After three career fairs and the start of Reignite!, I am exploring more ways to use my data science skills. My first lessons in business intelligence are helping me connect business questions with the decisions a dashboard needs to support.
In This Post
- A Busy Week of Career Fairs
- Time to Reignite!
- Data Analytics vs. Business Intelligence
- From Ask to Act
- Making Better Questions with SMART
- What I’m Taking Away So Far
- What’s Next
A Busy Week of Career Fairs
This busy week actually started the Friday before at CMU’s STEM Career Fair. I spoke with several companies and learned more about the ways data is being used across Michigan industries. One conversation that stood out was with Consumers Energy. Their focus on “Safety First” and their use of GIS and environmental data caught my attention, especially because I already have experience working with geospatial data in my own projects.
I also reconnected with Auto-Owners Insurance. I interviewed with them for a Predictive Modeler position in May and am still very interested in the opportunity. I recently submitted a new application for the role and enjoyed another chance to talk about my interest in predictive modeling and insurance.
 |
| My notebook for notes on career development, technology, and continuous improvement. |
On Tuesday, I traveled to Ann Arbor for Tech Homecoming, where I learned more about the area’s technology community. I met Jack from Visto360, who encouraged me to apply for their Junior AI Engineer position. Applied machine learning is one of the areas I am most interested in growing into, and I would love to work more with techniques such as neural networks while solving real-world problems.
Then, on Friday, I attended SVSU’s All Majors Career Fair. One of my most interesting conversations was with Nexteer, a company focused on automotive and manufacturing technology. They do not currently have a data science opening that fits me, but the team seemed interested in my data science and machine learning background. It gave me another company to watch for future opportunities.
I was also surprised by some of the industries where I found potential data opportunities. I spoke with The Andersons, an agriculture-focused company in Ohio, and learned more about the Michigan Department of Agriculture and Rural Development. Growing up on a ranch gives agriculture a personal connection for me. Combining that background with data science was something I had not seriously considered before.
After three career fairs in just over a week, one of my biggest takeaways was how much more comfortable I became talking about myself. Repeating my elevator pitch, explaining my projects, and answering questions about what I want to do helped it feel increasingly natural instead of rehearsed.
I left excited about several possibilities, including some I had not considered before the week began.
Time to Reignite!
Alongside the career fairs, I officially started the Michigan Council of Women in Technology Foundation’s Reignite! program. Reignite! supports women who are returning to the workforce or transitioning into technology through career development, networking, and professional learning opportunities.
One part of the program is completing a Google Professional Certificate. There are several options, ranging from UX Design to Project Management, and I need to complete at least one within the next four months. I decided to start with the Google Business Intelligence Professional Certificate.
The BI certificate builds on Google’s Data Analytics material. Since I already have a degree in Data Science, I felt that completing the entire Data Analytics certificate would repeat too much of what I recently studied. BI gives me more new material while encouraging me to revisit familiar concepts from a business perspective.
I have already found myself going back to the Data Analytics videos and glossary when I need a refresher. Hopefully, I will not need to backtrack too often, but I do not want to rush past a concept just because I have seen something similar before.
I already have some experience with Power BI, and the certificate will give me more practice with Tableau. That will likely bring another learning curve and a little more review along the way.
Because the BI certificate is relatively short, I also plan to complete the Google Advanced Data Analytics Professional Certificate afterward. Its seven courses go further into statistics and machine learning, so it should refresh skills from my degree while giving me another structured portfolio project to build on.
So far, I have attended my first two Reignite! meetings. One of my favorite parts has been hearing about the goals of the other women in the program. Everyone brings a different background and a different idea of where they want technology to take them.
Our first face-to-face meeting is coming up, and I am looking forward to meeting everyone in person.
Data Analytics vs. Business Intelligence
One of the first concepts I had to understand was the difference between data analytics and business intelligence. They overlap, but the course helped me see a useful distinction: data analytics can answer a specific business question, while BI emphasizes systems that keep important information available for ongoing decisions.
That monitoring can use live data or information refreshed every few minutes, daily, or weekly. For example, an insurance analyst might ask, “Why did claim volume increase last quarter?” A BI system could track claim volume and other important measures over time so decision-makers can see when something changes and respond.
From Ask to Act
The data analytics process I am reviewing follows six steps:
Ask → Prepare → Process → Analyze → Share → Act.
We define the business question, gather relevant data, clean and organize it, analyze it, communicate the findings, and use them to support a decision.
Using the insurance question from the previous section, we would first identify the data needed and combine relevant sources. During processing, we might standardize formats, handle missing values, and check for data quality issues. In the analysis stage, we could compare claim types, locations, or time periods using statistics and visualizations.
We would then share the findings with stakeholders such as claims or operations managers. Depending on what we discovered, they might adjust staffing for a high-volume period or investigate an increase in a particular type of claim.
Keeping the Question Alive: The BI Cycle
Business intelligence takes many of these ideas and keeps them running over time.
The BI cycle is Capture → Analyze → Monitor.
Capture brings in the data needed for the business problem, Analyze turns it into useful information, and Monitor tracks new data so the organization can respond as conditions change.
For our insurance example, a BI system could continue tracking claim volume, average claim severity, and processing time. When a measure is tied to an important business goal, it can serve as a key performance indicator, or KPI. For a team working to settle claims more quickly, processing time could be one such indicator. A dashboard could track it and alert a claims manager when it moves outside an expected range.
That ongoing monitoring is the part of BI I am starting to understand more clearly.
Making Better Questions with SMART
The first step in data analytics is asking the right questions. One framework for doing that is SMART: Specific, Measurable, Action-oriented, Relevant, and Time-bound.
A weak question might be, “Why are claims high?” It is too broad to guide a useful analysis. A stronger SMART version would be:
“Compared with the previous quarter, how did claim volume change by category last quarter, and which categories contributed most to the increase?”
This question is Specific because it focuses on claim volume by category. It is Measurable because claim counts can be compared. It is Action-oriented because the results could help identify where further investigation or operational changes are needed. It is Relevant to understanding changes in claims activity, and Time-bound because it defines the periods being compared.
SMART questions give the analysis a clearer direction before the data work begins.
What I’m Taking Away So Far
Starting this certificate, I have begun to realize how much work sits behind a useful dashboard. I am looking forward to working with Tableau, but I also want to understand why a particular measure belongs on the screen and what someone should be able to do with it.
For example, the total number of insurance claims may matter, but by itself it does not explain what changed. Comparing claim categories or previous periods gives that number context. Looking at severity and processing time answers different questions about cost and workload.
So far, my biggest takeaway is that BI is as much about understanding the business problem as it is about building the dashboard. I want to carry that thinking into my own projects.
What’s Next
For now, my main goal is to keep moving through the Business Intelligence certificate while making sure I understand what I am learning. I am taking notes, reflecting on the concepts, and quizzing myself instead of focusing only on finishing quickly. I am already more than halfway through the first of four courses.
My job search is moving forward too. In two weeks, I have a phone interview with United Wholesale Mortgage (UWM) to explore possible IT opportunities. I am interested in learning what that could mean for future data analytics or related roles within the company. I will also continue following up with the people I met at the career fairs and applying for positions that fit my interests.
At the same time, I am reading Software Engineering for Data Scientists and applying what I learn to a new end-to-end machine learning project focused on predicting whether patients with diabetes will be readmitted to the hospital within 30 days of discharge.
The project gives me a chance to practice the technical side of data science while thinking about how a prediction might be used. I am interested in whether readmission risk could help identify patients who need additional support or follow-up care. That connects the modeling problem to decisions that matter to hospitals, health systems, and insurers.
There is still a lot I want to learn, but the pieces are starting to connect. The career conversations are giving me more ideas about where I could contribute, and the certificate is helping me think more carefully about the problems I want my work to solve.
Comments
Post a Comment