How North Labs Revolutionizes Data Analytics

In this eye-opening episode of the CTRL+Listen Podcast, we chat with Collin Graves, CEO and Founder of North Labs to uncover how his company is leading the charge in transforming the data analytics landscape. Discover the cutting-edge strategies and technologies North Labs employs to empower Data Leaders and Executives with unified analytics capabilities that deliver measurable Day 1 value. From cloud computing breakthroughs to digital transformation stories, hear how North Labs is not just a service provider but a strategic partner in data analytics.

Join hosts James Sweetlove and Joseph Passmore as they explore the journey of North Labs from its inception to becoming a top-tier, veteran-owned IT services group. Whether you're an industry professional or just curious about the future of business through data, this conversation is packed with valuable takeaways.

Listen to the Episode:

Watch the Episode:

Episode Highlights:

• Why Data Analytics Matters?
• How PCB Manufacturers Benefit
•  Data Analytics and Compliance Regulations
•  Is Data the Most Valuable Commodity
• Advanced Tier AWS Company Insights

Links and Resources:

Learn more about NorthLabs here

Transcript:

James: Hi everyone, this is James from The Control Listen podcast, brought to you by Octopart. I am here with my cohost, Joseph Passmore. Today we have a guest for you, Collin Graves. He is the founder and CEO of the fascinating company NorthLabs. Thank you so much for coming on. It's great to have you on the show.

Collin: Yeah, thanks, James. It's a pleasure.

James: Anytime. So, for anyone who doesn't know your company and what you do, do you wanna maybe tell us a little bit about the, the company, its story and sort of what it's all about?

Collin: Sure. Yeah. I've been in the cloud computing industry now for 17 years. I got my start with Amazon Web Services back when it was about six months old when I was turning wrenches on planes in the military and read about this, this new concept called cloud computing and what Amazon was doing with it and knew I had to be a part of it. So, totally self-taught from a software engineering standpoint. Don't worry, I'm not, I'm not the greatest software engineer in the world, just, just good enough to be dangerous. I'm the least qualified person at our company. But, you know, 17 years in, and the mission is the same. We help mostly manufacturing and industrial organizations with their digital transformation around cloud data. So that could be anything from, you know, unified analytics to machine learning and AI, predictive forecasting and that sort of thing. Really just to help solve business problems with proper data motions.

James: Fascinating.

Joseph: Why is data analytics so important to the future of business?

Collin: I think I have a, I have a philosophy that every problem is a data problem, right? You always hear this, this notion of you can't improve what you can't measure. And now that data is even more prominent, it's even more complex and verbose across any organization. I think everyone sort of feels this, this pull toward, okay, I know I produce a lot of data, I know that I can be gleaning better insights and better capabilities from that data, but I really don't know where to begin or how to take that first bite of, of the elephant, so to speak. And that problem, that that concept, that philosophy is only going to become more prominent over time because we're only adding more and more data, right? On an individual basis. We're producing more data than ever before. But if you look like at an industry like manufacturing, manufacturing produces something like eight to 10 times more data than the second ranked industry. And that's government. And we all know how much data the government has, right? So when you think about complex supply chains and complex industrial operations, that data footprint is exploding. And historically, organizations in that vertical have had a hard time deciphering between what's important from a data perspective to leverage today. And what's maybe important to think about a year or two or five in the future. So really our mission is to say, here's how you can get the most value out of your data today, and how we can structure that to bring continued value delivery in the future. I equated a lot to building a house. A lot of people really wanna think about the home theater or the upstairs game room. But until you have that foundation pour that that concrete is set and that rebars put in, if you focus on that game room upstairs, at best, you have an expensive r and d experiment. And at worst, you actually build something that can't stand the test of time and becomes a huge sunk cost to the organization. And that's what we see all the time at North Labs is we're three years into this, we've spent $6 million on it, and we're no more mature today than we were when we started. And, and really that's the mission that, that we're, that we've set out to solve.

James:  So would you say that there's sort of a, a problem where data is an afterthought as opposed to something that's the key part of the equation from the very start?

Collin: I think so, and it's nobody's fault, right? We're shoehorning in this philosophy. When you think about manufacturing, industrial, or really any organization, most of these organizations have been running for years or decades prior to this data conversation entering the chat room, right? So you have your existing operating procedures, you have your existing ways of life, I call it existing laws of physics, right? And we're essentially coming in and saying, that's not good enough. You have to now sort of build on this, this concept of data maturity. But a lot of organizations are, are to the point where they go, yeah, that all sounds fine, but how, how do I start? Right? I already have the house built, you're telling me to tear my house down and and build a new one, and that seems like a lot of effort and, and I need somewhere to sleep at night, right? So it's not like you have an emerging industry where you can grab new philosophies and new concepts as you go. These organizations are trying to retrofit data maturity into the existing equation, which is a more difficult conversation to have, but ultimately, those who are committed to it and find themselves at least a little bit successful with it can outperform industry averages and outperform their peers as a result.

Joseph: I noticed on your website there's this phrase, single source of truth. Could you expand a little bit on what that means to, to you and to North Labs?

Collin: Yeah, and it goes back to that, to that home building equation, right? If we think about an average organization today, and whenever I use examples today, it'll be industrial related, but this applies to, we've gone from an average organization having between 18 and 24 sources of data 10 years ago to now. Today we have well over 100 in an average organization. All of those systems produce a data footprint. They're all keeping track of records, logs, whatever the case may be, but all of them are structured slightly differently. And so it becomes really difficult to analyze across those systems if you have two different laws of physics at play, right? Your old, you know, Microsoft database backed ERP versus a new cutting edge marketing tool or CRM, they're both producing data, but they're doing so in different ways. And so it becomes very difficult to be able to analyze across those systems in a meaningful fashion. And so when I say single source of truth, whether that comes in the form of actually ingesting everything into one place or putting those transla translation layers in place, so when I ask a question, my systems know what I mean. That's the most important thing because the value of data and problem solving in your business comes from the ability to ask questions of your business in a sort of cross pollination type of manner, right? If I just ask questions of my ERP, I can only get answers that are from the perspective and the data of that ERP, but that might not be the whole answer. And so if I'm able to unify these systems together so I can ask a question once it can go out and pull from multiple systems, now I have a much higher likelihood of finding answers that I wasn't previously privy to. And that's where a lot of this sort of data maturity and capability is unlocked.

Joseph: Is this something you're using AI new advancements in AI to, to help solve?

Collin: We can, there are applications for it for sure. I think AI is amazing, right? I'm, I'm a huge tech nerd. I've been a technology, you know, aficionado for a long time, but you're, you're right in calling that out, Joseph, right? Everyone today is interested in leveraging AI, everyone, right? That is the sort of, it's the topic du jour for sure, and it will be for a long time. The issue is, if we skip ahead in the book all the way to AI and the first five chapters, were all about laying that solid foundation and making our data capable of being ready for AI. Now we vastly, severely limit ourselves and the capabilities we can actually introduce to the company. So what we find today is that groups are adopting AI, but they're doing so in that siloed capacity, this system over here, that system way over there. And while that's great, don't get me wrong, the most advantageous thing an organization can do today is to be able to use AI across all of their systems on that unified source of truth, right? What does it mean for me to use a, not, you know, inference or predictions on all of my data as opposed to one sliver of my data? And so a lot of groups come to us and say, we have this great vision for AI, we wanna be using it a year from now. And my encouragement to these groups is take the time to sharpen the ax before you start chopping the tree down, right? If you can unify that data before you leverage AI, you're gonna find yourself many steps ahead of where you would otherwise be. And most of the time, to be honest, when we go through that unification effort and give those even more basic analytics capabilities to organizations, the light bulbs start going off. Right now I have visibility across all of my systems. Now I can ask questions when they come to mind. Now I have all of my systems updating in real time so I can have a, a real time view of my industrial operations. This is more than enough for a lot of groups, and some will still want to dip their toe into AI, don't get me wrong. But if we just jump all the way into AI, we're gonna find ourselves hitting our heads against the ceiling in fairly short order. Whereas this really helps us pull back the, the slingshot and create that kinetic energy for, you know, really meaningful AI adoption in the future.

James: Just to bring it back to our specific audience who listens to this podcast, I want to ask like a sort of a focus question. Obviously you work in the manufacturing space, but, so let's say looking at a PCB manufacturer, something along those lines, how would they benefit from investing in this type of data modernization?

Collin: It's a great question. And we see this all the time across, you know, other, other realms of electronics engineering, CAD design, that sort of thing is if we can, if we can take lessons learned from the past that maybe we didn't even extract from our systems with humans, right? But if we can unify our data and have AI take a look at designs, take a look at specifications, and pair that with quality control efforts, supply chain efforts, you know, end product feedback from our, from our teams or from our customers, we put ourselves in a position to really understand how to support and build in the future, right? We all know that PCB design is one that, you know, you're, you're typically building something pretty darn novel. I love the idea of, of, you know, PCBs, my father was an electrical engineer for, for many years, building pacemakers for, you know, a global 100 organization before he went out on his own. So that's near and dear to my heart. So some of this design oriented stuff, whether it's PCB, whether it's architecture, whether you're making furniture, whatever the case may be, with unified analytics, you have the ability to combine that whole value chain from idea through to production, through to delivery to the customer's door to understand where your weak links are in the process and optimize your throughput as an organization. Obviously you can use AI to help with the design itself, but I view that sort of macro concept of that whole value chain being optimized as the most valuable thing an organization can focus on.

James: Great. Thanks.

Joseph: How can a strong data core help businesses react in real time to world events?

Collin: That's a great question. Obviously the low hanging fruit in manufacturing and industrial is your supply chain, right? And typically supply chain has been a very reactionary process, right? So we're having to learn about an event process, that event, and then get together in the morning and go, oh my gosh, Joseph, this is happening in this region of the world. What do we do? We're about to run out of products. We have an order to that needs to go out in 48 hours. What should we do? And it's flipping the script to proactive motions, right? So if we can really understand our supply chain and use proper forecasting, use proper optimization analysis and real time insights, we put ourself in a position to not only where we can make decisions faster, but we work with many customers who now are able to take automated actions, right? Without a human in the loop for, for their supply chain, right? So there are systems today that you can integrate with that can call out, you know, global, you know, stuff that's global events that are occurring. You can use that data to say, all right, we should pivot our order from this region 'cause it'll be delayed to our backup supplier over here, where maybe we pay four more, four additional points on the product. But we know that our opportunity cost is going to be higher if we try and wait for that order to be fulfilled during this event. So really for us, it's all about, you know, taking as much of the reactionary nature of a complex business and making it as proactive as possible.

James:  And can, can data analytics and AI help companies deal with these increasingly complex compliance regulations that are being brought in?

Collin: Absolutely. We work on that stuff all the time, right? So when you think about cloud and data, there's a natural sort of ancillary motion of security, right? But security can mean, right? It can mean intrusion detection, sure. But it can also mean are we compliant and, and well governed as an organization. And so using AI to automatically keep track of that compliance posture for your organization has saved our customers hundreds or thousands of hours per year. Whereas otherwise it would've been a, a manual effort of Joseph going in and having to, you know, modify a, an Excel spreadsheet with asset tags and last review dates and things like that. Now we can sort of use AI to keep track on things in a more real time manner, which is great because the sooner you can detect compliance drift, the easier it is to fix, right? So if you make a change and your compliance posture changes, you wanna be notified as quickly as possible so you can go in, in the same context that you're currently in and go, oh yeah, I should have, I should have done this instead of this, right? The last thing you want to do is learn about that at a quarterly review or during an actual, a formal review because now you have a lot more pain because you know, your job, your job performance is seen as suboptimal, right? And you put your, you put your organization in a, in a precarious position. So we call that automated compliance posture and it's something we see being used a lot with, with ai.

James: Fascinating. 

Joseph: Could you tell us about any trends you've seen in the US data analytics market?

Collin: I can, yeah, obviously Gen AI is the talk of the town from everyone and their aunt, right? So, and it's a, that's a blessing and a curse, right? Because anytime you have this sort of hype cycle taking place, it does drum up interest in the field I work in, right? So we get a lot of customers reaching out and saying, Hey, we wanna leverage gen AI for this use case within our organization. But again, it goes back to how mature are we today as an organization? Because if we deploy large language models onto data that is of suboptimal quality AI has the potential to make matters worse, right? Because Gen AI is just sort of teasing the boundaries of the data that's laid out. And if there's a wide dispersion of quality within our data sets, you can get some some pretty gnarly outputs from large language models because they're not designed to sort of double check things and go, does this seem right? Right? It'll sound as confident as you want it to, but it doesn't mean it's going to be accurate. So we've, we've had a lot of customers reach out with that sort of narrative in mind. We wanna leverage gen ai, we wanna privately host our own large language models with something like Amazon Bedrock. And the conversation always goes and should always go back to, okay, but let's take a current state analysis of the foundation of our house, because now we're talking about that upstairs game room, right? And so it's a bitter pill to swallow for some organizations who realize that, hey, yeah, we have probably a year of work to do before we can really leverage Gen AI. You can do so in a sandbox environment, you can kick the tires and explore the art of the possible, but if this is something you really want to put into production it, I always recommend that we're starting with those foundational bits first, because otherwise you're just limiting your potential in the future.

James: Would you say that data itself has become the most valuable commodity in the world these days? I've heard that said,

Collin: Yeah, data's the new oil, right? That's, that's everybody had, that's what everybody says. Yeah. I think, I think data's important, but I think organizations that will see the greatest leap in enterprise value are not just those who have a ton of data, but those who have organized it well and orchestrated it well and have the, the potential to leverage it well. Right? And that's a big distinction. If I just collect all the data, like their baseball cards, that's fine, right? But if I can demonstrate that I have a mature operation with my data and can leverage it to make decisions both with humans as well as automatically, that's where we're gonna see large leaps in, in enterprise value. And I think we're starting to see that play out in the private equity markets. It's not just about collect at all just to have it anymore. It's about how do you collect it and then rationally leverage it for, for your organization's growth. So there's work to be done on that front for a lot of groups yet. But the good news is, right, anyone can begin taking steps toward that maturity.

James: And do you think that's an area where AI is going to really help?

Collin: Absolutely. Yeah. I think for me, what's most interesting with AI is not necessarily how am I gonna apply AI to my end product, to my end customers, or anything like that. I think where AI has the shortest term benefit to most organizations is to help in that decision making around where I should be placing my focus with my data, right? If it can help identify gaps in quality, if it can help identify gaps in my ability to reason across my, my core business value chain, right? That's where we see a lot of promise. 'cause at the end of the day, think about it, manufacturing, we could apply AI to customer service, we can apply AI to marketing, you know, yada yada yada. But what's the core mission of a manufacturing group is to take raw material, convert that into a finished product and get that throughput out the door into the hands of customers, right? That's the core mission. Anything that's ancillary to that is a nice to have as far as I'm concerned. And maybe it's just 'cause I'm the, the old, old Greyhound at the track, right? But if you can leverage AI to improve your core mission, your core value chain, you put yourself in a, in a much greater position relative to your, to your market peers. And so that's where I think AI has a lot of short term upside for most organizations is help me with my decision making. If I have 150 systems that produce data, help me figure out what 20 or 25 systems are most important to be focusing on. Because if I try and boil the ocean across all 150 systems, I'll get nowhere. Right? Right. So if I can take that subset and focus a lot of time and on it, right? We can, we can show where our, our sort of investment potential might be.

James:  Yeah. Makes total sense.

Joseph: Could you tell us about being an advanced tier AWS partner?

Collin: Sure. Yeah. So, like I said, I've been in the AWS realm since 2007. It's been a, it's been a long road. I was around well before there was such a thing as Amazon Partners was one of the first 10 fully certified AWS folks in the world. And Amazon has really pioneered the way for the cloud hyperscalers and what it means to run a partner program. You know, you find elsewhere in the market that you know, who, you know, politically is really advantageous. It's easy to grow in the ranks wi within other cloud providers. Amazon has always set the standard for, you know, what it means to have a meaningful partner program. So they have four tiers. You have registered, you have select advanced and premier. And it's really difficult to grow in those tiers. It's not just about, Hey, I landed one big project and they said, attaboy, here's your next rank. It's really about a, a lot of different motions going on all the time from marketing to how many certified staff you have in place to demonstrating actual auditable capabilities internally to Amazon, right? So any designation you have as a partner goes through an audit process where they actually sign off on, Hey, these guys know what they're talking about, right? They haven't just checked some boxes and now expect to, to receive this, this cool badge. They go that extra mile to make sure that every partner that advances is really, really worth it. As far as advance versus Premier, that's sort of the top of the mountain. Most partners for the record will, will stay on the, the other half of that partner ecosystem. So either registered, which you get when you sign up or select, which is actually, you know, it's not super difficult to get, but you have to actually put work in that jump to advanced and Premier is quite difficult. And really the only difference between advanced and Premier is the size of the organization. Advanced tier partners and Premier Tier partners still qualify for the same programs. Amazon loves to contribute, you know, funding toward customer initiatives. So they'll actually help fund our work for a customer and they'll contribute marketing dollars for our activities and stuff like that. So it's been a great, a great ride. Obviously we haven't spread ourselves around to the other cloud providers and don't intend to, we're sort of all in with AWS we'll support other clouds if they're already in place within a customer organization. But in terms of our sort of go to market motion, we focus almost exclusively on Amazon. And when you're in the world of manufacturing and industrial, who's, who's a better representation of what it means to have an optimized supply chain than Amazon, right? Yeah. If you ever walk into one of their fulfillment centers, it will absolutely blow your mind. It's like a, a theme park with roller coasters and autonomous robots and lasers and just the most insane technology. They're just, they're just missing the, you know, your autonomous policing robots. From episode one. But it's, it's really remarkable. And so that's actually become a big marketing motion for us, is to bring executive thought leaders to an advanced fulfillment center with Amazon where they're leveraging these advanced AI capabilities and saying, right, you know, you're not gonna get all the way to Amazon tomorrow, but what, what's that little sliver that you could take away from this to bring your organization to the next level in terms of its throughput? And that's been really special. So proud to be a partner. I think we're one of five veteran owned AWS partners in the, in the ecosystem. And we're one of very few that's, that's certified in data and manufacturing. So we have a very specific focus within Amazon, which allows us to differentiate ourselves from the other partners.

James:  Wow. Congratulations. That's, that's really cool. Thanks.

Collin: Yeah, it's, I'm glad I, I'm glad I hitched my, my cart to the fastest horse on the track because, you know, it's, people like to, you know, think I knew what I was doing and it was really just, it was a gamble and, and it paid off. And I don't take that for granted with them. We, we really spend a lot of time on that, on that relationship. But they've been great to us.

James:  That's great. I had one more question I wanted to ask that I only just thought of actually. I think it's a good one. What would you say is the main area that manufacturers, I guess, overlook as far as data management goes?

Collin: That's a great question. I think the biggest focus for us right now is putting real time operational analytics into the hands of the executive suite or key players in an organization across IT information technology systems and ot, operational technology systems, right? In manufacturing, historically those have been two very siloed off parts of the business. They know separate networks really locked down from a security perspective. 'cause the last thing you'd want is someone to hack into your, you know, email server and take down your plant, right? So they've really kept those cordoned off historically, but now we're seeing a growing motion in Joseph talked about that single source of truth, right? That unifying motion within these organizations of being able to combine data from your IT side of the house. So your ERP, your CRM, whatever the case may be, and your OT system. So your actual, you know, machines, your IOT sensors, your asset tracking devices, MES and SCADA systems, right? Whatever the case may be. Unifying those together gives the potential for a manufacturing organization to have a real time view of how their organization is performing. Not just, Hey, here's what we did last week or last month, which is how a lot of organizations still operate today. I get my Excel spreadsheet, here's last month's numbers. But imagine being able to sit down as an executive and, and say, here's what our throughput was in the last hour. Here's what our defect rate was in the last hour compared to the hour before. Here's what it's forecasted to be this next hour because this machine is causing issues. It's not yet time for maintenance, but maybe we ought to have a slow down and take a look at it because that machine is causing defects. It gives you the ability to make decisions on the fly without having to wait until the next day, week, or month to turn this, the wheel of the ship, right? Because most organizations navigate like this. But if you can unify those systems and have real time operational insights, you have a lot better chance of being able to navigate in a straight line. And that's ultimately what our mission is at North Labs.

James: Fascinating. Well, that was all the, the questions we had for you. So just to wrap up, if someone wants to check out your company, see your offerings, contact you, what's the best way to do all of that?

Collin: Sure. I'm on LinkedIn. Colin Graves, obviously our company is North Labs. You can find us with a quick internet search. But one thing that I really like about our, our organization is for those companies that are just starting to plan what their data future might look like, what their data maturity could be, we actually architect a plan for them free of charge. So it's our way to provide a lot of value for these organizations to say, look, whether you use us to build or not, we at least want you to be successful in the steps you take in the next 3, 6, 12 months. And we design that plan for them, you know, free of charge. So for those interested in that, if there's anyone listening, you can go to build my data blueprint.com and engage with us there. It's a four week effort where we sort of peel back the layers of the onion for these different groups, really try and figure out where their biggest pains are, and put a plan in place to help them solve those pains so we can help eradicate this 80 to 90% failure rate that we currently have in our market of organizations. Failing to put a data program in place that returns predictable. ROI. So that's my, that's my teaser for your listeners.

James: Wow. That's a fantastic offering. It's really cool. Well thank you so much for coming on. It's been absolutely fascinating talking to you. It's an area we haven't really covered as much on this podcast, so it's really great to finally get across this side of things.

Collin: Yeah, love what you guys are doing with the podcast. I'm, I'm gonna be a, a continued listener. I think the content's been been outstanding, so hopefully, hopefully listeners, you know, gained a little something out of the, the conversation today. But keep doing what you're doing because it's, it's really special.

James: Thank you. Thanks so much. We appreciate that a lot, much. And for anyone listening at home, just come back next week and we will have another guest for you.