Join Tan in this engaging episode as she chats with Fergal McGovern, CEO of VisibleThread, about the nuances of AI and technology adoption. They delve into the challenges of integrating new technologies, spotlight common pitfalls, and share actionable strategies for success. The discussion also highlights the vital role of 'Internal Champions' in driving a culture of continuous improvement and active engagement.
Join Tan in this engaging episode as she chats with Fergal McGovern, CEO of VisibleThread, about the nuances of AI and technology adoption. They delve into the challenges of integrating new technologies, spotlight common pitfalls, and share actionable strategies for success. The discussion also highlights the vital role of 'Internal Champions' in driving a culture of continuous improvement and active engagement.
00:00:00:03 - 00:00:08:15
Tan Wilson
You also talk about the dangers of both losing nations and choosing the right, quote unquote, job to be done by solutions.
00:00:08:18 - 00:00:38:10
Fergal McGovern
Getting hallucinations as really a fancy way of saying you may be a little bit surprised and you may get a little bit inaccurate results, but you have to check it because it's very convincing.
00:00:38:12 - 00:01:05:15
Tan Wilson
Welcome to another episode of the Optimize podcast. Today will delve deep into technology and adoption, exploring the common hurdles and uncovering actual strategies to navigate these challenges. The pace at which technology is evolving necessitates a robust strategy for adoption within organizations, and that's what we aim to unravel in today's discussion. We have a distinguished guest who has graciously sponsored this episode and brings a wealth of knowledge to the subject.
00:01:05:17 - 00:01:32:24
Tan Wilson
Today's guest is Fergal McGovern, the CEO of Visible Threat. And Fergal has worked extensively helping organizations become more effective with technology and I. His journey spanning over three decades reflects a blend of visionary leadership and practical insights that have significantly impacted the tech adoption landscape. The ethos of visible thread resonates through Fargo's endeavors, offering a clear lens to view and navigate the complex tech terrain.
00:01:32:26 - 00:01:42:28
Tan Wilson
We thank Fergal on his dedicated team of visible threads for their support and sponsorship of this episode. Fergal, we're thrilled to have you with us here today and look forward to hearing your insights.
00:01:43:00 - 00:01:44:16
Fergal McGovern
Great. Nice to be here.
00:01:44:19 - 00:02:02:18
Tan Wilson
So it's been a while since we've had a chance to kind of meet in person. It was a it's been a very great year for visible thread and obviously technology. So reminiscing about the inception of visible thread. What pivotal moment spurred you to embark on this venture?
00:02:02:21 - 00:02:41:10
Fergal McGovern
How long do we have, Tom? It's a great question. So if I reflect back to the early parts of my career, one of the things that I had the misfortune or perhaps fortune to be involved in, was working in the context of large organizations, large enterprises, you know, very large banks like Bank of Montreal. I spent a lot of time in the U.S. working primarily in the nineties, and what I would often see is wasted energy and time and miscommunication going on, particularly when important content like, for instance, business or technical requirements lived and documents.
00:02:41:12 - 00:03:10:01
Fergal McGovern
So there was a lot of friction, a lot of inefficiency and really what drove me to kind of start visible thread is to aim to provide a better way to assist discovery and finding key requirements, port requirements such that you avoid risk and you improve efficiency of review. So really it's all about document review, making that as efficient as possible and helping normal people who don't have time to get involved in tech.
00:03:10:04 - 00:03:18:12
Fergal McGovern
When I say normal people, I mean people involved in business. Use automation for that purpose, but in an easy way.
00:03:18:15 - 00:03:41:00
Tan Wilson
Yeah, that's quite a universal theme and challenge that I think a lot of businesses have. Regardless of the industry. So yeah, we see that on this side of the pond here too. And a lot of the things and especially the government sector too, there's, you know, government contracts tend to be very, very tight and time and cost constraints.
00:03:41:03 - 00:03:52:09
Tan Wilson
So I think some of the solutions that you've provided, I think really do help our industry, particularly to kind of help maneuver some of those challenges.
00:03:52:14 - 00:04:16:29
Fergal McGovern
And that's a great point because I think in the context of kind of government contracting, specifically the pressure, the time pressure to actually digest content, you've got a very short window of time to both digest a an important solicitation. You may even have, you know, six or eight weeks, go through hundreds of pages, make sure that you're compliant, make sure that you've got the appropriate risk profile.
00:04:17:02 - 00:04:43:26
Fergal McGovern
Make sure, in fact, that maybe it's inappropriate to even bid on a certain opportunity. So in that context, a lot of, you know, kind of long hours are part of the equation and there's a lot of burnout and so forth. So what we find with our customers is that when they apply automation in the right ways, in the right for the right tasks, that it can significantly alleviate the matter, improve the efficiency, is probably a better way to put it.
00:04:43:28 - 00:04:53:26
Fergal McGovern
And also improve the likely outcome, which of course for our customers is to win more business, is to be more compliant and therefore deliver on the promise that our shareholders know.
00:04:53:26 - 00:05:17:02
Tan Wilson
Absolutely. And I think this is even more pivotal too, with the onset of the large or. GEEWAX and the larger IDR queues, these task orders, I know for a lot of our customers are ten less days to respond and without any type of automation or templates and things like that, it just really wouldn't be, you know, possible in order to do that.
00:05:17:02 - 00:05:30:16
Tan Wilson
So but that's, that's quite the journey. And with like the API landscape rapidly evolving, how do you perceive some of these advancements influencing or shaping the technology adoption realm?
00:05:30:18 - 00:05:52:11
Fergal McGovern
So AI is yet another layer of kind of consideration when you think about automation and the environment. And I think there is a kind of a rush that's like pretty much every hype cycle. I'm not suggesting that generative AI is a hype cycle, but with every kind of new innovation, the Internet itself, the advent of the smartphone, A.I. is of a similar kind of stature.
00:05:52:15 - 00:06:18:03
Fergal McGovern
And the situation that's most important to keep in mind is that A.I. has been around an awful long time. It's the generative AI type of A.I. that is new, and that is something that's very good at certain use cases, certain jobs, and it's not so good at other jobs. So to make that concrete generative A.I. is not so good at mathematical jobs.
00:06:18:05 - 00:06:45:00
Fergal McGovern
So if I want to reliably, you know, do something in line, like read and write to a database. So user management, for instance, you would never use generative AI for user management or for workspace management or for parsing out a microsoft Word file, but you would absolutely use it for other jobs. So, for instance, summarization of text, the creative element of writing good proposals, that's where a generative I can really shine.
00:06:45:02 - 00:07:12:25
Fergal McGovern
So I think when you're thinking about A.I. and the advances, it's important to be a little bit more considered and not rush into situation and maybe avoid some of the or at least consider some of the products that are effectively just, you know, thin wrappers around a large language model, which is the engine of of generative A.I.. So there are a lot of new products out there who claim to deliver all kinds of nirvana and, you know, suburban type outcomes.
00:07:12:27 - 00:07:29:00
Fergal McGovern
Just be a little bit careful of those. And I don't say that as a vendor. I say that as a vendor who's been 15 years building product, invisible thread, and sometimes technology and engineering takes a long time. That's just the nature of it. You can't do things overnight and suddenly deliver a robust solution.
00:07:29:03 - 00:07:56:13
Tan Wilson
This an unfortunately, we're a society that is so into the immediate, like how can you sell my problem now and how can I go to one place to kind of solve that problem? And I'm super critical myself about really thoughtful adoption of technology to really go back and do the hallmark of evaluating what are your needs and what is the problem that you're trying to solve.
00:07:56:15 - 00:08:28:18
Tan Wilson
And obviously, if you do find a solution or tool that supports or can help, you know, at the enterprise level, at a more global level, then then that's great. But you do really have to understand first what you need and then understand exactly what it is that you are trying to adopt in order for those to to be successful or even to have those within your organization want to adopt it or understand why they are adopting or, you know, trying to implement a new technology.
00:08:28:18 - 00:08:36:13
Tan Wilson
It's it's that whole why are we doing this element that I think always kind of gets this sometimes.
00:08:36:16 - 00:09:04:22
Fergal McGovern
Yeah, that that's absolutely right And I think it's not that we're suggesting that you should be kind of, you know, overtly analytical about this stuff. But when you step back, there's a lot of budget right now available in large corporations. And effectively the budget is coming from the C-suite. And it's basically, you know, go get us onto A.I., generative A.I. and a lot of that money is going to be spent and waste, and some of it will be spent wisely.
00:09:04:24 - 00:09:28:28
Fergal McGovern
But the biggest thing that people should keep in mind is that ultimately, why do you adopt technology in the first instance? It is to make things more efficient, more productive, better outcomes. So it's a combination of efficiency improvements and better quality outcomes. So you can kind of map whatever it is you're looking to do to the core sets of problems, which is back to boring old business process analysis.
00:09:29:05 - 00:09:48:12
Fergal McGovern
It's kind of what is your asset state and what is your anticipated future state? And let's try to map the technology irrespective of whether it's A.I. or standard automation, it doesn't really matter. Let's try to map it back to those individual steps. And that way you can measure success much easier and avoid the idea of flushing cash down the drain.
00:09:48:14 - 00:10:08:09
Tan Wilson
Now, now, people, I think there is a mass, you know, movement to throw money at things. And, you know, just like everything else, you can't throw money at stuff and they get better. So, so true. But, you know, you and I are on LinkedIn a lot. I think it's been LinkedIn has been a blessing and a curse, let's just say.
00:10:08:09 - 00:10:31:23
Tan Wilson
Right. At the same time, everyone's talking about technology and AI and adopting, you know, their tool, that tool or, you know, the solution. And there's a lot of, you know, trust but verify type of, you know, challenges with LinkedIn and with, you know, with everything that we're reading out there. You don't quite know what your reading is. You know, it's true.
00:10:31:25 - 00:10:58:27
Tan Wilson
You've pointed out a lot, and I love following your posts because I feel like I get so much smarter, like every time I, I read those. But you've pointed out that that merely introducing new technology isn't going to be, you know, like we said, the panacea for many organizations. You know, that what they're hoping for. Can you expand on some of the, like common pitfalls that you encounter when companies try to adopt new technologies?
00:10:58:29 - 00:11:17:24
Fergal McGovern
Yeah, there's kind of certain patterns. It's like everything changes and nothing changes. Human beings are human beings and we all kind of, you know, we all have the same advantages and the same quirks and the same kind of set of things. By and large, irrespective of whether you're talking about 30 years ago or you're talking about this decade, it doesn't really matter.
00:11:17:26 - 00:11:44:25
Fergal McGovern
So the biggest challenge in adoption adopting excuse me, automation is a human challenge. It's fundamentally that people overestimate the ability for normal employees, typically in organizations to absorb behavior change, because the biggest issue is a behavior change issue. If you adopt new tools, if you introduce new stuff, new ways of doing things, you're going to get the boat curve of life.
00:11:44:25 - 00:12:07:24
Fergal McGovern
You're going to get people who are way ahead of this and really embracing it on one end. On the other end, you're going to get the naysayers, the people who reject change and are good with change. And that's okay. You know, they're people. And then somewhere in the middle, you're going to have people who, you know, have a normal life who don't you know, they work, but they work to live and they're not going to spend a lot of time engaging.
00:12:07:24 - 00:12:29:20
Fergal McGovern
So when management looks to adopt new technology, they really need to consider it from an adoption standpoint regarding the behavior change that's required. So how does that look in practice? You need to get by and you need to have people bought into the idea. You can't just drop new products or new technology in unexpected to suddenly magically work.
00:12:29:22 - 00:12:49:04
Fergal McGovern
So education, training, not training where you basically say, okay, we're going to send everybody in the organization off for three days and they're going to kind of sit in the same classroom format where it's old school kind of, you know, some lecture lecturing and it's got to be a much more refined approach to training, has got to be choosing individual groups.
00:12:49:06 - 00:13:12:07
Fergal McGovern
Typically those groups and organizations that you think will actually adopt this fastest and will lead the way and would set the template out for correct and good usage. So then it can start to trickle down. So I'm dating myself now, but it feels like, you know, trickle down economics from Reaganomics back in the day. Right. But there's a lot to be said for the idea of actually getting the best people.
00:13:12:09 - 00:13:32:17
Fergal McGovern
That doesn't mean the most senior people. It means the people with the best acumen, the people with the best vision about how this might work, getting those people involved and then trickling down those learnings which are now specific to the organizations way of doing business because a lot of time vendors will want around and you know, consultants will come in, they won't really know the business.
00:13:32:20 - 00:13:58:01
Fergal McGovern
Yeah, I mean they're slick, they've got PowerPoints, it's all good. It looks good. C-suite likes it. It feels good, but it fails and you can have a colossal waste of money. You need a proper kind of program, engagement plan, a change management plan. And ultimately, it's about people embracing this new technology. The second thing that's really important is that you've got to measure benefit.
00:13:58:04 - 00:14:25:10
Fergal McGovern
You know, if you're doing 180 day plan for a change management program or 180 day, you know, kind of I like to do things in tight increments, I would divide up into, you know, 30 day increments first, 30 days is going to be whatever under center, as is process. Second, 30 days is going to be getting our kind of key stakeholders involved, getting people to start to, you know, use the solution for real business objectives.
00:14:25:13 - 00:14:37:28
Fergal McGovern
And then you start to layer in more and more depth. And that's great because you get early warning signals at that point. So Big Bang approach is rarely work. It's much better to incrementally approach the entire adoption process.
00:14:38:01 - 00:15:03:09
Tan Wilson
Now, I think you said a couple things that really kind of resonate with me is the fact that I think with any type of new technology adoption, new process, whatever it might be, that you're always going to have people on different ends of that bell curve and that that's okay. I think a lot of times executives measure success as to how many people are on board or have adopted it.
00:15:03:11 - 00:15:25:27
Tan Wilson
And so I think it's sometimes, you know, until you rule it out and I like your you know, roll it out, you know, small and incremental and within an organization in our functional area where I think it will have the most impact or it's going to be more heavily use where you can see benefits earlier, I think that will help show what some of the key performance indicators of what success looks like.
00:15:26:00 - 00:15:55:16
Tan Wilson
And with that said, though, success within one functional area, within one organization is going to look very differently than maybe somewhere else. And sometimes executives like to broad brush success in one area and think that they're going to be able to replicate that somewhere else at that same level of the intensity or the same utility. Maybe they're looking for a percentage or hours or whatever of, you know, use.
00:15:55:18 - 00:16:24:17
Tan Wilson
It just doesn't translate like that. And that's that's okay. And that's still could be viewed as success. But I think you need to know how and what you're measuring first before you can kind of go in and, you know, set that set that measurement or try to claim success for sure. Yeah. So, you know, I know when you when you talk about adoption and everything, you you also need to stress the need for heightened security with adoption.
00:16:24:17 - 00:16:29:15
Tan Wilson
Can you talk to little about security and elaborate on that?
00:16:29:18 - 00:17:08:11
Fergal McGovern
Yeah. So security and A.I. is kind of a a really important factor, particularly when you're dealing with regulated industries. And again, I don't know if the awareness is out there to the extent it needs to be when you're looking at vendors of software. And I guess I can speak from that point of view because that's the area that we women, there are an awful lot of vendors out there, particularly ones that pop up in the last, you know, short period of time, let's say, between one and three years who are promising all kinds of stuff and who are kind of not necessarily keeping in mind what it takes to deploy software in an enterprise setting.
00:17:08:13 - 00:17:30:13
Fergal McGovern
When you deal with information security folk in an enterprise setting, they have legitimate and valid concerns about the content and the content. Typically, they don't want that content going outside of their firewall. So that's a tech way of saying it just has to be 100% locked down. If you think about defense space, you think about IT services, financial services, health care.
00:17:30:16 - 00:18:16:14
Fergal McGovern
You just don't want that going into the public Internet. So a lot of companies will set up private clouds for that exact reason in defense of space or in kind of more, you know, situations where you've got secure considerations and you've got, you know, skiffs going on or whatnot. These private clouds are really you're kind of your entry point if you're using Lem's large language models, which are fundamentally the engine of generative AI, if you're using GPT, for instance, which, you know, Openai was the first to democratize the whole generative Hey, I one year ago actually with GPT three and creativity, once you put that content into a GPT type arch language model, it's basically
00:18:16:14 - 00:18:35:27
Fergal McGovern
violating the security stance and posture because you don't know at this. There's a number of factors here. You could be training inadvertently, but that's mostly locked down. Now. I think most of the asking of model guys have kind of got options for that, but effectively your content is going outside your environment and there's a compliance implication around this.
00:18:36:00 - 00:18:54:06
Fergal McGovern
So you don't know if non-nationals are going to be able to somehow interact with that software or that large language model. And the reason, you know, a lot of the kind of intent of the executive order from Biden, which now is about two or three weeks old, was we can debate the ins and outs of why it's good and why it's not so good.
00:18:54:09 - 00:19:18:23
Fergal McGovern
But the intent was the security intent. So what do corporates need to do? I think they need to have a checklist. They need to ask the vendor, where is your Al-Alam living? They need to understand what are infosec people be able to deploy. That allowed them my view, and this is our stance as visible threat because we are very conscious of our security posture.
00:19:18:23 - 00:19:40:17
Fergal McGovern
It's one of our big differentiation points for the last 15 years, working with the top defense contractors, for instance, we've always, you know, nailed our colors to the Mars security as the number one priority for us. That's why we are deployed into skiffs. So for us, you need to deploy and Al-Alam, which is completely isolated, it cannot be trained.
00:19:40:19 - 00:20:02:22
Fergal McGovern
It should not be leaking any data outside of the firewall. And the chief information security officer needs to be 100% comfortable that that is the case. Yes, there will be evolution very quickly in the next six months, 12 months around security. But I think in this industry, in this business, you just can't take chances on your content, your proprietary content.
00:20:02:22 - 00:20:05:12
Fergal McGovern
You just can't do that now.
00:20:05:15 - 00:20:44:02
Tan Wilson
And I think fortunately, unfortunately, the cybersecurity industry is going to be very successful and very profitable for decades just because every time we do introduce a new technology, every time we do introduce something new within our organization that increases our vulnerability is tenfold. And then it makes security even more complex than what it was before. Right? Again, not to date myself, but I remember when we used to fax data, you know, to to other data.
00:20:44:04 - 00:20:44:22
Fergal McGovern
I was there.
00:20:44:22 - 00:21:07:17
Tan Wilson
Right? Yeah. And then it was even more secure because you would phone and call someone and then emails and all of that. All the World Wide Web kind of made, you know, working in this environment even more difficult. So again, we've some of us have lived through several life cycles of this, and so too we shall overcome this.
00:21:07:17 - 00:21:28:06
Tan Wilson
We just need to kind of stay smarter than those who are, you know, trying to to kind of find those breaches and stuff like that. So, you know, you like so let's take a of but from security because I know we could talk about security and scare, you know, all of our listeners for 4 hours on some of those challenges.
00:21:28:06 - 00:22:00:16
Tan Wilson
But you know, you also talk about the dangers both loose the nations in choosing the right quote unquote, job to be done by A.I. solutions. And I think that's that's such a great way of of coining. That is the job to be done. Right. And that's that's really kind of an interesting take on it. Like talk to me a little bit more about that, because I think a lot of people think that, you know, AI and certain technologies is that one hammer that does everything or it's just one tool in the toolbox, but it does have a pros and cons for sure.
00:22:00:17 - 00:22:28:29
Fergal McGovern
So I think this is maybe, you know, it's going to be a learning opportunity for people to understand what our language models are, what AI generative A.I. is, what jobs it's good at doing, versus what jobs it's not so good at doing so. So if we kind of strip it all back and we say, well, what is generative A.I. and I don't want to get into a Ph.D. about kind of AI in general, but fundamentally generative A.I. at its core is a predictive, creative technology.
00:22:29:02 - 00:22:48:17
Fergal McGovern
You know, if you think about the good use cases that the places where it plays, it's creating new content based off a massive training model where it can actually, you know, be quite impressive. People. In the early days of Toshiba, people were, you know, doom and gloom naysayers. You know, the earth is going to end because this thing can think it can't think.
00:22:48:17 - 00:23:15:05
Fergal McGovern
What it's doing is it's got a bunch a huge corpus of, you know, stuff in that model. The entire Internet is fueling the model and it's predicting what is the likelihood of actually sequencing a number of words. Back to you in response legitimately to your question. So it appears very compelling. It is fear appears very convincing. But what you get back, you have to verify.
00:23:15:08 - 00:23:38:06
Fergal McGovern
So it's trust, but verify at time. And we, of course, as professionals should always do this. So the good use cases, you know, summarize this piece of content, create a first draft of a of a blog post or create a first draft of an RFP. That's all good. But you would be remiss as a professional to believe that and to actually solely depend on that.
00:23:38:06 - 00:24:04:01
Fergal McGovern
It's kind of, you know, again, believing in silver bullets. It's it's Coca-Cola and stuff. You use it to help you speed up your process. You use it for the right use cases. You do not use it. You know, I saw this actually friends of mine in the early times of our lands, which is only like eight months ago, I had one particular friend who was coding against this to give him back in technical land what's called Jason.
00:24:04:05 - 00:24:25:15
Fergal McGovern
Jason is an underlying way to represent data and he said, this is great. I'm going to speed up that entire bit of the problem I had to do using traditional approaches. I don't need to do that anymore. My that will just do it and it'll be so easy. And Barbara, the problem with that is that the LEM is what's referred to in computer science circles as non deterministic.
00:24:25:17 - 00:24:52:03
Fergal McGovern
If you run the same command a million times, you will not get the same outcomes because it's generating the contents. And if the model changes or something changes in its tuning, which you can, then you get a different outcome. So getting hallucinations is really a fancy way of saying you may be a little bit surprise side and you may get a little bit inaccurate results, but you have to check it because it's very convincing.
00:24:52:06 - 00:25:16:20
Fergal McGovern
My favorite example is I asked three other labs how many words are in this sentence? Very simple question. It seems simple, right? And the sentence was, in fact, 45 words. Bing gave me a 35 word answer, saying there was 35 words in the sentence barred, Give me a better answer. Which was, I can't give you a good answer because I'm an alum and I'm not very good at giving answers for math type questions.
00:25:16:23 - 00:25:50:05
Fergal McGovern
Well, that's and then to Beatty, which was appropriate. Exactly. But it wasn't so good at doing that thing. And Chatbot gave me false seven instead of 45. So it was a little bit off, but not too much off. So now if I'm a person going into a, you know, going into the DMV for for whatever it might be, and I'm relying on a system to give me accurate answers for questions, I better hope their standard software player that gives us the right answer is 100% of the time, every time.
00:25:50:07 - 00:26:16:03
Fergal McGovern
And alarms are not guaranteed to do that. But they're very good. A creative type Use cases, tech summarization, photo filling in stuff, diagnostics, predictive stuff. That is just enough for the job at hand. So for me it's all about this job to be done. Again, it comes back to the earlier part of the conversation. What is it you're trying to automate and maybe it's appropriate to use to herself, or maybe it's appropriate to use all of them type software.
00:26:16:06 - 00:26:37:20
Fergal McGovern
Or maybe in reality it's appropriate to use a hybrid. You inner Paula, we use a lot of AI, but not generative A.I.. And as we look in our kind of roadmap, we will be layering in a lot of generative A.I., but would be very considered about the job it will do for us. And there are times in places where it's really good at doing certain things.
00:26:37:23 - 00:26:48:20
Fergal McGovern
So I think it's not a binary question. It's a question of understanding what this stuff is good at and what it's not good at, and making your judgment and then mapping it back to the process that you're trying to automate.
00:26:48:23 - 00:27:14:22
Tan Wilson
Yeah, exactly. I love collecting examples of hallucinations, and I've heard you talk about that before. You know the word count. And I'm going to put it into like my list of examples to kind of provide other people. But I've got a hallucination story myself, too. I was it's a it's very good for creating resumes and being able to summarize them.
00:27:14:26 - 00:27:35:27
Tan Wilson
And it's very compelling. It's probably one of the most compelling writers, you know, I've ever encountered. It's like it's actually convinced me sometimes, yeah, that's a great idea and it's great to kind of throw. I kind of use Chad as my sounding board, right? Like, I've got this lame, crazy idea of doing this. What do you think?
00:27:35:27 - 00:27:57:10
Tan Wilson
You know, what do I need to, like, consider and kind of like work through some of that planning. But I was using it to do a resume one time and I was just trying to, you know, throw in all these, you know, background things and it actually hallucinated and said that the candidate was a former CIO at the Marines or something.
00:27:57:12 - 00:28:22:29
Tan Wilson
And I was like, and that he was fluent in Japanese. And I thought to myself, I never said that he spoke Japanese. I never said that that was a position that he had once held. I'm like, Where do you like? Where did it possibly create that? Yeah, And it turns out it was because there were a couple of bullets from the candidate where he was at Okinawa.
00:28:23:02 - 00:28:46:00
Tan Wilson
He did consult with the Marines and he has consulted at the skills level. So of course, like he had hallucinating and says, yes, you're the former deputy CEO of the Marines and you are fluent in Japanese. So nice. That was. Yes, that was very, very far off. I was like, wow, that's that's an example. I'm going to keep an in my bucket here because people need to know this.
00:28:46:00 - 00:29:10:12
Tan Wilson
People need to understand and, you know, like it's very I think we're hearing more and more about the lost nations. And now that people have more time to kind of experience and kind of share their success stories of their, you know, challenges with that, where I think we'll we'll hear more and more of that and hopefully more people will verify what's coming out there.
00:29:10:15 - 00:29:36:05
Tan Wilson
But, you know, you mentioned the notion of having a champion to guide you through this adoption process. And I think that's so important. It brings a sense of like ownership and accountability, especially given like the evolution pace. It's such a rapid pace right now. How do you think organizations could effectively identify and empower, like these internal, quote unquote, champions to navigate through this adoption?
00:29:36:05 - 00:29:39:21
Tan Wilson
You know, like voyage.
00:29:39:23 - 00:30:12:08
Fergal McGovern
Yeah, it's it's a great question and it's just a thought process from my standpoint. I think the idea of having a champion internally in any enterprise organization is critical to any significant kind of behavior change process with automation. What are the characteristics of that? Champion That's a great question. The person needs to be okay. Well, firstly, it can't be a part time role and this is a huge issue, right?
00:30:12:08 - 00:30:33:18
Fergal McGovern
A lot of times it'll be, well, I got my day job and whatever, and I'm supposed to be doing this secondary job, which isn't really being empowered to actually deliver anything significant. So the person has got to be more or less full time on this. The second element of that person is that they need to have a combination of two skills.
00:30:33:18 - 00:31:11:03
Fergal McGovern
One is effective communication and the second is effective or effective enough understanding of tech. They don't need to be a deep dive tech person. They don't need to be a dev person, but they need to have a logical disposition. That's a tricky mix. An effective communicator with good, logical talents and skills they are around. You will find them, I think, biasing towards a younger person not to be a just and I'm not exactly a spring chicken myself, but I would certainly bias towards a less aged person.
00:31:11:05 - 00:31:37:25
Fergal McGovern
Not because they don't have the capacity, they certainly could. They're probably towards the back end of their career. They probably don't have enough energy to kind of push it through because it requires a lot of energy. It's effectively an internal consultant role and you're liaising with a lot of different teams. You're trying to do a combination of business modeling, you're trying to do a combination of kind of education and get people on board, understand the tech landscape.
00:31:37:27 - 00:32:00:16
Fergal McGovern
So there's quite a bit going on in that role. It is not a part time role, it's a full time role. It also can often be trodden into the kind of PMO program management office that in my experience has been the kiss of death because a lot of that office or those offices can often be on a part and they don't have any kind of real clout in an organization.
00:32:00:18 - 00:32:18:20
Fergal McGovern
So the reporting structure of the person and there should be multiple of these people in on the organization. The reporting structure is important, so they need to report more or less when, say, directly into the C-suite, certainly 1 to 2 steps marks away from the C-suite, because that gives them power and they will need to bring people along the journey.
00:32:18:20 - 00:32:28:26
Fergal McGovern
So they need to have the ability to crack a whip every so often, so that there are the characteristics that gives you success. You don't do that. You just waste your time.
00:32:28:28 - 00:32:32:18
Tan Wilson
It almost sounds like a proposal manager, right? You know, an.
00:32:32:18 - 00:32:58:05
Fergal McGovern
Empowered Proposal manager, An Empowered proposal manager. And the challenge with proposal managers is that they're considered, in many cases a kind of a back office function. They're not considered as strategic as really they should be. So, yeah, I mean, that challenge that you guys were asking, you've got an unimpaired proposal. Manager That's the doom situation. But when you've got an empowered program manager or a proposal manager, excuse me, life is good.
00:32:58:07 - 00:33:27:22
Tan Wilson
Yeah, exactly. And you know, again, I've been I've been accused of being an agent myself, but I do love to work with, you know, certain certain professionals in a certain age group for certain things. Right. And I think having someone who is on the beginning of their career, they tend to be really they're they're like a sponge right there.
00:33:27:22 - 00:33:52:26
Tan Wilson
They're open to hearing and experiencing and trying and failing and trying and failing. And they can see if you can find the right person who is very futuristic, like you said, finding someone who is a good communicator and technically speaking, is really good and understands the technology and how the solution of technology is going to fall, solve a potential problem.
00:33:52:28 - 00:34:20:07
Tan Wilson
And having that future mindset, I think is is such a powerful combination. A lot of times when you have someone who is within an organization and they've been there and you know, they understand that that person has their their value and their role. And this whole process to they they know legacy wise, you know, some of the challenges that the organization's gone through, what works, what they need to do to kind of get, you know, to the end result, which is success.
00:34:20:10 - 00:34:38:24
Tan Wilson
But a lot of times there's a lot of baggage, there's a lot of things they have to unlearn. And so there's a lot of negative type of conversations of, well, we tried that before. It didn't work well, you tried it under a different technology or you tried it under a very different set of, you know, the factors are a little different.
00:34:38:24 - 00:35:03:07
Tan Wilson
Some of the, you know, things are it's it's not quite the same scenario as maybe before. So, yeah, definitely having someone I think it's kind of more difficult, though, when you're a smaller organization and you have some you you've have resource limitations as you do wear many hats. And I know that a lot of our you know, our listeners are from small business.
00:35:03:09 - 00:35:14:04
Tan Wilson
If they are our business, you know, how can they do that if someone can't do this full time? Is it even possible for them to to do this successfully?
00:35:14:06 - 00:35:40:16
Fergal McGovern
Well, small businesses are a little bit different and their complexion most significant decisions in small businesses are either directly or one step away from the CEO or the president, anything significant. So the technology adoption is a significant question so that president or CEO needs to be involved in that evaluation. They need to be involved in the successful rollout.
00:35:40:16 - 00:36:01:05
Fergal McGovern
They can't just say, well, okay, knock yourself up by a tool coat. Off you go. It it's it's more strategic than that because what's strategic for a small business isn't necessarily the same for a large business. So it just depends. We deal a lot with enterprises, but equally we deal with a lot of the same BS about a third of our customers or SMB.
00:36:01:07 - 00:36:42:00
Fergal McGovern
A third are over $1,000,000,000 in revenue. So it's it's a wide kind of church. What I've noticed are SMB customers do phenomenally well when there's vested interest in the solution and they understand what they're trying to get out of it. So it goes right back to the earlier part of this conversation, which is if you know what you're trying to automate, if you know that there's inefficiency or that there's potential massive efficiency, so you'd no longer have to hire the same people you thought you would have to hire if you know that upfront and you measure against that, then for sure that C-suite person, that that person in charge will be all over that because
00:36:42:00 - 00:37:03:24
Fergal McGovern
it's completely in their interest. So it's a slightly different situation In my context, what I was talking about earlier was in a very large organization and you do need dedicated resource in a smaller organization because you run faster because you're juggling money, more balls at the same time. As long as the C-suite is involved and leadership is involved, then you can get good outcomes.
00:37:03:27 - 00:37:43:00
Tan Wilson
Exactly. So I think the other part of not only having a champ in is important, but the communication. That's that's where everything breaks down, right? As you you don't have good communications, clear lines of communication, clear lines within the organization and and mechanisms for being able to provide communications and how do how do you think that the internal champions and all those kind of involved can effectively foster feedback and how and what does that feedback look like for success?
00:37:43:02 - 00:38:06:18
Fergal McGovern
So that's a great point about the kind of feedback loop on the two key points, and that kind of ties in with our previous discussions. When you do any kind of change management being a small to medium business or being an enterprise, it is really important to have an understanding of how you're going on the actual rollout, how you're going on the actual change of behavior.
00:38:06:20 - 00:38:33:09
Fergal McGovern
And the only real way you can do that effectively and yield a successful result ultimately is tight cycles, tight feedback loops, situations where instead of considering, okay, we're going to roll this out and six months in six months, we're going to see if it's working or not. Instead of that, you look at kind of two four week increments and you just tighten it down and you allow yourself the ability to course correct.
00:38:33:11 - 00:38:52:09
Fergal McGovern
So it's kind of a fundamental tenet of, you know, an agile approach. I mean, in the world of I.T, we talk about agile approaches, meaning tight delivery cycles. The rationale for doing that is that you tie box on a very tight timeframe and you check the results. You need to understand what results you want and you see whether you're in the right direction or not.
00:38:52:12 - 00:39:13:29
Fergal McGovern
And then you shift course as you need because life work in a static way. So it's not like we have this master plan. It's going to take, you know, four months to conceive and then we're going to roll it out. It's going to be on here for six months and then we're going to see the results. It's much more, okay, the first month is what's our initial place that we want to see change.
00:39:14:02 - 00:39:38:27
Fergal McGovern
It's a tiny sliver of the overall objectives. Let's measure that. Let's course correct us. We need a set of people involved. And that's an entirely you know, it's completely dependent on effective communication for the both the coordinating party and in a large organization that's a dedicated person in a smaller organization, that's going to be a part of the C-suite in some way.
00:39:39:00 - 00:39:56:04
Fergal McGovern
It'll be a part of their silos, their weekly status calls. Okay, we bought this software. We expect to have to do X. Is it doing X? Who's supposed to be using it? Are they using it? Have they been trained on it? Are they getting the results we thought we would get when we actually signed off on the purchase order on this and the investment?
00:39:56:06 - 00:40:26:14
Fergal McGovern
Because if you don't do that and we have examples where a customer is and it's befuddling, it's it's crazy. They buy our software and they pay no attention to the post-purchase point. It's almost as if, like they believe it's just going to magically work on its own. It isn't. And if you don't have your people using it effectively and we spend a lot of time and energy, you know, having our staff or customer success staff working closely with our customers once once the software is bought.
00:40:26:16 - 00:40:48:04
Fergal McGovern
But if if you can't get that engagement on the other side, it's a waste of time for all concerned. It's like the old phrase goes, You can bring the horse to order, but you cannot make the horse drink. And in this case we see it. And it's it's a very strange thing. You wonder why would somebody drop X thousand dollars on a piece of software when they're not going to actually give it a fair crack?
00:40:48:04 - 00:41:03:21
Fergal McGovern
It actually been successful. And that is right back to that whole idea of how do you do successful change in behavior. And that is predicated on a huge to huge extent on effective communication, both internal and external.
00:41:03:23 - 00:41:36:07
Tan Wilson
No, exactly. I've seen it. I've experienced that adoption challenge myself. But, you know, I love technology. So I'm willing to kind of try it and test it. But again, I wish I would have a champion sometimes to kind of help me maneuver that. But it sounds like this champion actually like I'm a huge superhero fan. So I don't know if you know this, but my son Henry, he and I are Marvel, DC, Marvel Ray superheroes.
00:41:36:07 - 00:41:53:17
Tan Wilson
So we were like, I love all the superhero movies and whatnot, but if you had one superpower to ensure that organizations seamlessly could integrate and engage across organizations, what would it be? What's that one? Superhero power.
00:41:53:20 - 00:42:09:27
Fergal McGovern
Yeah, I think it's kind of the theme of this entire conversation. It's the ability for me to help other people communicate effectively. Communication is a single superpower. I would love to be able to imbue upon teams of people.
00:42:09:29 - 00:42:41:06
Tan Wilson
Well, there you go. Couldn't have said that better myself. So before we wind down and and close out this podcast, I want to kind of spice it up a little bit. Right. This has been such a heavy conversation about something that is so exciting and new. So I'm going to put you on the hot seat a little bit and I'm going to ask you to respond with just a sentence or a phrase to some of these questions, and we'll just kind of have a little bit of fun here.
00:42:41:13 - 00:42:54:25
Tan Wilson
So today here, what was the first thing? Well, what was the first thing that crossed your mind when you heard the term artificial intelligence?
00:42:54:27 - 00:43:06:05
Fergal McGovern
So I first heard the term artificial intelligence in 1989. Believe it or not, I was a computer science major and I thought really interesting.
00:43:06:08 - 00:43:21:14
Tan Wilson
How you get extra points for giving it a year or two. So. there you go. I know our listeners know exactly how I'm going to answer this next question, but are you a morning person or a night owl?
00:43:21:17 - 00:43:22:23
Fergal McGovern
Morning person?
00:43:22:25 - 00:43:40:01
Tan Wilson
Yeah. Never got to do that. So now I know I ask Alexa every day, what time is it in Ireland so that I know when to send my first messages before I go to bed? So I'm your night owl. So that's the reason why you see emails from me, right when you wake up in the morning.
00:43:40:04 - 00:43:41:04
Fergal McGovern
Got it?
00:43:41:06 - 00:43:51:03
Tan Wilson
Yeah. So? So what was the last book you read that reshaped your thinking?
00:43:51:06 - 00:44:32:24
Fergal McGovern
I have to confess, I have. It's been years since I've read a book. I am absolutely a committed podcast listener. I consume all my information over podcasts, so I'm like, weird when it comes to books. So no books podcast. Why is tons in particular, you know, mandatory listening for people in my game, the All in podcast, which is basically a Silicon Valley type focus thing, I love the thing called the Restless Politics, which or your side of the pond won't be as familiar, but it's basically the former assistant to Tony Blair, a guy called Alastair Campbell, and Rory Stewart, who's a former conservative, which is the opposition party.
00:44:32:26 - 00:44:39:19
Fergal McGovern
They talk about a lot of topics and it includes technology. It includes politics as as obvious. But that's where I get most of my earnings.
00:44:39:21 - 00:44:45:02
Tan Wilson
Okay. Any e-books are with that now?
00:44:45:04 - 00:44:52:11
Fergal McGovern
Very occasionally. But really, honestly, it's all audio. It's it's that's where I learn everything. Yeah.
00:44:52:14 - 00:45:03:00
Tan Wilson
Yeah. That's interesting because my husband is a ferocious reader. He even still reads like newspapers on like his phone and I like newspapers. Do they still exist or.
00:45:03:00 - 00:45:03:23
Fergal McGovern
Do I.
00:45:03:26 - 00:45:12:25
Tan Wilson
Feel like, wow. And for the first five years of our marriage, you thought I was illiterate because I never picked up a book. It's like now I do read.
00:45:12:27 - 00:45:14:15
Fergal McGovern
I think I'm in that category.
00:45:14:15 - 00:45:29:23
Tan Wilson
Yes. Yes. You and I are both illiterate in his in his eyes, I suppose. But okay. So if you could give air voice, what famous person's voice would you use?
00:45:29:25 - 00:45:32:12
Fergal McGovern
It would have to be Arnold Schwarzenegger.
00:45:32:14 - 00:45:36:14
Tan Wilson
Leigh, I would love to hear this. Why?
00:45:36:17 - 00:45:57:07
Fergal McGovern
Because I was actually, funnily enough, listening to a podcast, Restless Politics, they interviewing Arnold Schwarzenegger as really taken what his story and his his fundamental optimism is, is really infectious. So I just kind of liked that. It's more it's just the voice tone. It's more of the actual attitude. Yeah.
00:45:57:09 - 00:46:21:03
Tan Wilson
Though I would agree with you on that one. I like his voice. I don't know what it maybe it's because we're on this side of the pond and we find all the accents on the other side so much more interesting than ours. Except for maybe down here in Texas. Y'all. But yes, I just hear, you know, I could speak to someone and there's a little threat all day long just because of the accent over there.
00:46:21:03 - 00:46:45:07
Tan Wilson
So for me, it's it's this attack. But Arnold Schwarzenegger is really kind of interesting. There's there is a Netflix documentary about his life and he's very introspective. So it's very interesting how he's kind of looked back and his view on on things and how he, you know, whether or not you believe that his politics are not super interesting journey that he's had.
00:46:45:09 - 00:46:53:23
Tan Wilson
So I'm a huge techie and I love new cool, shiny things. What's the coolest tech gadget that you own?
00:46:53:25 - 00:47:05:27
Fergal McGovern
I'm going to disappoint you massively. I'm not a gadget freak at all. Pen and paper is my belief system and building software. But I don't I don't go for gadgets.
00:47:06:00 - 00:47:22:23
Tan Wilson
Well, no, that is boy, that's that's very enlightening. I would have thought something totally different. Well, if you ever need advice on a tech gadget, I have them all. The coolest ones I will have. And I will test and break so you can come to me for regular news.
00:47:22:27 - 00:47:24:27
Fergal McGovern
I'd like to surprise there.
00:47:24:28 - 00:47:56:10
Tan Wilson
There you go. So this conversation day has been so much fun and so enlightening. And here are some kind of key takeaways from our discussions. For our listeners, engagement is really critical to seamless technology adoption. Be wary of those hallucinations and generative AI and then ask yourself what is the right, quote unquote job to be done? And the right internal champions are really critical when automating technology and within organizations, whether it's A.I., not A.I. solutions, are both.
00:47:56:12 - 00:48:25:05
Tan Wilson
So as we wrap up, another enlightening episode of the Optimize podcast, I want to thank Fergal McGovern for sharing his invaluable insights and exploring technology and AI adoption. Your expertise has undoubtedly shed light on actual strategies that our listeners can implement. To our listeners. The challenge lies in embracing the strategies, discuss and identifying your organization's internal champions to drive tech adoption and to adopt AI proactively.
00:48:25:07 - 00:48:45:04
Tan Wilson
It's about engaging and creating a culture, feedback and continuous improvement to fully harness the potential for technology and AI. So that's a wrap, folks. Keep sharing, sipping, strategizing and remember that the secret ingredients is a genuine connections and a glass of bourbon infused with a gentle splash of wit.