4 Types of Data Analytics
Here's the thing about the four types of data analytics. If you are new to the field, then you're probably still wondering what they are. But if you've been around the data feild a while, then you're probably wondering why are people still talking about these? So in today's quick video, I'm going to answer both of those questions and be sure to stick around to the end, because that's where I'm going to share with you a perfect example of the four types of data analytics discussion done right. And you can use that to pattern after in your discussions of this exact same topic. For the very best data leadership in business building advice, be sure to subscribe to my channel and the belkl icon to be notified when a new episode drops each week. Also, huge shout out to the community. Thank you. Thank you. Thank you for all of the love and support and comments and content requests. It keeps me going and keeps me inspired and it really helps this channel grow. So thank you. And please keep them coming. Now, if you're new around here, then let me say hi. I'm Lilian Pierson and I support data professionals to becoming a World-Class data leaders and entrepreneurs in terms of why I'm actually talking about this topic now. I'm in the process of updating or actually rewriting my book, Data Science for Dummies. So, of course, the four types of data analytics discussion needed to get weaved in there somehow. By the way, we have a launch party coming up for the third edition, or at least that's going to be happening in September if you'd like to be part of the party, then I got the link in the description below where you can sign up and RSVP. I have been a data profressional since 2012 and been running my own data business for eight of those years. And so I've seen this conversation floating around for year after year after year, going on decade. So if you're new to the industry and then I want to share with you what are the four types of data analytics. But if you've been around a while, I want to share with you an insider perspective on how you can use that same topic idea or that same bit of education with your community, but create that context that needs to actually make that conversation meaningful instead of just a rehash of what the same thing everyone else has been saying for the last 10 years.
I'm going to be real with you, though. I'm not going to spend a lot of time defining what are the core types of data analytics just because it's been done so much by everyone across industry. And it's not the important part of this discussion, but it will cover those. So let's get to it. The primary most basic type of data analytic is descriptive analytics. They show you what happened in the past. They're based on historical data and it's looking at the past and just describing them in a quantitative sense, what happened when, how many or where. Now, this is the most basic type of data analytic. And in terms of what types of deliverables associated with descriptive analytics, that would be something like ad hoc reports or even cante reports if you're continually creating the same type of descriptive analytic report. Moving on to the next level of data analytics. So here we're talking about diagnostic analytics. So what this is, is basically just looking at the descriptive analytics on things that happened in the past and trying to diagnose why they happen that way. These are commonly used in both engineering and the sciences, and they basically answer the question, why did this thing happen and how will we know if it happens again? Diagnostic analytics are perfect for diagnosing problems in processes or subcomponents of processes. I would love to hear from you about your experience working with any of the four types of data analytics. So let me know in the comments below, and I promise I will reply back. Moving on to the third type of data analytic. That is predictive analytic. And so what these analytics are communicating is what the data analyst thinks will happen in the future if present circumstances do not change.
So if all things stay the same, what's going to happen in the future? It's a predictive analytics. So this one, as opposed to the previous to this, is forward looking. But it's not a sophisticated prediction like a data scientist would create with machine learning models. But it's more of based on the correlation analysis. These types of predictive analytics are built on top of both historical and present data sets, and they simply use mathematical and statistical methods in order to predict future events and trends. Lastly, we have prescriptive analytics. These are the domain of the data scientist. And they essentially say what is going to happen in the future if we take these prescribed course of action? Usually prescriptive analytics are built on top of a whole series of random testing, experimentation and optimization and of course, prescriptive analytics being at the top of the stairway or at the top of the analytics, work up the analytics chart. And they do generally depend on sophisticated machine learning modeling. Now, the main thing you need to come away with from this discussion on the four types of data analytics is that you cannot accurately predict the future if you don't have an accurate picture of what happened in the past. So what this means is to create accurate, prescriptive and predictive analytics, they need to be based on accurate, descriptive and diagnostic analytics first. If you're enjoying this discussion on the four types of data analytics, then there's a pretty good chance you're going to like the video I did on the five reason, the best five reasons not to become a data analyst. So we will put a link to that in the description below, as well as within the cards on this video. Alright, for the juicy part, why the four types of data analytics don't actually matter in 2021. The thing about any conversation you'd have with anyone offline or online is that you need context and relevance in order for any sort of information to have any meaning whatsoever.
So information for the sake of information just completely sucked. And honestly, there is so much noise on the Internet now in the data space that it's completely overwhelming. It could push people away from even looking in online communities for new and fresh information just because there's so many people saying the same thing. And it doesn't have a lot of context. So if everyone is saying the same thing, then finding something truly valuable and meaningful is like finding a needle in the haystack. The thing about this discussion about the types of data analytics, is you do not want to contribute to the noise. The four types of data analytics yes, they are so relevant. They're still valid. You still need to know them. All of that if your data professional and you want to have a conversation about these four types of data analytics, you got to bake in the why into that conversation. What you don't want to do is repeat the same generic thing that you've heard other people saying or that you've read from a book, just because it's not novel, it's not new. So it doesn't really need to be said again. Not only does that contribute to the noise problem within the data community on the Internet, but also it's a waste of your precious time and really that your time is the most valuable asset you have. So when you take the time to share something online, just make sure that it's actually meaningful and helpful to your intended audience. So now it's for the good stuff. I want to share with you a few examples of the types of data analytics discussion done. Right? And when I say that, I mean that these conversations are done in a meaningful way with lots of context and lots of purpose. So it's not just a rehashing of the same information, but it helps you apply that information to actually get some sort of result in in your career or in your data business if you're an entrepreneur.
So first, let's look at my buddy Ken here, the video on the four types of sports analytics projects . What I love about this video is it talks about the types of analytics, but it does it in the context of how you can do projects to attract opportunities for yourself in order to land paid work as a data analytics professional. Another thing I like about this video is it's applied to the sports domain. So any sort of discussion you're having online about data needs to be targeted to a specific avatar specific listener or someone in your audience that you intend to have receiving this information. And it needs to be targeted to their exact needs. So it can't be just a generic everyone in the world like a dictionary, like four types of data analytics. But it needs to be like three types of data analytics for sports or health care for media. And really going deep into why these types of things actually matter within the industry need to the actual listener. Right? So that is how you can get traction for the personal brand you're building using the same topic as we've been covering for 10 years, but actually get traction because it's targeted, it's targeted and meaningful to an individual in an area of application. And Ken does this really, really well in his sports analytics projects video. If you want to check that video out, I will leave a link to it in the description below. Now, another cool example I found was by a company called Retalon. Retalon is an award winning service provider of predictive analytics for retail companies. They do A.I. solutions for supply chain planning, merchandizing, inventory management, pricing optimization. And they really take a transformational approach to analytics in the retail sector. So how are they using this conversation to get traction for their business? Well, they're targeting it and they're narrowing it down to their ideal client. So in the case of a retail client, they have taken the four types of data analytics and narrowed it down to a very specific discussion on how it actually impacts the fashion industry.
And guess what? They are reaping the benefits of that specificity by enjoying a nice high ranking in Google search when it comes to analytics in the fashion industry. That, in turn, is bringing them new clients for their fashion and retail analytics business. This is really the how in the way that it makes sense to talk about the four types of data analytics in 2021. And it's simply because the data field is pretty darn mature compared to what it was 10 years ago. And so we need to keep maturing and progressing and evolving in our conversations about data analytics, about data science, in how to use these superpowers to, you know, to make the world a better place. Or if you're an employee, of course, that would be to make more profits for your company. And hopefully their mission is to make the world a better place. Now, we get to the end of this video. You can see that this video really isn't about four types of data analytics at all. It's about having meaningful, relevant conversations about analytics and data science that actually have an impact to the people who are involved. Even this video itself was meant to add value in terms of helping other data professionals get their message out to the world and get more traction in their career by being more specific and more contextual with their communications. Now, if you've enjoyed this meta discussion on the four types of data analytics, then I'm pretty sure you're going to love my data super hero quiz, which is a fun, fast Forty five second quiz that helps you uncover the ideal role for you and the day of the profession, given your personality, skill sets and passions. We'll even link to it in the description below. Also, I'd like to invite you to our Facebook group called Becoming World Class Data Leaders and Entrepreneurs. I would love to see you in there. There is a link to that in the description below as well.
Then be sure to show it some love by giving it a thumbs up. And tell me in the comments below, what area of the debris field are you working? Also, be sure to subscribe to my channel so you'll be the first to know when the next episode drops.
I'm going to be real with you, though. I'm not going to spend a lot of time defining what are the core types of data analytics just because it's been done so much by everyone across industry. And it's not the important part of this discussion, but it will cover those. So let's get to it. The primary most basic type of data analytic is descriptive analytics. They show you what happened in the past. They're based on historical data and it's looking at the past and just describing them in a quantitative sense, what happened when, how many or where. Now, this is the most basic type of data analytic. And in terms of what types of deliverables associated with descriptive analytics, that would be something like ad hoc reports or even cante reports if you're continually creating the same type of descriptive analytic report. Moving on to the next level of data analytics. So here we're talking about diagnostic analytics. So what this is, is basically just looking at the descriptive analytics on things that happened in the past and trying to diagnose why they happen that way. These are commonly used in both engineering and the sciences, and they basically answer the question, why did this thing happen and how will we know if it happens again? Diagnostic analytics are perfect for diagnosing problems in processes or subcomponents of processes. I would love to hear from you about your experience working with any of the four types of data analytics. So let me know in the comments below, and I promise I will reply back. Moving on to the third type of data analytic. That is predictive analytic. And so what these analytics are communicating is what the data analyst thinks will happen in the future if present circumstances do not change.
So if all things stay the same, what's going to happen in the future? It's a predictive analytics. So this one, as opposed to the previous to this, is forward looking. But it's not a sophisticated prediction like a data scientist would create with machine learning models. But it's more of based on the correlation analysis. These types of predictive analytics are built on top of both historical and present data sets, and they simply use mathematical and statistical methods in order to predict future events and trends. Lastly, we have prescriptive analytics. These are the domain of the data scientist. And they essentially say what is going to happen in the future if we take these prescribed course of action? Usually prescriptive analytics are built on top of a whole series of random testing, experimentation and optimization and of course, prescriptive analytics being at the top of the stairway or at the top of the analytics, work up the analytics chart. And they do generally depend on sophisticated machine learning modeling. Now, the main thing you need to come away with from this discussion on the four types of data analytics is that you cannot accurately predict the future if you don't have an accurate picture of what happened in the past. So what this means is to create accurate, prescriptive and predictive analytics, they need to be based on accurate, descriptive and diagnostic analytics first. If you're enjoying this discussion on the four types of data analytics, then there's a pretty good chance you're going to like the video I did on the five reason, the best five reasons not to become a data analyst. So we will put a link to that in the description below, as well as within the cards on this video. Alright, for the juicy part, why the four types of data analytics don't actually matter in 2021. The thing about any conversation you'd have with anyone offline or online is that you need context and relevance in order for any sort of information to have any meaning whatsoever.
So information for the sake of information just completely sucked. And honestly, there is so much noise on the Internet now in the data space that it's completely overwhelming. It could push people away from even looking in online communities for new and fresh information just because there's so many people saying the same thing. And it doesn't have a lot of context. So if everyone is saying the same thing, then finding something truly valuable and meaningful is like finding a needle in the haystack. The thing about this discussion about the types of data analytics, is you do not want to contribute to the noise. The four types of data analytics yes, they are so relevant. They're still valid. You still need to know them. All of that if your data professional and you want to have a conversation about these four types of data analytics, you got to bake in the why into that conversation. What you don't want to do is repeat the same generic thing that you've heard other people saying or that you've read from a book, just because it's not novel, it's not new. So it doesn't really need to be said again. Not only does that contribute to the noise problem within the data community on the Internet, but also it's a waste of your precious time and really that your time is the most valuable asset you have. So when you take the time to share something online, just make sure that it's actually meaningful and helpful to your intended audience. So now it's for the good stuff. I want to share with you a few examples of the types of data analytics discussion done. Right? And when I say that, I mean that these conversations are done in a meaningful way with lots of context and lots of purpose. So it's not just a rehashing of the same information, but it helps you apply that information to actually get some sort of result in in your career or in your data business if you're an entrepreneur.
So first, let's look at my buddy Ken here, the video on the four types of sports analytics projects . What I love about this video is it talks about the types of analytics, but it does it in the context of how you can do projects to attract opportunities for yourself in order to land paid work as a data analytics professional. Another thing I like about this video is it's applied to the sports domain. So any sort of discussion you're having online about data needs to be targeted to a specific avatar specific listener or someone in your audience that you intend to have receiving this information. And it needs to be targeted to their exact needs. So it can't be just a generic everyone in the world like a dictionary, like four types of data analytics. But it needs to be like three types of data analytics for sports or health care for media. And really going deep into why these types of things actually matter within the industry need to the actual listener. Right? So that is how you can get traction for the personal brand you're building using the same topic as we've been covering for 10 years, but actually get traction because it's targeted, it's targeted and meaningful to an individual in an area of application. And Ken does this really, really well in his sports analytics projects video. If you want to check that video out, I will leave a link to it in the description below. Now, another cool example I found was by a company called Retalon. Retalon is an award winning service provider of predictive analytics for retail companies. They do A.I. solutions for supply chain planning, merchandizing, inventory management, pricing optimization. And they really take a transformational approach to analytics in the retail sector. So how are they using this conversation to get traction for their business? Well, they're targeting it and they're narrowing it down to their ideal client. So in the case of a retail client, they have taken the four types of data analytics and narrowed it down to a very specific discussion on how it actually impacts the fashion industry.
And guess what? They are reaping the benefits of that specificity by enjoying a nice high ranking in Google search when it comes to analytics in the fashion industry. That, in turn, is bringing them new clients for their fashion and retail analytics business. This is really the how in the way that it makes sense to talk about the four types of data analytics in 2021. And it's simply because the data field is pretty darn mature compared to what it was 10 years ago. And so we need to keep maturing and progressing and evolving in our conversations about data analytics, about data science, in how to use these superpowers to, you know, to make the world a better place. Or if you're an employee, of course, that would be to make more profits for your company. And hopefully their mission is to make the world a better place. Now, we get to the end of this video. You can see that this video really isn't about four types of data analytics at all. It's about having meaningful, relevant conversations about analytics and data science that actually have an impact to the people who are involved. Even this video itself was meant to add value in terms of helping other data professionals get their message out to the world and get more traction in their career by being more specific and more contextual with their communications. Now, if you've enjoyed this meta discussion on the four types of data analytics, then I'm pretty sure you're going to love my data super hero quiz, which is a fun, fast Forty five second quiz that helps you uncover the ideal role for you and the day of the profession, given your personality, skill sets and passions. We'll even link to it in the description below. Also, I'd like to invite you to our Facebook group called Becoming World Class Data Leaders and Entrepreneurs. I would love to see you in there. There is a link to that in the description below as well.
Then be sure to show it some love by giving it a thumbs up. And tell me in the comments below, what area of the debris field are you working? Also, be sure to subscribe to my channel so you'll be the first to know when the next episode drops.