The Data Point #11: The 4 Types of Data Analytics
Understanding the various ways of approaching data
Big thanks to this week’s sponsor who helps keep this newsletter free for everyone.
This week’s newsletter is sponsored by Udemy. Browse 1,000s of high-quality, on-demand courses that you can learn on your own schedule. Udemy was instrumental in my pivot to data analytics and continues to support my development with their expert-led, self-paced courses.
When analyzing data and fulfilling stakeholder requests, the work you’re doing will typically fall into 1 of 4 buckets.
Understanding the different types of analytics helps us to determine what kind of value we’re providing with our analysis.
Each of the 4 types also builds off of each other and increases in complexity from one to the next. Let’s dive into each one.
The 4 Types of Data Analytics
1 | Descriptive
Descriptive analytics covers the “What.” We’re looking at what has already happened or what is currently happening and describing it in detail.
An example would be looking at sales data from the previous year and providing key insights.
This type of analytics caters very well to visualization. Most dashboards you see tend to evaluate past data to tell a story of what happened.
2 | Diagnostic
Diagnostic analytics is the “Why.” Though we are still looking at past data, we’re going one layer deeper with this one.
We can examine “Why is this happening?” and attempt to diagnose the issue. This type of analysis helps us to get to the root of the problem rather than just looking at what happened.
This works well with visualization too but requires greater depth. You’ll need to dig into the data a bit more to be able to diagnose the results of the data. An example could include looking at which products brought in the greatest profit quarter-to-quarter and which ones might have taken away from profit.
3 | Predictive
Predictive analytics is the “If.” We’re starting to look towards the future now. “What might happen if we do this….”
This type of analytics takes data from the past to help us understand the future. We can provide more value with predictive analytics, but we are also getting increasingly more complex. This is where machine learning starts to come into play.
When it comes to data visualization, things like forecast charts can help us to provide predictive analysis. An example would be forecasting sales for Q4 with data we already have from Q1-3.
4 | Prescriptive
Prescriptive analytics gives recommendations. Now we’re not just looking at “What",” but rather “What should I do?”
This type of analytics is future-focused as well. It also provides the most value and is the most complex to explore. This might contain statistical methods for finding the necessary actions for a desired result.
We can give recommendations based on basic data visualization too. An example might look like recommending what times of the year to provide a sale as well as how long the sale should last, and what kinds of deals are provided by the sale. All of this is done using past data to understand the future and give a prescriptive recommendation on what to do moving forward.
Conclusion
Understanding the 4 types of data analytics helps us to comprehend what kind of value we’re providing with our work. It also helps us to determine how to approach a given task and the best way to solve it.
That’s it for this week.
See you next time ✌️
Whenever you’re ready, there are 2 ways I can help you:
If you’re looking to create a data portfolio but aren’t sure where to start, I’d recommend this ebook:
The Data Portfolio Guidebook: Learn how to think like an analyst, develop a portfolio and LinkedIn profile, and tackle the job hunt. Join 450+ learners here.
For help navigating the data job hunt, consider booking a 1:1 guidance call with me.