Trying to keep up with teenage slang is an incredibly futile task. While the emotions and themes of teenage life may seem to stay pretty consistent, the words and acronyms they use certainly do not. If you find yourself trying to ingratiate yourself into teenage culture, you’ll likely need more than a pocket-sized Merriam-Webster dictionary to keep up and fully understand today’s lingo.
Regardless of what stage you’re in on your data analytics journey, you may find yourself feeling just as lost and confused. Terms like machine learning, artificial intelligence, data science, and other semantics are the analytics equivalent of the ever-growing teenage slang vocabulary. As you dive deeper into the analytics world, you’ll undoubtedly discover the three primary types of analytics: descriptive, predictive, and prescriptive.
If you feel as lost as the last time you sat next to a table of Millennials in a Starbucks, you’ve come to the right place! In this blog, we’ll define these three primary types of analytics and the questions they are designed to answer. Not only will you leave with a better understanding of each type, but also tactical examples of how to apply each one in your organization.
What is Descriptive Analytics?
Before you can walk or run, you have to learn how to crawl. Descriptive analytics is the most basic form of analytics, focusing on describing what happened in the past or present. Descriptive analytics involves summarizing data to tell a story that has already happened and is easily interpreted by any audience. The majority of analysis performed on a regular basis falls into this category. Things like percent changes, averages, and totals, whether daily, monthly, or yearly, all fall into this category. Descriptive analytics is used when you need to understand what is going on in your organization or business at an aggregate level. If we’re still trying to understand teenage slang trends, descriptive analytics will help us answer questions like:
- What were the top 10 slang phrases in 2021?
- How many of those same words were in the top 10 the year prior?
- How many words did I not know the meaning of today?
While it may not be as flashy or sexy as predictive and prescriptive analytics, descriptive analytics is critical to making key decisions based on today’s reality. Building a solid foundation on descriptive use cases will help establish the prerequisites needed for predictive solutions.
What is Predictive Analytics?
We’ve graduated from crawling, and now we’re learning to walk. Predictive analytics seeks to use mathematical models to figure out what will happen in the future. This is where we begin to see actionable insights based on data. When estimating the likelihood of a future outcome, we’re able to adjust things like price, supply, or workforce to better position ourselves. While predictive analytics can be a powerful tool, it’s important to remember that no algorithm can predict the future with 100% accuracy. Foundationally, predictive analytics is based on probabilities. Applying predictive analytics to understand trends or relationships between data can be used in almost any department of any industry, and our pursuit of understanding teenage slang is still relevant! Using predictive analytics, we could answer questions like:
- What will the top 10 slang phrases be this year?
- How many times will I hear “on fleek” this week?
Predictive analytics can be as simple as a rolling average or as complex as a neural network. They are great tools that learn from the past and can incorporate other variables to help with business planning decisions.
What is Prescriptive Analytics?
Finally, we’re learning how to run! After a forecast is generated, prescriptive analytics goes one step further and optimizes on various potential scenarios to recommend a plan of action to get the best result. At their best, prescriptive analytics solutions predict not only what will happen, but why it will happen and what actions should be taken to take advantage of the prediction. This type of analytics goes beyond both descriptive and predictive analytics by giving one or more possible courses of action. Typically a robust prescriptive model will apply algorithms, business rules, and machine learning procedures against different data sets that include historical and real-time data. In our same example, prescriptive analytics might help us answer questions like:
- What phrases can I use to stay relevant for the next 6 months?
- If “yeet” is going to lose popularity next week, when is an appropriate time to stop using it?
Prescriptive analytics is complex to utilize, and most organizations are not yet ready to use them in their daily course of business. When implemented correctly, they can help establish a sustainable competitive advantage and drive business decisions.
While each of these types of analytics can provide value to your organization; it’s important to fully understand the purpose for each one. One of the biggest missteps in building out an analytics culture in your organization is over-investing in predictive or prescriptive analytics before you’ve spent the time to describe and understand what is foundationally occurring. Think of descriptive use cases as what ‘keeps the lights on’ in your analytics culture.
In the world of teenage slang, you wouldn’t get very far without first understanding what in the world all of these new words or phrases mean. These use cases will provide a solid foundation for your organization to grow and build upon as you use analytics to solve more complex problems.
To dive deeper into how to adopt all three analytics types strategically, check out this post on analytics maturity models. If you aren’t sure what level of analytics your organization is ready for, we’re here to help. phData is passionate about helping businesses of all sizes succeed with their data and analytics journey.
The primary difference between artificial intelligence and predictive analytics is that AI is automated from end to end and requires zero manual input. The advantage of using artificial intelligence is that variables and algorithms are constantly evolving compared to being limited to a specific model in predictive analytics. AI is always learning, whether you’re there to teach it or not. When thinking about predictive analytics, consider it a foundational stepping stone in advancing towards AI, similar to how descriptive analytics paves the way for predictive analytics.
Using different types of analytics in an organization enables you to inform different users at different levels. At some levels of your organization, simple KPIs or historical trends may be enough to inform their decisions, while others may need much more proactive data to create a game plan to address changing environments in their business. Let your technical team chase the high-value, more time intensive questions with predictive and prescriptive analytics while the majority of your business users capture the low-hanging fruit from descriptive use cases.
This post was originally written by Elizabeth Dinevski.