analyzed the customer data and based on that they will be creating some sort of predictive model. Now what our role was our role was when we started collecting the data that is where my major role was I started interacting with multiple teams because in my company not just a single team is owner of all the databases. So there were multiple teams which were owning different types of data like we had customer profile data, we had customer service related data, customer complaint related data, customer usage data coming in from multiple platforms, multiple teams. We tried to collect the data and try to you know make it structured data and we basically pushed the data into our own databases and our immediate task was to combine the data and analyze the data. So what we did is we did some sort of customer profiling where we try to analyze and do some sort of ADA exploratory data analysis to understand what are the characteristics of the people who are leaving the company. And once our EDA was done we handed over our EDA task or EDA insights to our data science team so they already started working on the model building part. While they were working on the model building part our major task was to create a dashboard out of it, create a dashboard out of the customer profiles. So what we did is me and one of my colleague we were who are working in the data analytics part we created a dashboard and our first thing which was going on my head was to create a dashboard in such a way that it is appealing to the higher management and appealing to the sales team. So I tried to give a website look to the dashboard so we had a homepage where we had different buttons so that we can navigate to different pages. Our second page was a customer profile data that means the customers who have not churned that means they are active in the system. I also created another page for the churner profile that means the people who have left the company. Now we had the entire dashboard, homepage, customer profile and churn profile. We kind of did it and published it and showcased it to the sales team and the data science team. Later on when the data science team created their predictive models so they basically came out with lot of customers along with the customers they also gave us a predictive score that what is the prediction that this customer will going to churn or not. For example customer A 0.6 which basically means there are 60% chances that the customer has churned. Customer B 36 which means there are 36% chances that this customer will churn and something like that. So again that data was handed over to me so what we did is we kind of expanded our dashboard from our second and third page to our fourth page. In our fourth page what we did is we created a churner churn reasons. Now what we did is we categorized the data the predictive data which was having the predictive scores into different segments high churn risk, low churn risk, medium churn risk something like that. That means people who had a score of 0.8 and above they are the high risk customers. High risk as in they are most likely to be churned 0.5 to 0.8 medium risk and below 0.5 is basically low risk. And then we created another kind of dashboard where we talked about different you know bar graph showing high churner profile, low churner profile, medium churner profile and what is the impacted income from these kind of customers. So this is what we did and it was a very you know pleasant experience we took almost like three months to finish this project and this is how our entire workflow was and later on we kind of pitched it to our sales team to our higher management and they really liked it. So this is one of the projects which we did in our previous organization. And then there will be follow-up questions from your interviewer that oh did you do this? Why did you use Power BI or why did you use this? Did you use page navigations or did you use tooltips? So all these things you have to be ready for the counter questions. That's all about this particular video. I hope you understood how to showcase yourself as a data analyst and explain one of your projects. Now in case you are more confident on Tableau or any other thing try to focus on that part. Let's say you are more confident on Tableau try to cook up the similar story by just removing Power BI and adding Tableau. Let's say you are not confident on Power BI or Tableau just talk about the EDA part that we did EDA and we did univariate analysis, we did bivariate analysis, we did correlation. So whatever you are doing try to explain your project in like 15-20 minutes because the more time you take in explaining a project you are reducing the total interview time. So if you take a lot of time then the interviewer will be left out with just 15-20 minutes to ask rest of the questions. So don't try to summarize it in just 5 minutes or 2 minutes try to take time and if you are confident on XYZ topics try to focus more on those topics and that is one of the tips which I usually give to my students. That's all about this particular video. I hope you enjoyed it. I enjoyed shooting this video and please like share and subscribe the channel comment down below what you want from me and if you like it please like it. I also need some comments that whether you are working as a data analyst, data scientist or you are a wannabe. So whatever it is please comment down so that I understand where my subscribers are basically from and based on that I will create my upcoming contents. That's it for this video. Thank you and see you in the next video.