I wasn’t sure what I was going to get out of Applied Analytics at RMIT. Some of the subjects I had done were pretty foundational, and I suppose from a professional statistics perspective, this subject might seem the same. But for me, it was a big leap forward—not just in my technical capabilities like using R or diving into more academic research, but also in how I think about the art of what is possible in analytics.
"To me, all data and analytics is simply an attempt to capture the truth as closely as possible and then make the best possible decisions based on it."
At the heart of everything I do in data and analytics is one nebulous yet guiding concept: "the truth." Things that really happened, or will happen. To me, all data and analytics is simply an attempt to capture the truth as closely as possible and then make the best possible decisions based on it. That’s why it’s so important to get your data right—to deeply understand what your raw data is describing before you try to derive insights from it.
But this subject opened my eyes to another layer of truth and insight. As my career progresses, I've come to see BI tools not as the pinnacle of analytics but as just one piece of the puzzle. They're great for showing vanity metrics or helping colleagues explore data, but more often than not, 
"ain't nobody got no time for that." People want confidence—confidence in a few sentences that make sense and drive direction. As an analyst, you need to build trust that what you're saying is robust and as close to the truth as possible.
From Classroom to Boardroom: Practical Applications of Applied Analytics
This subject gave me not just foundational knowledge but the tools to get closer to that elusive truth. One particularly profound moment was revisiting the normal distribution—a beautifully simple yet universal descriptor of behaviour (perhaps divine behaviour?). Understanding that and applying tools like regression and predictive analytics was both powerful and deeply satisfying.
"Clean, tidy data is just the beginning—you need to deeply understand what every feature means before synthesising insights."
I didn’t just leave these concepts in the classroom. I brought them straight into the workplace. For example, we applied regression analysis to our end-of-year Culture Study, looking at three colleague surveys across the year to determine which factors influenced employee satisfaction the most. By integrating interaction terms and leveraging our domain knowledge, we were able to uncover insights into areas where factors were intrinsically linked.
Real-World Applications: Beyond University Assignments
In university, the assignments demonstrated the power of these tools in scenarios such as:
- Predicting attendance at an event.
- Analysing the efficacy of flu vaccinations using dummy data.
- Conducting A/B studies on recommendation engines for a video streaming service, identifying the groups most affected.
In the real world, these methods become even more impactful:
- Customer segmentation: Using regression or clustering techniques to identify and target specific customer groups with tailored marketing strategies.
- Operational efficiency: Predicting demand to optimise staffing levels or resource allocation.
- Product development: Analysing user behaviour to improve product features or test prototypes in controlled experiments.
- Risk assessment: Developing predictive models to identify potential financial risks or fraud.
Each of these examples underscores how these statistical techniques extend far beyond academia and into the decision-making core of businesses.
"Stakeholders don’t want to know your normalisation methodology; they just want to know your model is more accurate and faster than a human."
Building Trust Through Analytics
At the end of the day, it all comes back to building trust. Stakeholders don’t want to know the intricacies of your normalisation methodology or the hyperparameters you tuned for your decision tree. They just want to know your model is more accurate than a human and delivers results in minutes instead of hours.
This course reinforced a lesson I’ve learned through 10 years of working with data: Clean, tidy data is just the beginning. You need to deeply understand every feature you’re working with—what it means and what it’s describing—before you synthesise insights. Without that understanding, there’s no foundation for truth, and without truth, there’s no trust.
Applied Analytics wasn’t just another class for me—it was a leap forward in the art and science of finding truth in data. And for that, I’m deeply grateful.