The Interconnectedness of Things

Tackling Turnover with Smart Automation: Upholding the Mission in Government

QFlow Systems, LLC

Share Your Thoughts With Us!

In this episode of The Interconnectedness of Things, host Emily Nava sits down with Dr. Andrew Hutson, COO of QFlow Systems, to explore how federal agencies can stay mission-focused amid workforce challenges like staff turnover, early retirements, and shrinking budgets.

They unpack the concept of “corporate amnesia” caused by institutional knowledge loss and discuss why traditional processes struggle when resources dwindle but mission demands remain constant. Dr. Hutson outlines the “power triangle” of standardization, automation, and integration as a pathway for agencies to modernize without scrapping legacy systems.

Listeners will gain insights into what smart automation really looks like, how case-based, rule-based, and probabilistic decision systems align with today’s AI and workflow automation, and why clean data and oversight are non-negotiable. From practical use cases like loan processing to the cultural shift required for adoption, this conversation highlights how technology can help federal employees “do more with what you have” while staying true to their mission.

If you’re a CIO, IT leader, or public servant navigating the future of government work, this episode offers a grounded, hopeful perspective on the role automation can play in building resilience and continuity.

About "The Interconnectedness of Things"
Welcome to "The Interconnectedness of Things," where hosts Dr. Andrew Hutson and Emily Nava explore the ever-evolving landscape of technology, innovation, and how these forces shape our world. Each episode dives deep into the critical topics of enterprise solutions, AI, document management, and more, offering insights and practical advice for businesses and tech enthusiasts alike.

Brought to you by QFlow Systems
QFlow helps manage your documents in a secure and organized way. It works with your existing software to make it easy for you to find all your documents in one place. Discover how QFlow can transform your organization at qflow.com

Follow Us!
Andrew Hutson - LinkedIn
Emily Nava - LinkedIn

Intro and Outro music provided by Marser.

Emily Nava: [00:00:00] Hi everyone. I'm Emily Nava, head of marketing at Q Flow Systems. I'm joined here today with Dr. Andrew Hudson. Um, he needs so introduction, but the COO of Q Flow Systems. 

Dr. Andew Hutson: Hey, thanks for having me. I'm excited for our topic today. 

Emily Nava: Yeah, absolutely. Um, so we're talking about what it means to uphold the mission even as federal agencies face some staff turnover, shrinking budgets and growing demand.

Emily Nava: Um, so I'm just gonna jump right into the questions here. 

Dr. Andew Hutson: Let's do it. 

Emily Nava: Um, so let's start with what you're seeing across government right now. Um, what are some workforce trends that you've observed, especially around staffing and workload? 

Dr. Andew Hutson: Yeah, I, I can't be too specific. Uh, just, uh, the nature of the [00:01:00] type of information I have access to is likely not public, and I certainly wouldn't want to put any of my friends in an awkward position.

Dr. Andew Hutson: But I can talk about the broader trends and we can see the immediate impact Doge has had on staffing, not just with layoffs. But with early retirement, that seems to be, at least with our interactions, the biggest disruption that agencies have been facing right now. At the same time, while the number of resources folks, people, humans, that are available to work.

Dr. Andew Hutson: On specific mission areas has decreased. The scope and size of the mission areas have not, and that's presenting some very burdensome [00:02:00] situations. And it made me think about a term that probably goes outside of private sector and works just fine here with public, but a term called corporate amnesia. Mm.

Emily Nava: Mm-hmm. This is something 

Dr. Andew Hutson: that was, uh, talked about a lot with layoffs or with an aging workforce, that there's just all of this knowledge that goes away as folks retire or staffing decreases. The extent to which that that happens is variable, but I'm starting to see it with our federal customers, where folks just.

Dr. Andew Hutson: Don't know exactly who to go speak to yet. They're uncertain on the processes that they used to know. And are they still valid today? Uh, and tasks that have traditionally been assigned to the roles that folks have, uh, they [00:03:00] haven't been trained up on how to do them. That's because they were reliant on others that could own different portions of the processes.

Dr. Andew Hutson: So all these are kind of combining together and it can create delays and errors and, and even interpersonal conflicts where folks have mis or, uh, disparate expectations of what people should own and how they should be showing up. 

Emily Nava: Yeah, I mean, you answered a few of my questions with that, um, great synopsis of, of what you're seeing out there.

Emily Nava: Um, so how are, um, traditional, or I guess why are the traditional processes that might've worked, um, when they were fully staffed, why are they struggling to keep up now? 

Dr. Andew Hutson: Oh, that's a great question and I won't take the immediate example. I'll, I'll go back. Years to some work that I did with the state of Alabama and their Medicaid process.[00:04:00] 

Dr. Andew Hutson: When something changes on the demand or the supply within a given system, it has to compensate somehow. You can't continue to do the same process in the same way if demand or resources change. So the Medicaid example, uh, with a new legislation that had passed at the time, the number of eligible, uh, folks, citizens in Alabama that would be covered by Medicare and Medicaid, doubled the processes that they had in place, were there to keep up with the current workload and didn't scale very well.

Dr. Andew Hutson: If you go to the opposite of that and say, I had a process where the demand is continuing the same, but the number of resources applied to that process has decreased. You have a similar [00:05:00] problem. It can't scale to the increased demand. So one option is to decrease the supply that you're providing on this particular process.

Dr. Andew Hutson: By the same rate that your number of resources has decreased. But if your mission doesn't change, you can't really do that. So then you have to find alternative methods to capture what had traditionally been done manually to try to be done using something else. And I am alluding to, yes, some type of computer automation.

Emily Nava: Automation why we're all here. Right. Um, so what does smart automation look like in the public sector? It can sound kind of intimidating or abstract, so, um, what could that really look like? 

Dr. Andew Hutson: Well, I'll, let me ask you some questions. So I'm gonna go back to. [00:06:00] Graduate school and three types of decision support systems that existed out there.

Dr. Andew Hutson: And for our purposes, we can tie this back to automation. So the first kind is called case based decision making. Can you imagine what that means? 

Emily Nava: I imagine it means, um, looking at past examples of something similar. 

Dr. Andew Hutson: That's exactly right. Nailed it. First try. Nice. 

Emily Nava: Uh, 

Dr. Andew Hutson: the, so it's basically saying here is a fully divine event, a case, and how closely does this current event or case batch, okay.

Dr. Andew Hutson: So it's a similarity comparison. The next is rule based. What do you think that means? 

Emily Nava: If this happens, this happens always. Mm-hmm. 

Dr. Andew Hutson: Perfect, perfect way to say it. If this, then that, uh, I think there's even an app out there that's called that, um, [00:07:00] very clear, very black and white. Here's a rule. It's either followed or it's not very nice for computers.

Dr. Andew Hutson: Not always great for processes. 

Emily Nava: Nope. 

Dr. Andew Hutson: So now we have the third kind, probabilistic based. What do you think that means? 

Emily Nava: Um, probably a combination of the two. Um, looking at a larger sample of similar cases and deciding the best course of action based on those outcomes. 

Dr. Andew Hutson: That's pretty, pretty good too. You almost like you're an expert.

Dr. Andew Hutson: Uh, the probability based systems are saying, what is the probability of the next best step given past rules and cases if we fast forward to, and that's like [00:08:00] 20 years old. Okay? Yes. That far back. If we come to today and we think about. LLMs and we think about automation, um, we think about how those things work together.

Dr. Andew Hutson: They're really kind of still the same. So automation on a process is a step by step or a state based, where we're saying perform action. Then based off of that action, go to the next step. And that could be a decision point or a very clear one directional step. And we, we can start to build this image in our head of steps and triangles and business process notation.

Dr. Andew Hutson: That works great for rules. And then for cases, we have something like a vector database where we can put all of our cases into a database and give it [00:09:00] a, a new case that is unique, uh, in and of itself, and it will pull back similar cases and uses something called co-sign similarity. Then you go to an LLM.

Dr. Andew Hutson: Well, the entire framework of a training model for LLM is using. Probabilities LLMs work by saying, based off of the context provided, here is the most likely next token. Or we can think of it as words or word parts that should be generated based off of that information. And you can pull all these together and with rules and with cases and with vectors, and with LLM start to create smart automation.

Dr. Andew Hutson: That can combine these things together to create a desired result that may be similar to or close enough to how a [00:10:00] person might perform a similar activity. But can you guess where this falls apart? 

Emily Nava: Falls apart, um, when there's no oversight. 

Dr. Andew Hutson: That's a great one. No oversight. Uh, which means. The system is just allowed to, to, to chug along, and no one is saying, uh oh, oh, oh.

Dr. Andew Hutson: Uh oh. No, no, no, no, no, no, no, no. That's not right. Do that. No, no, no, no, no. Uh, it can also go really bad if the cases are incomplete or incorrect. 

Emily Nava: So, bad data. 

Dr. Andew Hutson: Yeah. So bad data hurts this, and, and guess what? People traditionally are not really good at. When they're doing a process 

Emily Nava: keeping data clean. 

Dr. Andew Hutson: Yeah.

Dr. Andew Hutson: They're not really good at writing down all the instances where this did or didn't work. Uh, and we're really seeing that back with our, our [00:11:00] customers. And this is rampant, whether it's private or public industry. So we're not picking on anybody but the. Ability for these smart automations to work effectively if it's predicated on how much has been captured and how well it's been captured.

Dr. Andew Hutson: Well, if everything's in an an, an email inbox, for example, and no rhyme or reason, uh, or worse. There's tons of rhymes or reasons depending on which department or missionary you're in. Or 

Emily Nava: the person 

Dr. Andew Hutson: or, or, yeah. Then we even have the person anomaly where, and we've all probably done this, oh man, this was a really good idea six months ago, but I wanna switch it up now.

Emily Nava: Yeah, this process works for me, but 

Dr. Andew Hutson: this process works for me. Yeah. How many developers have we talked to and say, well it works on my machine. 

Emily Nava: Yeah. 

Dr. Andew Hutson: Uh, all that just makes it murky and difficult to try to fix, and that's why educating. [00:12:00] On how these systems can work is that much more important because it will change the behaviors on all these ancillary activities and data capture points that everybody's working with to help make that smart automation better.

Emily Nava: Right. Well, I mean that sounds all well and good, but a lot of agencies aren't. Set up to have smart automate automation for all the reasons that we just listed. Um, so how can agencies adopt this smart automation without ripping and replacing legacy systems just altogether? 

Dr. Andew Hutson: Well, it gets harder, right? Um, so the first thing that they have to consider is standardizing to some extent.

Emily Nava: Mm-hmm. 

Dr. Andew Hutson: So. That takes [00:13:00] commitment and discipline to create the habits for standards that might be more plausible in a public sector setting. Uh, but it, it isn't guaranteed. It really does take leadership and buy-in from those performing the work to follow those standards. But standards aren't enough because you have to integrate.

Dr. Andew Hutson: With these existing systems that are out there now to make sure that standards are consistent and you can pull in the relevant information across these different products. Once you have that set up, uh, you can introduce something called automation. Now, you may go out of sequence and say, I'm gonna start with Automation Hudson.

Dr. Andew Hutson: I'll work on these other things as I go and that's fine. You can even use automation to help start to enforce [00:14:00] standards, to create consistency and processes at a, at a low level. You can use automation to and rules to ensure that data entry is consistent across your disparate systems. But the, the kind of the trifecta, the, the power triangle.

Dr. Andew Hutson: That I'll say right now is that standardization, automation, and integration to help folks actually get value from their data and help them with their work. 

Emily Nava: Absolutely. Um, so you mentioned a few data entry, um, standards enforcement, um, I guess you could call those workflows. Um, but what other types of work workflows would be ideal to start?

Emily Nava: Automating for, let's say an agency that's ha hasn't before. 

Dr. Andew Hutson: Uh, so then you, you start to look at the low, you, everybody's already heard this, right? The low hanging [00:15:00] fruit. Mm-hmm. What does that mean? Um, you look for those things that are already prone to consistency. So I'll give you a healthcare example.

Dr. Andew Hutson: Uh, anesthesiology. That is, uh, a perfect example often used when we're talking about automation and systems because the calculations and dosages and approaches and metrics and monitoring all are consistent, and because they're consistent and repeatable re in multiple settings, it can be monitored and automated and standardized much more quickly.

Dr. Andew Hutson: But if, let's say we go into something like a loan process, same deal. Mm-hmm. The steps by which a loan process has to follow are relatively unchanging. Uh, you still have to meet certain. Thresholds and criteria. There's [00:16:00] certain paperwork that has to be filled out. There's updates along the way. There's follow up.

Dr. Andew Hutson: There's people that are involved, there's approvals. All these different things all have to happen in the same way each time. So that's a perfect opportunity to say, let me integrate with the systems. Let me codify the process, and then let me establish the standard. 

Emily Nava: Gotcha. Okay. Yeah, so they can just look at their processes overall and say like, this one, we've actually got it written down to some extent, so let's just automate these, these phases.

Emily Nava: Um, and like you mentioned, the loan loan processing, of course there's gonna be anomalies, but that has to hit these certain milestones every time. 

Dr. Andew Hutson: Mm-hmm. Anomalies are the thing that always comes up. Right. It always. Almost feels like that's the, the, the end of the conversation. Uh, and that can feel that way a lot, but that's, that's really majoring in the minor.

Dr. Andew Hutson: Anomalies will happen [00:17:00] and you need to be able to account for them. But are, if anomalies are a majority of what's happening, they're not really anomalies anymore. 

Emily Nava: Right. Then like maybe you need to take a look at where those are happening so much 

Dr. Andew Hutson: and, and so maybe you can only automate 80% of your process because everything else needs to be unique to the circumstance.

Dr. Andew Hutson: Well, that's still 80% that's automated. 

Emily Nava: Exactly. 

Dr. Andew Hutson: And it leaves you room for that other stuff. And then you might find that. There is actual room for automation in that 20%, but you couldn't see it before because the 80% was distracting you. 

Emily Nava: Yeah. We 

Dr. Andew Hutson: call that the, the whirlwind here internal a lot. The, the whirlwind of the day to day can distract from the improvement you need.

Emily Nava: So I kind of wanna go back to what we started talking about, beginning of upholding the mission. Um, so the mission doesn't end. Doesn't it? One, if people are laid off or retire early, that doesn't mean the [00:18:00] mission, like you said, changes. It certainly doesn't end. Um, so that's something that really resonates with, with federal workers.

Emily Nava: Um, but what does that mean and how does a company like Q Flow help agencies stay on track even when those resources are limited? 

Dr. Andew Hutson: One of the great things about. The folks we get to work with is their dedication to the mission that they have knowingly and willingly dedicated their life to serving the American people.

Dr. Andew Hutson: Many of the folks that I meet not only work for the agencies that we work with, but they started off. In the military and they have a, a profound sense of duty to the mission at hand. [00:19:00] And so when you speak with these folks and they say the mission doesn't end, it isn't a throwaway statement, it is impactful.

Dr. Andew Hutson: It is them saying it does not matter that we don't have the resources to make this easy. The mission doesn't end. They feel this incredible responsibility to the American people. And so when we talk about impractical terms for the federal agency, there's so many things that any agency provides to Americans that they still are obligated to provide regardless of the number of people available to provide it.

Dr. Andew Hutson: And so they really wanna step up in a great way and find. How they can still meet the demand with the resources they do have available. And that's where we try to come in and help. We try to come in and help by [00:20:00] introducing easy tools for constructing workflows. Try to make it really approachable for folks that don't have, uh, a coding background.

Dr. Andew Hutson: Make things drag and drop. Very visual, very familiar. Uh, and, and ultimately, uh, inspectable, I'm not sure if that's a word, but someone can look at this and read it in the same way as someone else, uh, on the team and understand how the process works, uh, that that's a critical element in any type of process improvement, as well as continuity because.

Dr. Andew Hutson: If that process is readable by virtually anyone, uh, then as there is turnover, uh, as there is retirement, the knowledge of how the process is supposed to operate can be not only understood, but improved. [00:21:00] Or modified as needed as things change either with legislation or mission areas or focus or, uh, external factors that can impact the system.

Emily Nava: Totally incredible answer. So, if an agency is wanting to start to explore automation, or maybe they already have and they're having some trouble, um, what's one thing that they could do today? 

Dr. Andew Hutson: Well, the best thing they should do is reach out to their technology office because likely there are tools already available that they can use.

Dr. Andew Hutson: We're, we're in the same boat. Um. We are a service that's cross industry. We're available on Stratus that you can get our licenses that you can use. We have authority to operate with the DOD as, as well as other agencies. Um, we are here not only to [00:22:00] help with the technology, but also with the people. So we don't provide staff augmentation, but we do provide training consulting.

Dr. Andew Hutson: Uh, and support as part of our licensing structure because we know just, uh, technology adoption is so much easier when you have someone that knows the tech and is, uh, immersed in the domain knowledge in which you're working. So that would be my, my first, uh, recommendation is, is to reach out and see. What tools are there?

Dr. Andew Hutson: You can also reach out to us. You can even knock on my door, virtual door and come ask me about it, uh, because there's likely some tool that hasn't been leveraged. Uh, and that's available to you right now. 

Emily Nava: Yep. So you kind of touched on a big point about support and, um, adopting new technologies without [00:23:00] that support is very, very hard.

Emily Nava: Um, so what should, so if I'm a CIO or IT leader for, for an agency and I'm looking for a new partner or platform. I'm assuming, making sure I'm gonna get the support I need is definitely something I should look for. Is there anything else that I should be looking for when it comes to selecting a partner?

Emily Nava: Technology partner? Yeah, I would talk 

Dr. Andew Hutson: with partners. Uh, and something that's always important to me, uh, is more than just do they have support? 'cause that's everybody has support. It's how are you going to help my staff be successful with this software? How are you going to prepare them now and in the future to get value from putting their work on this platform?

Dr. Andew Hutson: Those are really important questions that I think are often missed or certainly delayed in their [00:24:00] response. It's just not in our DNA. To go along with that, we have a lot of veteran, uh, software professionals. Uh, on our staff that we've earned all the scar tissue of all the ways not to do something and how we can help best is all that learning that has been done really comes down to the people.

Dr. Andew Hutson: You probably have all heard this to some extent. There's people, process, technology, put these things together. But the people side I think is often misunderstood because it's not just do you have the right people for your process, that that's not really it. It's are you helping and supporting. People in the process with the technology.

Dr. Andew Hutson: Too often we go to the tech, then we try to find people who are experts in the tech and then give them the process. But that's just not [00:25:00] feasible in most scenarios, and that's wrought with failure. What's better is get the people that know the technology and pair them with the people that are performing the work.

Dr. Andew Hutson: Have them work collaboratively. Put the experts in a support role. Not only that, but a consultative role to answer questions as they come up throughout the adoption process, as well as the maintenance process. When you have those in place, the success of the adoption. Increases and the return on investment is actually realized.

Dr. Andew Hutson: When you don't do those things, you're gonna spend a lot of money, and at the end of a couple years, you're gonna look at the outcomes and it's gonna be nowhere near you thought they would be two years ago. 

Emily Nava: And then you're canceling that contract 

Dr. Andew Hutson: and then you're starting the process over again to, to waste more money.

Dr. Andew Hutson: That's just Exactly. [00:26:00] It's hard to see when you're in it. Uh, and easy to say, looking at it from the outside. 

Emily Nava: Exactly. Um, all right, well, we're about time at time here, but before we wrap up, I wanted to ask you, what excites you the most about the future of workflow automation in government right now? 

Dr. Andew Hutson: I think one, uh, that there's so many different agencies that are adopting ai, and that's a very broad term to projects that they have.

Dr. Andew Hutson: And what that means is they're opening the door to try to leverage these tools in the best way for the American people. And so I'm stoked about it because. Understanding all of this pre-work that's been done before, calling back to our probabilistic rule-based and case-based decision support systems.

Dr. Andew Hutson: Being able to have something like an LLM reliably partner with people. [00:27:00] Can't replace, but in my view, they can partner to help increase the, the speed and quality of outcomes, uh, is really exciting. And how those can be folded together in a secure and performant way, uh, for our federal agencies. Man, it's, it's like an awesome spot to be right now because it's just, it almost feels like magic.

Emily Nava: I agree. It's been, it's been really fun, um, having these conversations with people and, um, hearing how they're also excited about implementing these things in their mission areas. Um, so I wanted to take a pause here. And say, for those of you listening who are interested in hearing more conversations like this, between me and Dr.

Emily Nava: Hudson and some other guests, uh, Q Flow also hosts their own podcast. It's called The Interconnectedness of Things. Um, you can find it at q flow.com/podcast. Um. We talk about things in [00:28:00] healthcare, uh, federal government, definitely automation, ai, um, all the tech goodness. Um, so if you liked this conversation, I urge you to go over to q flow.com/podcast and give us a listen over there.

Emily Nava: Um, so Dr. Hudson, thank you so much, um, for talking with me today. Um, it's been a really great conversation. 

Dr. Andew Hutson: Always a blast. Emily. Can't wait for the next one.