How Microsoft Power improves my data analytics and visualization workflow
especially those in new fields, are often believed to be that data science is simply collecting data from databases, using algorithms and deploying models.
But, not only that. Data analytics and visualization are important aspects of data science that help you understand complex data, understand it and create actionable insights.
In my early days in data science, I never saw the need for data visualization because I was not exposed to and familiar with the knowledge and appropriate tools to effectively solve visualization tasks.
I still remember spending hours burying on Excel sheets, manually updating the pivot table and endlessly adjusting the layout of the charts, just to build something that still doesn’t tell the story I wanted.
Don’t get me wrong, Excel is great, but sometimes it doesn’t cut it.
As a computer science major with an increasing interest in data, I knew there had to be a better approach – but I don’t know what that is yet.
My first real struggle was in a college program and I had to analyze student performance data over multiple semesters.
I know what you’re thinking; that should be easy.
OK, yes, yes.
But for me, not.
I have a row on scores, attendance, course code, etc., but turning all of that data into meaningful insights is like trying to teach Tony Stark to stay modest.
I tried everything: Excel formulas, conditional formats, and even dabbled with Matplotlib to generate some plots. No clicks; it’s overwhelming.
That was mentioned by a senior colleague Microsoft Power BI.
For those who don’t know, Power BI is a data visualization and business analytics tool developed by Microsoft that allows you to connect, transform, analyze and, most importantly, visualize your data.
At first, this sounds like another tool in a list of software that I haven’t quite mastered yet. So I had to do some personal reading.
I mastered the “Power BI Recipes: Creating Business Intelligence Solutions for Analyzing Data Models, Reports, and Dashboards,” a book written by Brett Powell, and it’s the beginning of something more granular.
This is not only a book, but it is more like a comprehensive guide to understanding the entire concept of creating interactive visualizations using Power BI.
After a few days of learning about the operation of Power BI, I have imported the dataset, cleaned it with Power Query, and built my first Interactive dashboard.
For me, I think it’s more than just a technological upgrade, it’s a mentality change that I don’t know I need to move forward. It changed my view of the data itself.
Moving forward this article, I will share the powerful Power BI that helps me in the process of data analytics and visualization, as well as personal stories and actionable takeaways that can help you develop your professionalism as a data scientist.
The day I stopped copying and started living
Yes, this is a breakthrough.
When I started analyzing the data, my workflow looked like a messy relay race: I opened an Excel file, copied the data, opened a new window, pasted it onto another piece of paper, crossed my fingers, and prayed to the sky without any breakage.
Guess what it is, it always breaks.
After copying and pasting one file to another, my folder is filled with files named something like this Sales_Q4_FINAL_final2.xlsxbut I still can’t track everything.
Power BI’s ability to get data from anywhere, database, spreadsheet, or even cloud services means I no longer need to play Data Tetris. With just a few clicks, I connect Excel tables, SQL databases, APIs and even my locally stored data files.
Don’t worry about whether you have a challenge to import datasets or not only work as you expect. It’s easy, trust me, you just need more practice.
Play with the dashboard and learn what buttons can be used and how to use them. You will be satisfied when you find your way out.
The first time I saw all the data updates, I just sat down and smiled. No copying, no mess, just clean and connect data.
Intuitive visualization with custom options
Like I said at the beginning, most people underestimate the power of good visual effects, especially when dealing with data. I think this is ridiculous because to be honest, the raw data doesn’t always tell stories.
According to a study published in the Journal Information visualizationpeople process visual effects 60,000 times faster than text.
If that doesn’t do it for you, even MIT suggests that the human brain can recognize images that are only 13 milliseconds.
In fact, these studies mean that your dashboard visuals are being absorbed and explained, and then someone finishes reading the chart title and even looks at the numbers you spent hours on.
My favorite Power BI features must be interactive and advanced data visualization. With its intuitive drag-and-drop interface, you can turn the darkest (as much data as I like, and sometimes it looks boring) into a dynamic dashboard.
With various visualization options, the range is:
- Matrix and table visual effects
- Instrument and KPI visual effects
- Slicer and filter
- Decompose the tree
- Waterfall map
- Map visual effects
There are many more, but I think these are my personal favorites.
Data scientists and analysts need the ability to successfully interpret data, identify trends and help businesses make better decisions.
As a pioneer in computer science, Ben Schneiderman correctly said:
“Visualization gives you the answers to questions you don’t know about yourself”
Power Query: The Silent MVP behind My Clean Data
You may ask, what is a power query?
Power Supply Query It is a data conversion wizard built into Power BI. This is a great feature that allows you to clean, reshape and prepare data before loading it into your model for analysis and visualization.
I think it is the engine that powers data preparation in Power BI.
The data is very confusing. That’s just part of the job. In addition, as companies and businesses expand, more and more data is being collected. It is very challenging for most data scientists and analysts to master a lot of raw data.
Remember the challenges I faced in my college program?
As it turns out, one of the reasons I found it difficult to do analysis is that my datasets are all confusing.
I was asked to analyze student performance, which took data from three different CSVs, each with quirks. One has an admission code instead of a name, the other uses an inconsistent date format, and the third uses the course title listed in all hats (screaming to me).
With power query, this is how I build a complete workflow:
- Replace the admission code with a readable name
- Convert date format
- Standardized text format
- Combine everything into one organized table
Data preparation accounts for up to 80% of data analyst time. Imagine how much time you will save and how productive you will become when you focus all your time and brain power on generating better insights. Time recycled coffee, yes, a real analysis.
Collaborative Sharing and Cloud Accessibility
I believe collaboration is a key player in the data science industry, and that’s why: Often no one has all the expertise needed to bring a project from raw data to a real real-world project.
Stay with me.
Consider a process of data science. It involves collecting data, storing it in a database, and creating algorithms and models to improve data quality, analytics, visualizations, and other essentials.
To effectively process data, these phases are often handled by a variety of professionals specializing in different fields, all working towards a common goal. Therefore, collaboration.
Power BI is a cloud-based platform that allows you to publish and share analytics reports with other data professionals.
Instead of emailing the Excel file (I’m sure we all did it once or twice), I’m able to publish the dashboard and share a live link with my team. They can make changes, share their ideas, and even update data sources in real time.
In a remote/hybrid work world, doing this seamless collaboration is a real game changer for data scientists.
Suitable takeaway
If you have tried data analysis and visualization before and find it difficult to understand, maybe you are not using the right tool.
Not only did Power BI help me solve the problems I had when I first started using data, it also changed the way I processed the data completely.
Most of us are already familiar with Power BI, and it is a new adventure for others. Whatever category you fall into, I strongly encourage constant learning of the tool and how to maximize its functionality.
I highly recommend you check out Guy on a cube on YouTube who teaches Power BI through his info video.
For verbal learners, you can get a lot of information from Brett Powell’s book. I mentioned it in the introduction, and personally, it is the best book I’ve read about data visualization.
Get familiar with these features and start improving your data analytics and visualization workflow.
