March 24, 2021

E14: Rick Hall at Aginity Final

E14: Rick Hall at Aginity Final

In the Tech Leaders Talk show, we chat with Rick Hall, CEO of Aginity. We chat about why data analytics is important within a business and how to get started with data analytics within your business.

We also discuss why it is important to enable non technical staff to use data analytics to drive data backed decisions within an organisation.


In the Tech Leaders Talk show, we chat with Rick Hall, CEO of Aginity. We chat about why data analytics is important within a business and how to get started with data analytics within your business.

We also discuss why it is important to enable non technical staff to use data analytics to drive data backed decisions within an organisation.

See omnystudio.com/listener for privacy information.

Transcript

Rick Hall at Aginity Final

Ernst: [00:00:00] Good morning and welcome back. We have good show for you today. Our next guest is the CEO of Aginity. Aginity's mission is to democratize data analytics and enable any business user to use corporate data to provide business and insights

I'm excited to introduce Rick hall. Rick has extensive experience in data analytics and shares with us.

How any business can get started.

With gaining insights into the business operations.

I start the show by asking Rick why businesses should care about data analytics?

Rick Hall: [00:00:33] Yeah, so that's a really good question. I mean, I think at this point, analytics has kind of evolved in everything that we do. So, whether it's innovation and we're using analytics to help us understand how well we're doing with a new idea.

So, if you think about, you know, like recently with COVID, of course we're using data and analytics. To understand how well the vaccines work. Right? So, all kinds of things that we innovate. We use analytics to, uh, to help us [00:01:00] understand, uh, in some cases, actually analytics is the centre of the innovation itself.

So, we're thinking about a new way of doing something where we're having some amount of intelligence speed up a process, or get to a better result. So, I spent some years working in, uh, consumer goods and retail. We used analytics to kind of predict the performance of pricing, uh, and that's done through all kinds of things.

Um, and then ultimately, we're just using analytics to measure how well we do in a given activity. So yeah, there's just very few things at this stage of the game where some degree of analytics, this is not critical to, uh, uh, helping a process go better.

Ernst: [00:01:46] Okay. Interesting. And then I know analytics can, even for small business can be really complex, especially if you don't have a diet data scientist, you know?

So where do we, where does a business [00:02:00] start with analytics?

Rick Hall: [00:02:03] Yeah, I, I mean, just like the central thing is what are the processes you care about the most, right. So. Uh, what's your output? What are you selling? Uh, how well are customers adopting it? Uh, and of course, how are you performing in terms of your ability to track your, uh, your business?

So, I think it starts with. What are your most important outcomes and being able to measure them? And I think it's just kind of a natural thing for business leaders to start there. Uh, but then ultimately it gets deeper and deeper and deeper into a business as they look for more and more processes where they can use analytics to improve performance.

But at the end of the day, we care about analytics because analytics is going to help us perform better.

Ernst: [00:02:52] Give us an example of some of the common use cases, right? It, what are the more common ones that you see on a day-to-day basis?

Rick Hall: [00:02:59] Yeah, [00:03:00] so I think that if you looked over the past, say 10 or 20 years where analytics has been applied, the most is on kind of the big, centralized processes and organization.

So, let's say sales performance analysis. So, if you've got a lot of customers. Understanding how well we're selling to different kinds of customers is a tremendous amount of analytics there. Uh, if you're building a complex product, uh, and it's coming from a lot of different places, then, you know, supply chain, uh, analytics become really kind of critical to your business.

Um, I think many companies apply it to finance. So, you know, particularly if you have a complicated set of financial activities is going to apply there. Um, so what I think we've seen over the past say 20 years is first, these big centralized processes have been tackled. Uh, and increasingly we're tackling them with more and more automation, more and more intelligence, trying to [00:04:00] use analytics, to look through large amounts of data that we don't have to on an individual basis.

Um, but I think we're also now spreading it analytics out across the organization and to kind of like every nook and cranny of the business.

Ernst: [00:04:16] I think looking at some of my sort of previous businesses, a lot of the times there's gone. Uh I've. Originally I started having an issue, just thinking about, okay, what exactly am I trying to achieve?

So just looking at those examples that you've given us, can you dig a little bit deeper on, when am I hearing there about trying to streamline a process or you also use a sales about trying to learn whether a product is going to be viable on the market? You know, can you dig a little bit deeper into those?

Rick Hall: [00:04:50] Sure. So, you know, you could kind of almost think of two categories based on that kind of question. What is, I'm just trying to monitor how well the process is doing. So, [00:05:00] you know, maybe I have some output from that process. Like I'm producing a new product every, you know, every few seconds in a big running supply chain.

And I just want to look for defects and see how often those things occur. So, it's kind of a monitoring. You know, kind of activity. So, we're looking at results that occurred, and we want to see when something happened that we didn't expect, uh, You know, what's kind of become more and more interesting is, is kind of predictive world of analytics where I'm trying to kind of ascertain, I'm thinking about doing some particular activities.

So, I'll take the pricing example. Like I'm thinking about raising the price of my product, and I want to kind of figure out in advance if I raise the price by. You know, 10, 10%, what's that going to do to my sales. Uh, and am I going to make more money or less money? And increasingly, particularly in consumer products, we're seeing, uh, [00:06:00] these kinds of decisions being made by predictive models so that we're using a set of analytics to kind of predict the performance of a pricing change or a new promotion or marketing campaign.

Um, so that we can, you know, kind of decide in advance, whether this is a good idea or bad idea, and we're kind of applying kind of what if or scenario-based analysis, uh, on our business. So, you could kind of say reporting retrospective analytics. That's where the whole industry really started. Uh, and really particularly in the past 10 years, and you know, it was kind of taken off.

With machine learning here recently, we're using more and more predictive analytics to get ahead of decisions and help provide insights into, uh, uh, into decision makers before they make an action.

Ernst: [00:06:51] Okay. So where does that journey generally start? Is that somebody within a particular team that goes right, we need to figure out this process more, or [00:07:00] do you find generally it comes from the leadership going, we we've got a specific challenge that we need to meet, right.

And then analytics could potentially help us solve this problem. Where do you see that request? Generally, originates from.

Rick Hall: [00:07:16] Yeah. So, I think historically it would have been from the top, right? You'd have an executive that would say, you know, we've got this problem, let's say it's pricing. And they would go out looking for a means of doing a better job with it and come up with some analytics solution.

But I think businesses have become kind of so complex and the expertise in many different areas. Is so diverse that we're increasingly seeing kind of analytics and decision-making pushed out into the business. So, we see, you know, kind of the most sophisticated customers are actually not relying on a central analytics team to provide all the answers and do all the work, [00:08:00] but we're seeing them.

Really seek to empower people at the edge of the business. People who have business problems with the capabilities to, uh, to build analytics. So, you're kind of almost teaching people in the business, how to look for and identify analytic problems, uh, how to articulate, how to solve them, and then how to go about, uh, seeking that solution.

Ernst: [00:08:26] Okay. For leadership team to make a decision, you know, w we want to use analytics to solve a particular business problem. Is there sort of something that the leader needs to think of and understand about leadership, about leadership, but analytics before they even start going down that path?

Rick Hall: [00:08:44] Uh, I think, you know, I think increasingly.

You want to kind of apply the test and learn, you know, kind of approach to analytics, right? So, you've got a problem. You think, you know, gee, if I had more data or I had more insight into what's going on with this particular, uh, you know, [00:09:00] part of my business that I I'd be better off. So, it starts with something that simple, uh, and, uh, in the, you know, kind of past we might've said, well, go off for six months and.

And try to build something. I think increasingly you want to say, okay, how can I quickly take an idea and test it out and learn whether it works and improve on it. Now that almost becomes the key. Infrastructure that a business organization needs to put in place centrally is I want to put an infrastructure in place that lets my business teams quickly test out new ideas.

Right. So, you know, if they don't have the right data platforms that are flexible and easy to accommodate data and the tooling, then maybe they'd have to go get it and that's going to slow them down. So, if I can put the, the, the capabilities in place. That let my business user’s experiment. [00:10:00] Then I think analytics will evolve, uh, in the most effective way.

Ernst: [00:10:07] That's very interesting. So, so what I'm hearing is that you basically almost putting infrastructure in place or, and the processes, uh, almost like a template so that other pieces of the business can basically go look, I want to do experiment on this particular item. Uh, therefore I can actually just use what you've already kind of built.

Um, is that understanding, correct?

Rick Hall: [00:10:28] Yeah. Yeah, it is. It's, it's really interesting because, you know, kind of up until maybe the emergence of the cloud and some of the new analytic platforms, when we were going to implement an analytic system, say 10 years ago, we would, we would go out and buy a piece of hardware, big computer, and a set of software.

And, you know, we would have to have that software for years and that hardware for years before, you know, kind of. It ultimately was retired. So, we couldn't really [00:11:00] accommodate new problems that we didn't think of when we made that purchase. Right. If that makes sense. So, I purchased X amount of capacity.

And I know what problems I'm going to solve with it. Well, now we live in this world where this computing power is really elastic, meaning it can be expanded and contracted easily, and the systems scale out to bigger and bigger workloads relatively quickly. So, given that new computing power, if I put that in place for my organization, now, if I teach my business users kind of how to fish.

I've given them a place where they can go and experiment and try things that I didn't anticipate and the capacity will expand or contract as, as needed. And it just creates a whole new possibility for this kind of pushed out at the edge’s kind of last mile innovation in analytics.

Ernst: [00:11:59] That's

very [00:12:00] interesting because one of the things I was thinking about is, you know, is analytics an organizational wide attitude, or is it more a look, a I've got a small team and a small, the small team thinks about, or I want to try and solve a problem and what I've taken away from that.

Um, those, those statements are basically it. You can very easily turn it into an organizational thing to have that experimental and learning. Um, attitude and be able to. Provide them with the capabilities to type to use those, those attitudes.

Rick Hall: [00:12:36] Yeah, I think so. I mean, I, I'm not sure how easy it is, but that's the transition.

I think that, you know, the whole industry is going through it. Right. So, you know, I spent most of my career developing those big centralized systems and then kind of watching people at the business edge. You know, kind of figure out how to do things or come back with problems that we couldn't solve for them.

So, I think that they approach that [00:13:00] we're seeing the, you know, the leading companies do is say, okay, I've got to recognize this, this whole different way of doing it. I got all these business people out there who need to make decisions who need analytics to do better. And what I'm going to do is create an infrastructure.

And then I'm going to think about how do I educate them around the use of data. And basic analytic capabilities. And then my central team becomes more like a, a set of coaches or they're supporting, you know, the complex problems. Uh, but they're not actually doing all the work themselves, if, if does that make sense?

Ernst: [00:13:38] Yes, absolutely. Okay. So, the leadership team decides, right. We want to try and build the infrastructure, um, to enable staff to, you know, you know, to generate some analytics, I'm assume that starts with a goal and then building a team. So, what would a team [00:14:00] look like or, uh, or do I need to really make clear what the goal is?

Rick Hall: [00:14:07] Well, so, so the goals, I think, as you, you know, for the big processes, like I've got to do a better job on my supply chain. Those goals are going to be right at the top of the organization, but kind of down in the organization at every little business unit, they're going to set their own goals. So, your team is going to be structured to support that process.

So, so what do you need? So of course, you need an analytic platform, right? So, you know, kind of, you know, if you're going to have this. Innovation approach. You probably need a platform that's kind of elastic and scalable, and that's going to mean, uh, you know, one of these modern data platforms that are out there.

So, you know, all the big cloud vendors have a platform, yet people like snowflake or Amazon have Redshift and, you know, uh, Microsoft and Google, they've all got these kinds of capabilities. Um, you need that. Any part of your [00:15:00] team that understands how to maintain and support that infrastructure. And then you need kind of at least two other roles.

You need some, uh, what I would call analytic engineers. And those are people who can, you know, kind of do the work for the really complicated bits. So, they're going to build the really sophisticated models, but they're also going to be there to support and train and coach. Uh, those people out in the business, right?

So, you're going to actually be trying to support and train these I'll call them business analysts that are semi-technical that have a affinity for data, and that have problems that we need to solve, and we want to make it so they can solve those problems and kind of the self-service manner. So, you know, kind of the old paradigm, the central analytics team did everything, you know, the new paradigm.

The central analytics team is doing some of the really hard bits, but they're [00:16:00] also then, uh, supporting, uh, this broader community of users. And I was, I was sitting down with, uh, the chief analytics officer for one of the largest healthcare companies in the U S it's pretty big outfit. And, uh, you know, and he said to me, he said, you know, I have 5,000 people in my team.

So, you think, well, that's a huge team, right? And he says, but there are a hundred thousand people in our business who are trying to make analytic decisions. And I can't possibly do all the work for them. So I've got to have an infrastructure and tooling and processes that allow me to teach these people how to do the work themselves.

I've got to give them self-service tools. I got to teach them how to fish. And I think for an organization that's going to try to evolve quickly and use analytics. Well, that's kind of the mindset that you know, [00:17:00] that we see now that's emerging,

Ernst: [00:17:05] I think with any program to try and get any program, which is going to be a are re you know, relatively large scale. It's important to try and hit quick wins that gives leaderships, uh, the confidence that we're actually investing something that's worth investing in as well as giving, um, you know, the business, some confidence.

Was there sort of any approaches that you can think of and rec or recommend where, you know, we're a business can hit quick wins, uh, just to prove that there is value to this.

Rick Hall: [00:17:36] Sure. Yeah. I mean, I think you start with, you know, some problems that, you know, that are at the top of the list, right? And you say, okay, you know, we're going to put some of this platform in place.

The good news about these, you know, cloud-based platforms is, you know, you can almost start with a credit card, right? You don't have to buy a tremendous amount of capacity to get started. Uh, and then you, you know, you need a small team of, [00:18:00] of analytic engineers. Who can do some hard work and work with people in the business, right?

So you probably want to prove out that you can make this collaboration between business people and engineers really actually work. And so start with the problem that's, uh, this well understood that you know, is important. Uh, start with a small team, let them get started and go from there. Uh, you know, I was, I was talking to another, uh, you know, kind of customer of ours recently and they had done just that.

So they had, I think it was 5,000 business analysts and they had served up to those analysts, always the past, uh, the actual results. And they said, well, we want to kind of create this self-service. So they created a small group of power users and they said, okay, within this. Particular business area. We're going to actually give these power users [00:19:00] some tools and see how well they can do themselves.

So it was a small experiment in a known area. They chose people who had an affinity for data, et cetera. And they said, okay, we're going to start there. We're going to see how well it works. And we're going to kind of expand, you know, from there. And, uh, I think they've, you know, I mean that particular customer is a recent conversation.

They're still. Early in that process, but you know, kind of the feedback was that the results of that, uh, are, are going well. And they're excited about that, that approach, but, you know, if you think about the alternative, right? So, uh, unfortunately the past, you know, year we've had this crazy experience with COVID right.

Uh, and it's just changed a lot of things, right. So, You know, if you were selling a consumer product in a retail store, or maybe you're selling it partly in retail. And then partly through restaurants say it's food. You might've had very sophisticated [00:20:00] models to predict what would happen, but you know, when COVID came along, all those models went out the window.

Right. Uh, and you have people in your business who are reacting all the time to these changes. You can't possibly have a centralized team, make all the changes and solve for this big unknown, uh, fast enough if you can't empower people at the edge to, uh, to, uh, react and change themselves.

Ernst: [00:20:31] Yeah, exactly.

Exactly. I want to come back to these power users in a minute, but, uh, I'm just trying to help, um, listeners visualize a sort of phases. A business would go through to try and get drive an outcomes through analytics. Can you talk about the sort of very high level phases that need to be considered and, uh, you know, what's involved with those phases.

Rick Hall: [00:20:55] Yeah. So, so first let's just kind of think about this transition I'm talking about. So let's [00:21:00] say let's call it centralized or data warehouse cantered, analytics, and then let's call this next phase. We're going to call it kind of a collaborative or distributed analytics. Right? So your first question is if you're trying to solve, if you're starting at the beginning, you're probably needed to answer these big centralized questions first.

Right. So, you know, you're just trying to basically. Monitor sales or monitor your supply chain and that's going to be done by your centralized team. Right? So your first step is have you put in place the basic analytics you need for your biggest core operations that everybody at the executive level knows about.

And your starting point there is just to recognize that that's what you need. Uh, you need a leader of your analytics team. That's going to. You know, kind of drive that capability, the centralized capabilities they're going to have, you know, members of their team that are gonna [00:22:00] include data architect, system architects, and, uh, uh, analytic engineers.

And you're going to start with your biggest problem. Uh, you're gonna put this team together to solve that problem. And you're gonna, you know, kind of go after it until it's solved, that may take you weeks or in some cases, months, or sometimes longer, but let's assume for a moment you've already done that.

Right. So I think many, many, many businesses have solved their big centralised analytic problems. Right. So your first question is, are you still trying to do that? Or have you kind of done that and now you're trying to create an environment where analytics becomes part of your culture, right. And helps you evolve in a very changing landscape.

Right. So if you've done the big bits and now you're trying to push it out in the organization and you want to evolve more quickly, that's when you start thinking about these different [00:23:00] roles and how capable your platform is. Of supporting rapid evolution, right? So you need a platform that's going to scale quickly, um, Scott to have to be elastic because you don't actually know what your users are going to need to do.

You're going to identify people who need better analytics in their, their business in parts of your business. And you're going to figure out where can I go to identify more power users that. Uh, can, you know, can actually do this work and that they're going to be, oftentimes you're going to find people who, uh, are really using a whole bunch of Excel or Google docs.

And, you know, they're, they're using all these end-user personal productivity tools, uh, to, to analyse data, right. And these tools, you know, while they do a lot, uh, they make it very difficult to collaborate. Uh, you can't really easily [00:24:00] track the results from one to the other. Uh, they can only handle kind of certain volumes of data, right?

So, but you're going to see this in your organization. You're going to know that this exists and you're going to figure out the places where these problems are biggest, and you're going to kind of try to tackle those. So, uh, and in tackling those. It then becomes this kind of empowerment process of how do I give those users?

Who've been doing all this stuff on their own or complaining because they can't get it done. How do I give them access to my, my new platform? That's scalable and, uh, give them the right tooling. And you know, then I'm going to have somebody on my side is going to support them. Right. So we're going to kind of.

Now my engineers, some of them will just be building the centralized stuff, but some of them, I might identify a subgroup that is going to really be in the role of helping coach and support these broader users in the, [00:25:00] in my organization.

Ernst: [00:25:02] So I'm hearing that not, not every organization has a data scientist hanging around, um, or, you know, but, so I'm kind of also hearing that you, you will need somebody.

Who's got a really good understanding of your internal business processes, uh, where those, that, where the data critical pieces of data live, uh, then you need to have somebody who can bring all that, those pieces of data together into a centralized location. Um, and then you need somebody who's able to, you know, well, cleanse that data to imagine, and then to do the analysis around that.

Would that be a fair statement?

Rick Hall: [00:25:42] Yeah. Yeah. So we, we talk about something, we call the analytics journey, right? And so of course you have to start with the problem that you're trying to solve. And then, you know, there's certain data that you think you need to solve it. And that journey is both acquiring the data.

Ingesting it into your platform, [00:26:00] doing data quality on that data. In some cases you need to integrate that data with other data, then you need to calculate that data into whatever type of analysis you're trying to do. And then also you need to provision it or analyse it, right? So that, that, that journey from the acquisition of data, to the analysis, that's common across us, uh, almost across all analytic problems, right.

And in the big centralized world, all of those steps were done by engineers, right? And the user was way down at the other end, waiting for the output in this kind of new democratized world. Uh, you're giving them tools where they can in fact, come up with their own data, ingest it, check the quality of it.

Integrate it provide calculations and do the analysis with kind of self-service tools that a semi-technical [00:27:00] person, somebody who's maybe just skilled with Excel can do themselves. And so, you know, in where we all came from the old world, uh, you would never let a user ingest their own data into your data warehouse that, you know, kind of when I was coming of age, You know, we would, that would be hypocrisy or, uh, not hypocrisy.

It would have been heresy. Right. Uh, but in the new world, we're going to let them do that because our platform can support it. Right. So we'll have centralized data that we've acquired and brought into the system because we already knew it was important, but the end-user's going to have some other data that they might need that I didn't know, they needed.

And, uh, uh, I want to make it so they can easily ingest that data themselves. Right? So the journey is, you know, kind of adjust, cleanse, integrate, calculate, analyse, right? [00:28:00] Uh, in the centralized world, all those things would be done. A bunch of them were done by it until you get to analysis in the new world.

I want to empower users to do that whenever they need to. And then add that to my system and

Ernst: [00:28:16] free up also free up your it people to be able to do some other, other stuff.

Rick Hall: [00:28:21] Yeah, for sure. Yeah. I mean, it's always just more work to do, uh, then, uh, than there is resources or most of the time anyway. Yes,

Ernst: [00:28:31] exactly.

Exactly. So let's talk about the tooling. So what type of tooling do we need to be consider considering at each one of those sorts of phases?

Rick Hall: [00:28:40] Yeah. So, uh, you know, kind of first is that the let's call it the analytic platform, right? So, you know, you need this, uh, this engine, that's going to allow you to ingest data and do what you're going to do with it.

Right. Then, then you really need a tool. That's gonna sit on top of that. It's going to be an end-user experience. [00:29:00] That's going to let you do these different activities, right. So, you know, in our world, certainly the Aginity the perspective is. We want one tool that lets a business user, do everything from ingesting the data, to querying it and writing a calculation and, and analyse it.

But, uh, historically you would see different tools for each of those steps. And in fact, if you go to the ingestion of data in the way that we did things in the past, I would have ingested the data. And pre-calculated what I think I'm at a use. I the process of loading the data in the system. So we used to try to build the logic into the loading of the data so that it already comes pre formatted for the analysis I need.

Uh, and that sounds like a good idea, but the problem is that that means that the data I'm ingesting. Is only going to answer the question that I knew I [00:30:00] was going to ask, right. Or it's going to do that well. So what we try to do now is let you point at any data, ingest that data in its native format. And that way you're not going to lose any of the value of that data.

Uh, and then multiple users can build processes on top of it that create those, uh, uh, those outputs. So. Uh, if you had a tool, like what we provided agility, you know, pro or premium, we're going to provide you an experience for that. Uh, if you're the kind of centralized engineering function, you're going to have maybe one tool to load the data, another tool for data quality, uh, maybe even a master data system to merge and integrate data.

And then you're going to write, you know, calculations in a SQL type engine to analysis, analyse it, that you're going to then provision into either a [00:31:00] data science platform or, uh, you know, BI tool. You know, kind of our view is that having all these different tools really makes it difficult for the average business user to consume things.

And. You know, w we think that this experience should become more and more fluid, uh, over time, but you want to, you know, kind of, if you're heading for empowerment, you want a simple tool. I mean, that's kind of the key question. You kind of ask yourself, are you happy doing all this work in a centralized world?

I don't think you will be, but let's just say, if you are then what I'm suggesting, maybe doesn't make that much sense. But if you are trying to push analytics out to your business, because your central team can't move fast enough, or, you know, there's too many different things for them to actually take on.

Now you need tools that a semi technical person can use. And so, you know, the, the, the [00:32:00] key thing for that is it's got to be easy to use. Right? Uh, it's got to be something that, um, I don't have to, you know, go through a year of training to get at, and that a compass is my whole job. Right. So I don't have to put six tools together and figure out how they connect, you know, the average person who maybe has been doing this in Excel or.

You know, some local database, uh, you want a simple environment, they can work in and you want that environment to take advantage of this big data platforms. Right? So these data platforms, hyper scalable data platform. So like snowflake or Redshift, or, you know, the others I'm talking about, they do the work, right?

So they're, these engines are really complicated and sophisticated. Uh, then I think you want to use your experience that sits on top of that, that. The average business person can use.

Ernst: [00:32:57] Okay. So this is a pretty good segue. [00:33:00] Tell me more about an agility, the problem you're solving, um, and maybe the journey that, um, agility takes a customer

through.

Rick Hall: [00:33:09] Yeah, so, so we, we see ourselves in this world of collaborative analytics, right. Which is kind of a new category within the analytics space.

Uh,

and we think of it as collaborative analytics because we want. The engineers, the technical team and the business users, the semi-technical team to be able to work together effectively.

Right.

Uh,

and,

uh,

so what we have created is a tool set.

Uh,

Aginity pro is for an individual,

uh,

Aginity premium is supports multiple users together.

Uh,

and there they create, it's a simple tool set that lets somebody ingest data,

uh,

calculate that data. Build analysis on it. And ultimately either analyse it directly in our tool or provision it to [00:34:00] a BI or a,

uh,

an AI machine learning tool downstream from there.

Um,

we believe that these data platforms have created all this power and we think that in the old world, We would have different platforms for different steps in the process. So, like I said before, maybe I have one tool that does data quality and other tool and just data and something else to do analysis.

Well,

um,

that creates a lot of complexity, right? So,

uh, you know,

we see this need to empower business users.

Uh,

these analysts,

uh,

who are out in the business, Who maybe didn't come with a technology degree. We need, we need more and more of these capabilities for these people, but they can't do all the work themselves.

Right. So,

uh,

we wanted to create a tool set that empowered those users, but also linked back to the engineers so that the [00:35:00] engineers become this. It, it was kind of like a force multiplier. Right. So if I can create an environment where one engineer. Can help support 10 or 20 or 50 analysts, each of whom are going to do,

you know,

maybe 80% of the work themselves, but need a little help with the complicated parts.

You know,

we think that's,

uh, you know, that that's

where the world is right now. Right.

You know,

you hear you, you mentioned not every organization has a data scientist or a data science team. Well, there just aren't enough data scientists out there. To solve all these analytic problems. Right.

You know,

there was a study IBM did in North America a few years ago and they said, okay, we're going to need,

uh,

2 million,

uh,

data scientists in North America.

Uh,

by 2020,

um,

and that's how many job openings there will be. And by the way, we create like 40,000 of these people a year when [00:36:00] we go across education of our, of the whole system. Well, the answer that IBM came up with is what we're going to educate more of that, right? I'm like, well, you're not going to go from 40,000 to 2 million.

I mean, I'm sorry, you're just not going to get that many people to learn hardcore data science. So. Our view is what we've got to do is create tooling that the business person can use that gets them a long ways down through the process, right? Because it's not going to happen that they're going to create enough engineers with sophisticated data science capabilities, but we can create tooling that can create this collaboration between the engineers and the, and the business users.

Um,

and so that's what we set out to do.

Uh,

and we came upon that,

you know,

kind of, if I go back in my own career,

uh,

I was building these big analytic pipelines and then I would see the users take the output of my big analytic [00:37:00] pipelines that do other things with it. Right. And,

you know,

it used to frustrate me like, well, no, no, I produced the answer.

Right.

Uh,

and it was actually somebody who worked for me, who finally came to me and said,

you know,

What's really going on here is they actually have different problems because we didn't solve every one of their problems. And they all have these little bits of variation and there's all these people out there and we're sitting around talking about and say, okay, well, we can't possibly ever in our centralized team solve all of these problems for all of these users that have these little bits of variation.

Right. So,

you know,

that set us on the journey of how do we create tools to empower.

Um,

and how do we do that in a way that also takes advantage of the tremendous skills,

uh,

of the central analytics teams and that's,

you know,

that's what set us on collaborative analytics and where we are today. And

I, I, I think also [00:38:00] providing people with those tools instead of waiting for instance, months and months for engineer to produce the output,

Ernst: [00:38:06] um, you know,

Rick Hall: [00:38:08] if they're.

Business. Person's able to produce that outfit.

Ernst: [00:38:11] You know,

Rick Hall: [00:38:11] they'd probably get a better, not a better, necessarily, always a better result, but a quicker result and exactly what they're looking for. Hopefully because that's, what's in their

mind. Yeah, that's right. And I think that's a really key thing that you're hitting on.

You know, when we did these centralized things, we used to say, you kind of almost had to have two people in a box, right? Every time you were doing something, you had the engineer, who's going to program it. And the business person who actually understood the problem, and they almost had to always work together to solve a problem.

Right. You could never do it just by, you know, one or the other. Right. And, uh, uh, that sometimes. Created these long life cycles because the business users didn't always understand exactly what they needed. Right. Um, you know, they might know the problem, they don't know the solution. So the you [00:39:00] know, the engineers go off, do all this work, bring it back and say, well, actually it doesn't work exactly that way.

Right. Um, so both time and quality impact, right. But if I can empower the business user to use the tool, then they can go directly from their head to the tool and they can work in a much more. You know, kind of organic, uh, manner, uh, than a centrally engineered manner. And I think that's super important to where we are today.

Ernst: [00:39:29] Excellent. So tell me a little bit about the, uh, process. If, if my business, for instance, wanted to use Aginity what are the steps? What's the process, um, you know, to get to an outcome.

Rick Hall: [00:39:43] Yeah. Sure. So, so the, you know, let's assume you've got this, uh, you know, you're trying to implement more and more analytics in your organization and you're, you know, you're thinking about how to do that.

So you probably either already have one of these big data platforms. You're, you're seriously looking at, [00:40:00] uh, looking at these things. Right. So, uh, what we're going to provide is tooling to help your business users, right? So, uh, you've got, uh, a team of engineers. And you've got a team of business people in those business, people have been on their own or using self service tools or doing it in Excel.

And you want to empower them with, uh, the, you know, a platform like ours. Now, the first thing for us is, look, we have a tool agility pro you can go to our website, download it and start using it tomorrow. Right. You don't actually need a massive amount of process to start trying this thing out. Right. So.

We want to make it easy for an analytic engineer or a business analyst to get started. And all they have to do is put that tool on their local machine and identify a connection to a database. So you do have to have a data platform in place in your organization. [00:41:00] That you can connect to, but we connect to all kinds of data platforms.

So it could be snowflake could be Redshift, could be a hive, which is essentially this big data Hadoop thing. It could be a Microsoft sin apps. I mean, we, we do a dozen of them today and we're going to be adding new ones all the time. So you have a data platform. You could just download a Aginity pro. And connect to it and you're off going and that's how an individual can start on their own.

Right? So our view is we're trying to make it really easy for individuals to do things. When there's a few individuals and many individuals that are trying to do similar things, then we move them from pro, which is this individual tool to premium and premium, uh, has a lot of the same functionality, but it creates a shared.

Environment where you can collaborate. Right? So we have this concept of a shared catalogue where an [00:42:00] engineer or an analyst or one analyst could share with another analyst, uh, to, uh, you know, to, to, to get their work. Right. What I think is the biggest difference outside of the, this, uh, this idea of empowering the, the analyst is that, and in how we did analytics for so many years, We had a central architect who defined the data architecture, you know, for, for analytics.

And that was all done and managed in a central fashion. Uh, if we're going to empower individual users, then in fact, uh, they're all going to create their own stuff. And if you just go that way, you can end up with chaos, 10 different versions of the same. Problem. Right. So you do need, uh, a team of data analysts who are call them data architects [00:43:00] who are actually going to help you kind of collate what the best examples are and promote those to becoming standard.

So I think of this as much more of like a biology metaphor. So I'm not going to engineer the perfect answer upfront. I'm going to let my users. Uh, do trial and error test and learn evolve, build their own stuff. And then I'm going to find a way to pick out what's really working and standardize it.

And then make it available for other users to, uh, to use. But that process of implementing, uh, that standardization, you don't have to start there. Right? You start by saying, I got a group of users, maybe they were power users who are used to doing a lot of work in Excel. Um, I've got this super powerful data platform and I want to see if they can use it.

Right. So we're going to give them Aginity pro we're going to let them. Uh, work directly against [00:44:00] that platform and see how it goes or, uh, and if I'm a business analyst sitting in the business, well, maybe I'm just going to go download Aginity pro, and then I'm going to call up my analytics team and say, can I connect to the database?

Right. And we have lots of users who do exactly that, right. It's happening. Uh, this democratization is happening because the need is so is so big. Right? Uh, and, uh, uh, so you evolved from individual, uh, power are users and engineers accessing the platform and doing their own work to an environment where you're collaborating to an environment where you're ultimately promoting the best outcomes of those collaborations into standards.

And, and that's kind of how I see the journey. Does that, does that make sense?

Ernst: [00:44:50] Yeah,

Rick Hall: [00:44:50] absolutely.

Ernst: [00:44:51] Absolutely. So where can people learn a little bit more about agility?

Rick Hall: [00:44:56] So our website, uh, [00:45:00] Aginity.com is, uh, you know, the simplest place to go. You know, like I said, you can download Aginity pro for free, so that's easy enough to do if you like it, uh, you know, you can pay for it.

And, uh, uh, you can start using more and more of it. And if you really want collaboration, then you know, you, you step up to a Aginity premium. So, it's on our website. We have an active, uh, um, you know, uh, LinkedIn, uh, social media set of activity as well. So, you can certainly go there, you know, I'm, uh, I'm on LinkedIn and I am often, you know, kind of posting ideas about how this kind of collaborative analytic, uh, world works.

And so, you can always kind of look at our stuff and. You know, reach out to me, but, uh, you know, we're trying to make this democratized as easy as possible. So, try it out. Tell us what you think. Uh, we're always learning about, you know, what our users really want and trying to evolve. We're producing a new [00:46:00] release or a software, like almost every month.

So, uh, hopefully we can help you out.

Ernst: [00:46:06] Excellent. Excellent. Rick, I really, really enjoyed the conversation. Thank you very much for your time today.

Rick Hall: [00:46:12] Thank you for having me.

Ernst: [00:46:14] Thank you.

The key takeaways for me during this conversation was to focus on a business problem first, make sure that you've got that business problem defined.

Second key takeaway is engaged somebody in your team who understands the processes and understands the business

And the third takeaway for me was

ensure you have a strong team.

that can coach the business

to drive those business outcomes. [00:47:00]