Alexis
Hi everyone, welcome to Talk Tech with Data Dave. I’m Alexis, your host of Talk Tech with Data Dave, and I’m here with my dear friend Data Dave, to answer your questions about all things data, all things cloud, all things technology, and all things D3Clarity. 

Data Dave
Good afternoon, Alexis, and welcome. Thank you. You’re looking great today, and I’m happy to be here. 

Alexis
Got my blue shirt on. I’m feeling good.  

Data Dave
Yes. Likewise, we match today. 

Alexis
I do have a question for you today, Data Dave, but as I like to say, if you have a question for Data Dave, we would far rather answer your questions than just all of my questions. Now, don’t get me wrong, I have questions for days, but listener questions are always better. So, if you have a question for Data Dave, please feel free to e-mail us. At talktech@d3clarity.com, or you can contact us right on the D3Clarity website if it works better for you that way. Either way, we’d love to hear from you.  

So, Dave, I do have a question today. 

Data Dave
Oh, excellent. 

Alexis
Data science versus data analytics. What’s the difference? 

Data Dave
Ah, that’s a good question. We hear a lot about them both at the moment, don’t we? We talk about them quite a bit, and we hear about them quite a bit. So, data analytics versus data science: they’re not really competitive, so we can’t really say verses because they don’t compete with each other. You don’t stack one against the other and say, “I’m going to do, you know, which is going to win this game, data science and data analytics.” Realistically, data science is probably… I’m just thinking about myself… data science is probably a subset of data analytics, and they’re linked together. So, if you think about data analytics being the total view of analyzing all data, then data science is a particular set of techniques and approaches in that analysis of data. 

Alexis
Oh, so definitely not versus, but more of a data analytics would be row 1, and data science would be row 1A. 

Data Dave
Yes, to a certain extent, it’s an approach or a technique. So, data analytics is really just the skill set really, I suppose. Or the approach of analyzing the data to give you meaningful information, meaningful knowledge in order to take action. I’m going to analyze a set of sales data, and I’m going to come up with a decision that I should be able to sell more if I open the shop in Hawaii. 

Alexis
Okay. 

Data Dave
That’s straightforward data analytics. I’m analyzing this data. I’m coming up with an approach, same as business intelligence or business reporting, right. Business intelligence is just having intelligence about your business. Knowing what your accounts were last month, etcetera, straightforward business intelligence. Then analytics is to say, “I’m going to analyze this data set to see whether I can come up with something I can take action on.” 

Data science is a little more defined, a little more specific. It’s where data science starts to say is, “I’m going to look at this data. I’ve got this huge cadre of data in front of me,” and essentially I want to use fundamentally coming from the scientific method the, “I’ve got a hypothesis. I would like to test it in this cade of data in the set.” 

Alexis
So, like basic 8th grade definition there is a scientific method. I like it. Okay. 

Data Dave
Basically right. basic scientific method of, “I’ve got a hypothesis that is a little more abstract and I want to test it in this set of data,” or, “I want the data to start to tell me something from it. I want to explore this set of data with a scientific approach to start say what is in here. What can it tell me versus what does it tell me.”  

Basic data analysis would say, well, “I know it tells me that I want and collected the data.” So it would tell me that data science starts to look at it a little differently, and say “I’ve collected this ccadre of data. I’ve collected all this data and I’ve got this hypothesis and I’d like to test it, I’d like to try it out. What mathematical – what approach can I use to drill into the data to build a hypothesis and to test the hypothesis.”  

Alexis
Because mathematics is what we use to tell the future. 

Data Dave
Right, exactly. You’ve listened to one of our previous podcasts before, Alexis. 

Alexis
I’m learning from the podcasts.  

Data Dave
Mathematics is the language of prediction and data is the language of history. So, if you start to think about that, then what we’re starting to look at can I apply a hypothesis of my history. 

Alexis
I like it. 

Data Dave
Apply the scientific method on to that to find the evidence. Explore the evidence and further use that evidence to build a mathematical model that now lets me predict the future. 

Alexis
Which is what I would, you know an uneducated brain, would say is data analytics. I’m going to use some sort of mathematical model to help me predict the future, but I love, the specifics of have a hypothesis, and I need to see if my hypothesis is correct. 

Data Dave
That’s really the fundamentals of the differences. Often data analytics or straightforward analytics is simply there in a predetermined structure, if you like, which is, the data was collected to reinforce and to tell us some details about a pre-existing model. The way I would put it and this is these terms overlap tremendously in the way people talk about them in the industry and in society, right? They massively overlap. Some people will have slightly different definitions. I’m probably a little more strict in my definitions here. The way I would put it is data analytics is more, “I’m analyzing the data to tell me something that I probably already know the approach to getting it.” Things like, “How many did I sell over the course of the last three months. Therefore, what should I expect to sell next month?” That’s pretty straightforward. That’s a linear extrapolation. I’m not doing anything strange or weird.  

If you start to talk about data science, then you’re probably gonna start to talk about, “How do my sales of winter coats correlate with the weather forecast in different cities around the state?” That’s a more complex hypothesis and a more complex prediction structure. So, I’ve got to explore and test my hypothesis and test the way that’s going to work as temperatures drop and different things in my cadre of data. Is that understandable? 

Alexis
I love that you said that how this is your stance on this. This is your definition because it gives me a great opportunity to pitch our bonus podcast, which is Data Dave Dives Deeper. If you would like to join us on Data Dave Dives Deeper and have this discussion with Data Dave because you have a different approach to these definitions, we would love for you to join us. You can reach out to us by emailing us at talktech@d3clarity.com or reaching out to either of us on LinkedIn as well. We’ll be happy to have you on the other podcast. There we go. That was my shameless plug because we’re good at shameless plugs. 

Data Dave
That’s excellent. Thank you. 

That doesn’t mean that data science doesn’t turn into analytics. What we usually find is the difference in skill set as well. You might have a business person that analyzes their data in the model that they’re used to to make predictions for their future and to have better knowledge of their business. Whereas you would usually hire a data scientist or a mathematician to do the data science and then as that model evolves with that data scientist that will get turned over to that business user to run that model, execute that model and follow up on that.  

So, the model from data scientist becomes the model that is executed as part of analytics. Data science in this definition is upstream of data analytics. The data science uses the data to determine the analytic model, and then the analytic model is used by the business to predict the business. 

Alexis
We’re definitely in an “all data science is data analytics, but not all data analytics is data science” realm. 

Data Dave
Yes, I can go with that 

Alexis
All is a rough word. I don’t love using like “always” and “never” because the truth is there’s always exceptions to that but that was an easy kind of recap of that concept of what’s the difference. 

Data Dave
Yes, they’re very strong words “all” and “never”. 

Alexis
So, probably the more important question is if we have a listener out there who needs help in data analytics or data science. Is that something that D3clarity can help them with? 

Data Dave
YES! I said that very quickly, but yes, we can. 

Alexis
Of course we can. Alexis, come on. 

Data Dave
Come on, Alexis. Don’t be ridiculous.  

Of Course, the interesting point in this is where we particularly specialize. Yes, we can help with data science and data analytics, where we help particularly is making sure that your data actually describes what you want it to described.  

If you go back to our previous podcast and look at “What is data?’” and “Can we ever expect perfect data?” I think it was the other one where we talked about this. Which is, is your data actually good enough to perform data science and does your cadre of data describe your patterns of business or your patterns and your structure well enough that if you determine some patterns and some predictions, some mathematical formula from it, are you comfortable with those because you’re interrogating your history and determine the future? 

I’ve dropped 1000 balls as an experiment and to predict acceleration due to gravity. That’s my cadre of data. I’m now exploring for what is the acceleration? What is the force of gravity? Is it 10 meters per second per second? Is it 9.8? Is it 7? What is it? I’m getting this prediction from mathematics based off the evidence in my data. Is it good enough? Is it clean enough? Does it describe my business well enough? Is it complete enough? Is it accurate enough for me to perform data science in order to come up with a valid prediction? Because the problem with prediction is if your history isn’t valid enough, then your prediction starts to be a little less than people can trust. 

Alexis
I need you to correct me if I’m wrong but, we predominantly focus on master data management and data governance, both of which are topics that we’ve talked about in previous podcasts, to some extent, and we use those two tools mostly to help us make sure that the data is in the best place it can be. 

Data Dave
Yes, exactly.  

Alexis
I did learn something. 

Data Dave
We use various tools in the data governance realm that start to say, “How do you govern your data through your business?” such that when you reach the point of “I’ve collected it all in my data lake or my data warehouse or wherever you want to put it, is it good enough? Am I confident in it enough to make the predictions based off it?” And master data management is a tool, one of the tools, one of the areas that we spend some time on to get people’s data at a high enough quality to be able to do that.  

So yes, those areas come into this to start to guarantee that the data you’re collecting and the data that is flowing into your data lake is clean enough, valid and accurate enough so that you can perform data science. So, you can jump in there with your hypotheses, swim around for a little bit, and come out with a prediction that you can trust and you can depend upon. 

Alexis
Thank you. That’s very, very helpful. Those are two terms that I hear people say all the time, obviously within D3Clarity, but also just kind of in the world. I’ve got friends on LinkedIn who are data scientists or who are data analysts and every time I see that, I wonder what the difference is, but now I have an idea. 

Data Dave
As I said, people do have different definitions, and these terms are used somewhat interchangeably in the market. And as you’ve got used to with me in particular, I do tend to be fairly strict on my definitions, but these terms do vary, and you’ll hear people with Ph.D.’s in mathematics and lots of mathematics who are data scientists. And people who are being data analysts, it doesn’t really matter. It’s where did the models come from? How did that get established? And some people will have stricter definitions than I do, and I welcome anybody who can define these terms in more detail. But that’s the working definitions that I like.  

Alexis
I like that. If you do, like I said earlier, want to join us to have this discussion, by all means, reach out to us, and we’d love to have you on an episode of Data Dave Dives Deeper. Otherwise, we are always open to answering your questions. You can e-mail us at talktech@d3clarity.com or submit a question, right on our web page. Which is kind of new and up and running, if you haven’t gone and checked on our web page lately, Dave actually led that mission, and it’s looking pretty great. Dave, anything else as far as a, “What I should walk away with from this conversation?” 

Data Dave
The only thing I would say is walking away from this conversation – again, to something we’ve talked about in previous sessions. The amount of data that we’re collecting is growing exponentially. It’s going through the roof, always across every walk of life that we’re in. The ability to understand and to apply new and interesting techniques into that is fascinating and continuing to explode, whether analytics, which is just using the data and the models has been created. The data science of producing a new model. And then we get into the realm next, the machine learning and the AI kind of structures where we’re using computers now to start to predict and do some of this learning and some of this interaction on our behalf around the data that we’re collecting. And we just keep looking at the quality and the structure of the data that all these tools are operating on because they are realistically, only as good as the history that they’re using to predict upon. 

Alexis
If you haven’t listened to “Can data ever really be perfect?”, which is one of the episodes that we had earlier in the podcast, I’d highly recommend you go back and listen to it. That was such an insightful episode, I think that is a really good jumping point to this conversation and to that point you just made Dave.  Thank you so much for talking today. I really appreciate it. Thank you everyone for joining us. 

Data Dave
And thank you, everybody. And if you do have questions, then please let us know. Thank you.