@msPartner, facebook.com/mspartner, linkedin.com/company/microsoft-partner-network

Today Microsoft is hosting an online event called Data Amp to share how to make the most of data to accelerate the pace of business innovation. This is a popular topic for businesses of all sizes, so our post today focuses on how to accelerate business innovation with data and analytics in small businesses. Many small businesses are interested in leveraging their data to increase their productivity and market share, but finding a starting point that makes sense is often difficult. Without large budgets to invest in data science, many small businesses think any ROI on their structured and unstructured data is simply out of reach. But learning to look the problem in a different way can help simplify the process and empower users for positive business outcomes.

Making Sense of the Noise

Understanding what Big Data means at the ground level is often one of the first challenges businesses come across. The discipline itself is changing rapidly and the utility of data science is unique for every business and every context it’s found at play. Industry verticals that are seeing significant changes in response to data and analytics include manufacturing, through predictive and preventive maintenance; financial services, through increased customer satisfaction and understanding; retail, through more personalized service; and healthcare, through enabling more positive patient outcomes and connecting widespread patient data.

Whatever the market, the business benefit of making sense of data is clear. But keep in mind that analytics need to enable action, otherwise they are just another expense. Businesses need to be able to gain insight that allows them to gain an edge over their competitors for the return to be worth the investment.

The Business Data Mindset

In the latest episode of the Microsoft Partner Network Podcast, we had the chance to sit down with CEO of Neal Analytics, Dylan Dias, to talk about the business of Big Data. Neal Analytics is a Microsoft partner focused on solving business problems with analytics and a management consulting perspective. Dylan was able to shed some light on what it takes for businesses to hit a fast-moving target like Big Data.

“What we like to say we’re good at is the modern data intelligence practice. We stay a level above the specific technologies, just because it changes so much. Instead, we choose to focus on the modernization of data intelligence, that way we can stay relevant.”
-Dylan Dias, CEO of Neal Analytics

Dylan said that it’s easy to become too focused on the technology and not focused enough on the business question users are looking to answer. Driving success from data resources requires strategic thinking and the knowledge that data is only of value if you collect it with the end use case in mind.

Tips for Big Data Success

1. Think about data as a key business asset

All information, whether it’s being collected or not, could potentially be used to answer a question or improve a process. Finding the value in the seemingly insignificant data points can be difficult. But that information when reviewed in tandem with all related data points can identify hidden trends and inconsistencies, and even provide reliable forecasts of what is likely to happen in the future.

2. Be strategic in your data collection

While it might be tempting to collect all data just in case it may be useful in the future, it’s critical to be strategic. Think about the business question you are trying to answer by collecting that data, and how it might be applicable to finding a solution.

“Oftentimes, we’ll be asked to come in and help a customer figure out what they want to do with their data. But I think that’s the wrong time to be asking that question. There’s a school of thought that says just collect all the data and we’ll figure it out. But, we come from the angle of driving ROI, so we always ask our customers that question first.”
-Dylan Dias, CEO of Neal Analytics

3. Fail fast and learn fast

Dylan said that coming from a background in management consulting, he sees the value in starting small and failing fast. With data and analytics practices, businesses often run the risk of trying to do too much. This inevitably leads to costly mistakes and data which doesn’t deliver the actionable insights that provide value.

“I come from the management consulting mindset which teaches us to start small and get wrong quick because you do run the risk of bottling the ocean, or of trying to do too much, which is not good.”
-Dylan Dias, CEO of Neal Analytics

To hear more from Dylan Dias about building a business on Big Data, listen and subscribe to the latest episode of the Microsoft Partner Network Podcast. This and past episodes are available for download on iTunes, SoundCloud, iHeartRadio, Google Play Music, and YouTube. Subscribe to get the latest episode as it’s released every week.

For a full transcript of today’s podcast, please see below.

Rachel Braunstein: Welcome to the Microsoft Partner Network Podcast. Every week, we bring in industry leaders and Microsoft partners to talk about the big ideas shaping business and technology today.

In today’s episode, we’re talking with CEO of Neal Analytics, Dylan Dias, about building a business on big data. Hey, Dylan, thanks for being here today. Great to have you.

Dylan Dias: Hello, Rachel. It’s nice to be here. Thank you.

Braunstein: It’s funny; Dylan and I were just talking about his morning. At 6:00 a.m., where you today?

Dias: So, I was in the parking lot of a large fast food chain, one of their local stores out here in Kirkland, and we were having some fun with applying cognitive services in the real world of fast food drive thru.

Braunstein: So, that kind of shows the possibility of analytics, right?

Dias: Absolutely.

Braunstein: And I’d love to hear more about you. I know you have an engineering background to consulting to Neal Analytics, so a little bit about you. And then what is Neal Analytics all about?

Dias: Absolutely. So, I like to tell people that I have one of the most fun jobs in the world. I get to apply a variety of different hats. So, I have an engineering background. I’m trained as an engineer. I spent a few years as a technology consultant. I had a detour into management consulting, where I went and got my business degree and then spent some time doing strategy. And the beauty of doing a job as a practitioner in the space of analytics and cognitive services is that you get to bring all of this together. There are some parts of the day in which you’ve got to be creative and kind of unconstrained to just kind of dream about the possibilities.

Braunstein: Yeah.

And other parts of the day where you have to be pretty precise and structured and detail oriented to make those possibilities come to fruition. And I can run the gamut, which is really a fun part of doing what I do.

So, Neal Analytics spends a lot of time playing in all of these different use cases and scenarios and technologies we talked about. But what we like to say we’re good at is the modern data intelligence and practice. Right? We stay a level above specific technologies just because it changes so much, but we understand that there’s a new evolution and modernization. And so, we focus on understanding that and putting it into practice in impactful ways for customers and for people.

Braunstein: How big is your company now?

Dias: We have about 50 people.

Braunstein: 50 people.

Dias: So, we’re still focused. We’re a boutique. I like to describe to our customers, right, we don’t hire by the truckload; we’re very selective. And we have a certain profile that we look for. It’s a balanced mind that can understand the variety of aspects that I kind of touched on in this conversation and bring that to bear in a practitioner.

Braunstein: I’m assuming you’ve had to kind of create a new job description. I mean if you were to look at a job description for a company like Neal Analytics 20 years ago, probably didn’t even exist. I don’t know. Did it? I imagine finding the right kind of talent that can do what you just said in terms of being analytical and think about business problems and customers.

Dias: Yeah. I would say so. I think people talk about the data science boom, but I think that’s in line with just having a new type of professional that can span a variety of form factors and boundaries and be effective.

Braunstein: Yeah, open up your mind a little bit, yeah.

And what exactly is analytics for you? What does that mean?

Dias: So, analytics it means a lot of different things to different people. But at the end of the day, it’s the ability to understand, to harness, to utilize, and to apply information or data in a way that is accretive to a business— in the increasing of revenue, in managing efficiencies into processes, or reducing cost. You have information. You can measure it and you can get what you inspect. Right? So, if you use that philosophy, you can have a continuously backed loop to keep improving and optimizing the business. So for me, analytics is the ability either through traditional techniques of kind of dashboards or EDWs, some of the more mainstream techniques or some of the more emerging techniques around applying structured and unstructured learning and/or kind of cognitive services or automation. It’s using this array of possibilities and technologies to make this promise of analytics come to fruition.

Braunstein: Can you talk a little bit about what that range looks like? If you are a partner and you’re a business looking and thinking, I’ve got to start doing this analytics thing. What is the range of things that you could start          thinking about or the potential opportunities?

Dias: Yeah. As cliched as it sounds, the possibilities are endless. It’s really true. People are saying these current technology enablers like IoT and machine learning or automation, bots, cognitive services, there’s quite a few of them, so it’s a bit of an alphabet soup. But if you just kind of put the array together and if you then find places to apply this. In things as simple as a financial function, right? So, one of the other cases we’re dealing with is how can you apply, for example, machine learning techniques for audit and risk and fraud detection to a financial function? To CPA firms and what they do on a day to day basis all the way to some things that are a little more sexy, if you may, where you’re trying to apply voice learning techniques to help with translation in a hospital. So, there’s such a gamut and as a partner, you typically want to begin where you have some area of strength. So, you’re dealing with some customers. You’ve provided some legacy services, potentially previously, and you understand their business. And then you have the opportunity to, with the knowledge of your customer’s business, apply some of these new enablers to dream up new possibilities.

Braunstein: So, you’re listening to the customer, seeing where your strengths are. Do you have any more information or kind of how to actually make this happen for partners that are really thinking about where to start?

Dias: Yeah. So, I like to think of kind of three areas. One is just capabilities. There are some markedly different capabilities that you need to bring to bear. People talk about data scientists. There’s a bit of management consulting.

Braunstein: Yeah.

Dias: There’s kind of advanced techniques in data engineering, with the cloud especially, similar to traditional on-prem but different in the fact that it’s a new hybrid approach. So, just kind of think about your capabilities. The second is kind of business smarts. I do think that a lot of these new technologies make the most sense in the context of use cases or scenarios. So, do you have familiarity with let’s say retail? Do you understand the merchandising function or are you in energy and do you understand kind of distribution? So, there’s a play in smart meters and internet of things, right? So, just having a business focus and being kind of declarative around where you can bring some net new skills I think is important. And finally, I think it’s just understanding your go to market. As a partner, there’s a lot of opportunity, but there’s also a lot of activity. So, understanding where as a partner you can shine in your go to market, be it relationships, be it some kind of associations that you have with industries. Identifying that and then doubling down on that is probably a recipe for success.

Braunstein: That’s great. What do you see happening in the industry? Where have you seen kind of industries picking up for analytics to be most impactful maybe?

Dias: Yeah. There’s a part of me which kind of recognizes that almost all industries are moving. I would say there are definitely some that are moving quicker than others.

Braunstein: Yeah.

Dias: The key drivers are there’s an organic opportunity with some industries. Like, for example, retail, right? Retail, there’s a lot of potential around understanding your customer better, understanding why people buy, how often they come back. When they buy a certain item from you, are they buying more than just that one item? So, there’s a lot of kind of inherent opportunity with retail. There are other places where the presence or lack of regulation can either hinder or support the adoption of analytics in the industry. So, healthcare, obviously, comes to mind. Finance is an interesting one. Some parts of finance have traditionally been very analytically inclined.

Braunstein: Yeah.

Dias: But there’s a lot of PII and privacy and information security and risk aspects that kind of help or hinder that. One new one that has been kind of popped up on our radar and I think this is pretty interesting is public sector. Public sector, governments, cities—cities are becoming smarter. The public sector knows that at the end of the day, they do have to serve us, the common man, and they have a responsibility as well as the opportunity. The unique opportunity—no one can get to the data that the public sector entities can, right? And so, there has been an uptick in activity out there. Airports are trying to get smarter to help you through the terminal better. Cities are trying to have better services on how they manage let’s say their billing relationship with you. So, the sky’s the limit, so to speak.

Braunstein: I love that public sector. I know we have a few partners using IoT to really look at the cities and lights and having the public citizens putting inputs. It seems that that especially with infrastructure, it makes sense that that would be a place where analytics and IoT would be a huge opportunity. What’s the journey that you’ve seen of big data? And where do you see the future? What are you excited about?

Dias: Absolutely. So, there’s a continuous evolution of these enabling technologies that I kind of referred to. I want to say maybe about three, maybe three to four years, your big data was all the craze, right?

Braunstein: Right.

Dias: I think between then and now, came a few of the technologies. I know right now, people are talking about automation and bots, right?

Braunstein: Yeah.

Dias: There’s cognitive services, so there’s a variety of        things that continue to pop up and will continue to pop up, right? So, I like to kind of boil it down to basic principles, first principles. I keep coming back to this, but I truly believe in this. There’s a business first kind of grounding in what you’re doing and why you’re doing it. So, do you understand the strategy? Do you understand the sector? Do you understand the underlying kind of pushes and pulls in a certain space that you choose to play? And then once you understand that, you understand what’s happening to funding to customer relationships to the ability to kind of drive a transformation. Another thing people talk about nowadays is a digital transformation, right?

Braunstein: Yeah.

Dias: So, just boiling it down to do you really understand that? And then with the evolution of technology, I tend to not get too caught up in technology. For me technology, even though we are a technology company, technology is an enabler. It’s kind of table stakes. You need to reinvest in individuals, so individually as well ad collectively as the company reinvests in just trying to keep on top of what’s happening. So, it’s a collection of internal kind of pilots, sandboxing, to just stay on top of the technology. So, that’s a given; you need to have that. That’s table stakes. But then, we try to bring it together with this good understanding of the external view of the world we are playing in so that both put together makes sure that we stay on top of things.

Braunstein: Yeah. What’s one of your favorite projects that you’ve worked on?

Dias: Wow, there’s so many. I’ll try to think of that. I want to say one of my favorite ones is going to be different. It’s going to be from education.

Braunstein: Yeah?

Dias: It’s not necessarily the most profitable. But it’s one of my favorites because of the potential social impact. So, I have educators in my family. Education is kind of near and dear to me. We did some work and I believe this is a publicized case study, so I can speak about it with the Cleveland School District. And the focus out there was kind of early intervention. The utilization of information to try to prevent students from dropping out of school. Once again, it’s a data problem. You know attendance records. You know if there’s lunches provided. You know if there are lunches that are skipped. You understand if the grades, of course, which we are central to education, are dropping. There’s a trajectory in that. So, there’s a lot of signals that are coming in from these students that walk through your classrooms every day and then as a whole, you want to obviously make sure that you hit the metrics. But each student is an individual. So, how can you have a macro view of what’s happening which potentially there’s already dashboards that do that? But then also how can you take a micro view of a student? It’s a human being. The unit of one is as important as the whole, right?

Braunstein: Yeah.

Dias: And so, applying big data and machine learning techniques, I know we worked with the school district to create some level of dashboarding, some level of modeling, predictive modeling to identify an at risk student and to create an intervention earlier in the cycle so that you don’t have to react after the fact, after the dropout has happened. And so for me, this is fun because you just apply the data.

Braunstein: That’s it. Yeah.

Dias: You did all of your predictive goodness, but you made it real in a true social and a real life way.

Braunstein: You’re talking really about not so much digital, I mean digital transformation, but human and human transformation and how can we use data and technology to help each other? And that’s kind of amazing. Right? That’s where the beauty I think we as Microsoft or really me as a human, Rachel, love being in this industry.

Dias: Absolutely.

Braunstein: Right? I mean those stories are the best.

Dias: Yeah, yeah. And we get the opportunity to have everyone live up to their potential as human beings, as companies, right, our partners.

Braunstein: Okay. Well, if the partners are starting to think about—I hate using big data—I’ll say #bigdata, what is kind of some your advice? What are things that you’ve learned to not do and things that you have to do?

Dias: I think beginning with the end in mind is important. Just having a good sensibility around what the outcome is, why we’re trying to do what we’re trying to do? What are the decisions you make? What are the technology choices you make? I think that’s critical. I’ve found that if you just jump in and try to embrace the technology, technology’s cool, don’t get me wrong. But just getting too over focused on the technology can lead into problems. We’ve often cleaned up work where other partners have been at play and we find ourselves asking the question, well, why did we build that Hadoop cluster? Or, we have all of this data in a big data store, now what do we do with it? And I think that’s the wrong time to be asking that question. There’s a school of thought that says, just collect all the data and we’ll figure it out.

Braunstein: Yeah.

Dias: So, there’s money on that approach as well. So, I’m not going to necessarily be completely counter to that, but we come from the school of thought where we want to generate ROI and create a value to our customers and so we ask the business question first.

Braunstein: Stick to the business question, listen.

Dias: Yeah, yeah.

Braunstein: Start with the problem, stick to what you’re trying to solve.

Dias: Yeah. And then the other thing is which I borrow from the management consulting side, which is kind of start small and get wrong quick.

Braunstein: Fail fast.

Dias: Fail fast is another way to say it. Because you do run the risk of boiling the ocean, of trying to do too much, and that’s not good.

Braunstein: Yeah.

Dias: So, if you have an outcome driven approach, if you narrow down what you’re trying to do to things that are useful and have a quick feedback loop, a learning loop, I think you’re fine.

Braunstein: And are there any resources or tools that you use or have used, your company?

Dias: Yeah. I encourage people, Microsoft has been putting a lot of effort into building out the Cortana Intelligence Suite, and so just hopping onto Cortana Intelligence Suite and just jumping off from there into a variety of different resources. There’s learning material. There’s certifications. There’s, I believe, even a data science course. There’s a lot of different aspects that can be brought to bear. And Microsoft has done a good job curating all of this. So, I would highly recommend that resource.

Braunstein: That’s awesome. Thank you so much, Dylan, for coming in today.

Dias: You’re welcome. Thank you for having me.

Braunstein: Thanks for listening today and check out the podcast description for show notes. Be sure to subscribe and keep in touch with us on LinkedIn, Facebook, and Twitter at MS Partner.