Who the Hell Are Business Intelligence Analysts?
A Bite-Sized Explanation of The Data Role That Is Rising in Popularity
When I just graduated high school in 2019, I didn’t know that Business Intelligence, or BI, existed. Many of my friends often ask me what exactly is the career that I’m pursuing and every time I answer, they will be more confused about what it is. Therefore, I decided to write this article based on my experiences and research to help those interested in this field but are still not sure what BI is.
Definition
Google ‘Business Intelligence’ and you’ll find keywords such as: ‘transforming data into insights’, ‘analytics’, ‘data visualization’, and ‘making data-driven actionable business decisions’. Before I was a BI intern, I couldn’t imagine the day-to-day tasks after reading these textbook definitions. But, lucky for you guys, I recently found an interview from Crystal Widjaja, the Former SVP of Business Intelligence at Gojek, who defined it beautifully:
Business Intelligence involves building data foundations, visualizations, and tools for other people in the company to make strategic decisions independently. We also constantly look for correlations in data to improve the business by being creative and exploring supposedly unrelated data points.
— Crystal Widjaja
The output of BI tasks can be cleaned raw data, data visualizations or dashboards, analysis findings, recommendations, and self-served ad-hoc tools.
What does a Business Intelligence Analyst do daily?
I’ll provide a typical job scenario to make the day-to-day job easier to understand. Say that you’re working in a small e-commerce startup. The marketing team wants to know which cities should be the main focus of an upcoming campaign focusing on fashion products. To increase market share, they want to increase brand awareness in the top-performing cities. As a Business Intelligence, you’re tasked to provide the marketing team with relevant data to help decision-making.
Here are the typical steps that you’ll go through:
- Seek the tasks’ purpose, needs, and expected output by liaising with the business user. In this scenario, the primary goal is to increase the company’s market share by elevating the brand awareness in top-performing cities, hence the output can be a simple list of the top-performing cities. If you want to dive deeper, the top fashion products in the respective cities can also be analyzed.
- Assess the availability of the required data to deliver the output. BIs need to understand the company’s data structure, as to how and where the data is sourced and stored. For this example, transaction, inventory, and customer data are the fundamental data needed to analyze the top-performing cities for fashion products.
- Clean and extract the data you want and combine them together using the query language such as SQL, a programming language to retrieve data from databases, since, in most cases, a tremendous amount of data is stored in different tables. In cases when data is not that huge, Microsoft Excel is more than enough to process and visualize the insights.
- After successfully going through the painful query process (just kidding), the fun part of analysis and visualization comes in. Utilize your go-to visualization tool, such as Power BI or Tableau, to choose the proper visualization and make it easy for business users to understand. You can go with a sorted bar chart to display the top 10 performing cities for fashion products in this case.
- Voila! You’re set to present your finding to the marketing team and promote the data-driven decision-making culture.
Do note that this scenario is a simplified illustration and is used for educational purposes only :).
Besides taking requests from business teams, BIs also have the opportunity to do their analysis and self-exploration, in addition to training business teams on how to use the BI tools (query and visualization) to be independent data people.
Business Intelligence (BI) vs. Data Science (DS)
Since both BI and DS are data roles, their difference is often asked about. In short, BI focuses on the past and present, answering the question of ‘What has happened’ to make actionable decisions. On the other hand, DS looks towards the future, to answer ‘What will happen’ and predict outcomes or potential risks. However, the two disciplines are correlated to each other since the BI process is the first step to take before applying DS methods. In order for DS to do exploratory data analysis and create models, the present and historical situation needs to be assessed first.
Back to the top-performing cities example above, BI illustrates the current and past state of the cities’ performances, and based on the insights, DS can build machine learning models to predict which city will perform the best next month.
Conclusion
Business Intelligence is a mind-provoking field where we have to always be creative throughout the process of turning data into valuable insights since the data can be very vague and unrelated. However, there is a different sense of satisfaction when the insights you generated impact the users!!
For any inquiries, I can be contacted via LinkedIn. Last but not least, I hope this article can act as an introduction to the BI field :D
