Calculus of IT - Season 2 Episode 6 - Creating Technical Diversity without Chaos

In Episode 6 of The Calculus of IT, we tackle a critical challenge: How do you create technical diversity without unleashing chaos? Joined by special guest Kevin Dushney, Mike and I explore the delicate balance between maintaining technological flexibility and preventing system sprawl. We dive into the "pendulum swing of standardization," discuss how to build sustainable technological diversity, and examine ways to measure success through diversity health indicators.

From establishing clear system boundaries to managing vendor relationships, we share practical strategies for maintaining technological independence while avoiding the pitfalls of both rigid standardization and unchecked chaos. Plus, we explore why sometimes a little chaos might not be such a bad thing - as long as you know how to manage it.

Transcript

Nate McBride: [00:00:00] You are, you are bundled up. Holy cow. I'm recording. Dude, it's fucking freezing out, man. I had a heater in that thing. I do. It's on and it's still, it's going to be, I'm okay. I'm all right. All right. If you see it, if you see the clouds of air coming out of my mouth, it's all, it's all good. I'm fine.

So tonight is going on Kevin. How are you Mike? 

Michael Crispin: I'm doing well. Doing well. Yeah. 

Nate McBride: Look it's the breakfast of champions 

Kevin Dushney: long time. Oh boy. That's that's on the that's on the no fly less for me 

Nate McBride: Well, if you do shots of it, yeah, it's gonna ruin anyone's night But if you just sip it, it's actually quite quite quite pleasant.

I don't believe it It's true story and it's just the effort the smell it's kind of like at that menthol You know just clears out the the pipes. 

Kevin Dushney: That's the problem as soon as you smell it. It's like nope That and Southern Comfort. [00:01:00] 

Nate McBride: Well, Southern Comfort just should never have been invented. I mean, that's some terrible shit.

I agree. It's an abomination. 

Kevin Dushney: Oh, 

Nate McBride: good.

Kevin Dushney: This podcast is brought to you by Southern Comfort. 

Nate McBride: Oh, yum yum. This podcast is brought to you by Stomach churn all night. Chunky churn. AI churn.

We're churning AI. 

Kevin Dushney: I'm in AI hell right now. 

Nate McBride: Well, is there an AI heaven? Like, is there, is there an alternative? I don't believe in it. There's maybe purgatory, but. I was at a demo this afternoon in Lexington for a company whose name I shall not say, but they, they profess to be a leader in, in deploying their, their like little robot among all of your unstructured data.

Mm hmm. And then giving you a map of confidentiality and risk and all of that, and I [00:02:00] asked him, like, how does it work? He said, well, we have an AI agent that does it, you know, that understands all the context. And I said, well, how does that work? How does it work? Oh, we don't know. Just, it's an AI agent. I mean, it's a sparking AI.

Michael Crispin: It just doesn't. It just doesn't. 

Kevin Dushney: It just doesn't. Can I just bamboozle you with a buzzword and you'll go away? Alright. Yeah. 

Nate McBride: I literally, I bet if we started a company tomorrow, and we called it, you know, uh, Psydually. If we invented Psyhoolie tomorrow, 

Kevin Dushney: Yeah. 

Nate McBride: We said it had, it was an AI, It was an AI platform.

AI platform to catalyze your corporate data. We'd be fucking billionaires by the end of the week. 

Kevin Dushney: Yeah. Can it fill in all the missing metadata for me too? 

Nate McBride: Yeah, it does. It does metadata. The agent does it. 

Kevin Dushney: Yeah. I just have an 

Michael Crispin: agent installed. It doesn't do that 

Kevin Dushney: when the, you know, the chemist that [00:03:00] should have put the metadata in is no longer here.

It, it 

Nate McBride: replaces the chemist with an AI agent.

Count me out, I'm 

Michael Crispin: in, count me in. 

Nate McBride: This podcast tonight is sponsored by SciHooli and brought, and also brought to you by Southern Comfort and AI Agents Incorporated, LLC. Okay. Uh, 

Kevin Dushney: a logistical question up front. Are you currently, uh, devoid of Islay Scotches? 

Nate McBride: Devoid? Well, I am. I, I went and bought Auckintoshan tonight, but I also have, I have Woodford.

Okay. Over, over there I have a gallon of wild turkey. Actually, no, sorry. It's right there. And then all, everything else is up in Maine. So I was looking for tonight buying the Auckintoshan and I'm like, you know what would be really good tonight? Not [00:04:00] Auckintoshan, but Jägermeister. And that's why I'm having it.

All right. But yes, otherwise I'm, I'm, I have nothing. I have a bottle of core vodka that I bought from Mike like three years ago. He had, he never finished. 

Michael Crispin: I still have to have that. Definitely still need that. 

Kevin Dushney: Yeah. I have more for you, so I'll, I'll bring, uh, Oh, thank you. Next time I'm on site, I will, uh, deliver the goods.

Nate McBride: I will happily take it. No doubt. And as we all do on this show, we just take it, Kevin. I don't know if you've, if you've followed the season, but 

Kevin Dushney: I am following. 

Nate McBride: Yeah, we just, we just take it. We just take it. 

Michael Crispin: Just take it. Just take it. 

Nate McBride: Take wherever 

Kevin Dushney: you want. Go wherever you want. Cafe Press. 

Nate McBride: Yeah, just take it.

I'm designing the St. Paddy's Day shirts for Calculus of IT right now, the limited edition ones, and I'm going to be doing one that says just take it. 

Kevin Dushney: Yeah. 

Nate McBride: St. Paddy, you know, big, [00:05:00] big shamrock. Yeah, and capitalize 

Kevin Dushney: the I and T and take it. 

Nate McBride: Oh, hey, look at you. Portion of the proceeds to the Dushni Foundation.

Okay, uh, well, Everybody, welcome back to the Calculus of IT, Season 2, Episode 6. This is the paradox of all paradoxes. The place where the salad meets the sad salad. The nexus that surrounds the nexus and the intranet for the extranet. It's just a paradox. You should get over it with us tonight. We're very, very psyched to have back our very, very special guest, Kevin Dushney, who, for the third time in his career, is willing to, um, take an, take a chance of destroying his reputation by being on this.

This podcast, I mean, have the 

Kevin Dushney: distinct pleasure of joining my esteemed colleagues on this podcast. 

Nate McBride: Ah, we're always happy to have you. Uh, this is next to Kevin is Mike Crispin and I'm Nate McBride. And at [00:06:00] least one of us here tonight believes that Phil Collins and Ringo Starr had something to do with an unsolved early 1980s homicide.

Now we don't have, we don't have all the facts, but I have enough. Which one? Which 

Michael Crispin: one? 

Nate McBride: Which one what? 

Michael Crispin: Which homicide? 

Nate McBride: It's unsolved. I mean, it's unknown. It could be Any one of thousands of homicides. I mean, who knows how many were, I'm only, I'm only, I'm only thinking that there was one that we are sure of at this point.

Okay. Yeah, we don't, it's a, it's a John Doe, Jane Doe situation. That's a good question. But, um, yeah, the dots are coming together. The dots are coming together. 

Michael Crispin: Was the person drowning? 

Nate McBride: They died from drowning. 

Michael Crispin: Oh, okay, 

Nate McBride: got it. Intentional drowning, is to be believed. Um, and so, yeah, due to the cuts of 75 percent of our staff, uh, we're operating at a 4 to 1 ratio now.[00:07:00] 

We're going to have to do this over Zoom from now on. And we'll be heavily relying on AI to do all the work for us. So, just some station changes to the podcast. I don't know if you know this, Mike, but we had to cut off 75 percent of our workforce. And everything has to be done through AI. 

Michael Crispin: I'm glad you had to do that.

I didn't have to do it. So, thanks. Yeah, we had a lot, we had a lot 

Nate McBride: of dead weight around here. So, psh, gone. We're gonna fucking get it done now. We're gonna move faster. 

Michael Crispin: Cleaner. 

Nate McBride: And ironically, we're probably spending more. So, who knows? I mean, it'll be, it'll work out. I'm not worried about that. We're so nimble now.

We're so nimble. Finding the biotech winter, as it were. Yeah, it's nimble. Don't, don't give me details. Don't, details are for other people. We are just taking it. Just take 

Michael Crispin: it, man. Just take it. 

Nate McBride: Uh, if you missed last week's episode, what the fuck, oops, I don't know what else to say anymore. What the hell is wrong with you?

We finally, after however many episodes, [00:08:00] got to the big meaty topic. The 24 ounce T bone steak of topics. The wedge salad with extra bacon top egg. How do we actively preserve autonomy and IT? Mike? 

Michael Crispin: No idea. I have no clue. 

Nate McBride: Kevin? 

Kevin Dushney: Uh, 

Nate McBride: Okay. It's a big 

Kevin Dushney: question. I don't, I don't have a crisp answer for you. This is why we're going to talk about it, I presume, but 

Nate McBride: Oh, well, we talked about it last week, but tonight we're talking about a new thing.

But that was what we talked about last week. And it wasn't just about reactively protecting it when someone tries to take it away. No. We actually talked about building systems and processes that intentionally preserve it. That's what we did last week. Yes. It was written right in the script. So you both, you both could have just read that one sentence in the script.

Yeah. 

Kevin Dushney: Gemini didn't do a good job of summarizing it for me. 

Nate McBride: Okay. But man, we just laid right into it. I mean, we teed it up. We knocked it out of the park. [00:09:00] Actually, I think that's a mixed sports metaphor, but unless you're playing golf with a bat or you're playing adult t ball, in which case it's actually a perfect metaphor, as always.

So Mike and I were playing adult t ball last week and just fucking slayed the other team last week. You should, you should, you should go ahead and 

Michael Crispin: listen. 

Nate McBride: Uh, not just because we're awesome, which is now a fact, uh, from the internet survey we did, but because that show leads to this show. So 

Michael Crispin: naturally 

Nate McBride: five.

Segues perfectly into this episode six. Uh, so if you're just joining us for the first time ever, you're either lost, which is, which is possible. I mean, there's tons of IT leadership podcasts out there with similar names, the calculus of IT. So maybe you're in the wrong podcast. That's okay. Um, either way, you're missing out on some context.

If you didn't catch the last episode, but here's the TLDR version. Uh, the loss of [00:10:00] autonomy is happening and IT leaders need to actively start preserving it. Boom. There it is. 

Kevin Dushney: Yeah. 

Nate McBride: Basically, we could just stop right there. The whole season's done at this point. Uh, so that brings us to this week. So if we knocked a tee ball out of the park last week.

This week we are knocking it twice as far outta the park, double the distance

Trance: a.

I'm the calculus of I. T. We value autonomy. Through the code we [00:11:00] weave our fate. In the data seas we escape. Zeroes, ones that can't debate. We control it. It's innate.

Nate McBride: The topic is what to do when LinkedIn just shuts off your profile spontaneously. 

Michael Crispin: Oh, man. 

Nate McBride: Is that a thing? Actually, no. Um, that's the wrong script. But that is a thing. Let me tell you, let me tell you a little story. It's just going to segue for just one second. 

Kevin Dushney: I'll discuss LinkedIn. I see that in the script.

Yeah. 

Nate McBride: Yeah. Yeah. So, so there I was chilling, as I do, and I was getting alerts from LinkedIn that unusual activity was detected on my account. But you know, it's like somebody logged in from this location and it was me. I mean, I was at the location, I logged in. But I was bouncing [00:12:00] around between Waltham and Stowe and Maine and just kind of like triangulating and I would get these alerts and I'd be like, yeah, dude, it's cool.

It's me. No problem. Ignoring the emails. Well, on Sunday, uh, the folks at LinkedIn HQ just eliminated my account from LinkedIn and not just eliminated it, eliminated every trace of me. So anybody that looked had nothing. You couldn't find me. There was no, no chats. There was no nothing. 

Michael Crispin: No history at all.

Nothing. 

Nate McBride: Nothing. No history. Nothing. Like, it never existed. So So, instead of disabling, Scorched Earth? Yeah. So weird. Um, Scorched Earth. That's just gone. Lovely. So, I got an email. Well, so anyway, I tried to log in, and I got this notice on the screen that said, my account had been terminated due to suspicious activity, but there was an option to, um, To check, click on a link to verify myself and attempt to get it back, but no [00:13:00] promises.

So the link went to this website called Persona and Persona, if you know anything about them, they're this hugely huge, nearly as big as meta, but hugely huge data aggregation platform on people profiles that they then just remarket and sell. Like clear, but a little bit more like not on the level anyway.

So they're like, okay, you gotta, you gotta do a video of your face from side to side, and then also upload a picture of your driver's license, both sides, which normally for any website that I got locked out of, I'd be like, okay, here's a picture of my double middle fingers. I'm never coming back, but this is LinkedIn 15 or 16 years of my life, uh, mapped out on this thing.

So I did it. And then I sent it in, and I sat around and waited, and meanwhile I was waiting, I did the absolute wrong thing, which is I went to reddit, and I started reading reddit posts about what happens to other people that this happened to. Nothing good. Hundreds and hundreds of posts, [00:14:00] zero good. All bad.

So now I'm like, thinking to myself, well shit, like what about, what would a world look like without LinkedIn? And if you just sit back and think for like a minute about this, yeah, we'd all survive. I still have a job, you know, get a paycheck and all those things, but think about what would happen. I mean, it's unilaterally like a very terrible thing if that happened.

Yeah. Fortunately, I was a very rare case and that this morning I was notified. That they had reset, they had, they were unhappy with all the suspicious logins. They had sent me multiple warnings, which I had ignored because I'm supposed to respond to them. And they reset all my stuff and I was able to log back in and turn off my life was, was returned.

It does bring up that question of what would you do? What would you do in the case where your LinkedIn account got compromised and [00:15:00] someone just shitposted all this terrible stuff, thus ruining your reputation forever, or, in my case, just eliminated? And if you read Section 8, I think it was like Part 1 or Part 2 of their Terms of Service.

It's this wonderfully clear, explicit sort of paragraph about if we terminate your account for any reason, we don't have to tell you what it is. And we can do it at any time, 

Michael Crispin: period, 

Nate McBride: period, point, point. It's true 

Michael Crispin: of all of them. 

Nate McBride: So, um. People have lost 

Michael Crispin: full Gmail accounts and everything as well. They've just been completely shut down from everything.

Yeah. Completely lost all their accounts. That had happened. 

Kevin Dushney: And it's a similar thing that, hey, we'll, we'll look into it and maybe you'll get your account back. In the case that time, it was, the answer was no. 

Nate McBride: You're just done. So I don't have, I don't have like an answer to the question like what's the alternative?

Obviously lots of ideas went through going through my mind like I could go ahead and build out a profile on my website and Point people [00:16:00] to that, but that's not really interactive. So something to think about but 

Michael Crispin: Substack could be a good place You could do a lot on Substack. 

Nate McBride: Yeah, 

Kevin Dushney: Substack is good.

You've lost all your reach though, right? Yeah, yeah. I mean, I lost. 1, 500 people or whatever, it doesn't really matter. It's in the thousands, so. Yeah. You lose all that reach, which is the big problem that we create, right? 

Nate McBride: So it's very interesting, and I urge everyone to A, make sure that your account is totally secure in LinkedIn, B, also if you get emails from vendors that are like, hey, unsuspicious, suspicious activity has been detected, obviously verify that it's you, and then B, read the whole email and see if there's other actions you need to take, especially if it's from LinkedIn.

Um, I do pay, I do pay for the premium subscription, and I'm hoping that that in some way expedited my case. I 

Kevin Dushney: was going to ask, because if it's the free, like. You know, you're the product, but you're paying in this case. So maybe that, what's that, what's that 

Nate McBride: saying? Like, if it's [00:17:00] free, you're the product, you're the product.

Yeah, you said it. So anyway, that's not the script. I just wanted to go off on that little tangent there, but the script is tonight about creating technical diversity without chaos. And I, as I'm writing that and I was writing the outline for the season, I'm like, without chaos, I mean, it's gotta be a little bit of chaos.

Like I'm going to artificially manufacture chaos when I do this anyway. So then I'm going to, change that to say. About creating technical diversity with just a little bit of chaos, but not so much that anyone would notice. So basically without chaos, but that's too long of a title. So it's basically creating technical diversity without chaos.

And we'll call it that. Although I will stipulate there's chaos, which is a paradox, which is what we do on this, on this show. We just create paradoxes everywhere. I would have done a jobs update this week. I was planning on it, but I only got my LinkedIn account open this morning and I didn't have time today to go and look and see what all the open jobs were.

So once again, urge you to go find them. If you're looking for a new [00:18:00] job, it's probably the same as two weeks ago, but next week I will have a jobs update. It will be awesome. Cause it will probably be the same as two weeks ago. Also with every podcast, I say this every week, but I'm going to keep saying it because it's important that we're going to release it in two formats.

One will be us talking, uh, and the other will be an AI version redone by our good friends at Notebook LM. Um, because we have Kevin with us here tonight, and there's only two AI bots available. It'll be interesting to see how the two AI bots deal with having three guests. Um, because usually one will take on my persona, one will take on Mike's.

Maybe Kevin will get his own AI bot. We'll find out. Suspense is terrible. I know. And we've been saying all season that we will get to an hour long episode time frame. That's not gonna happen tonight, and not in any fucking way. But we were close, like, at one time. I think we, like, episode two, we broke into two parts that were, like, Hour and 15 each.

[00:19:00] So not so bad. Um, so anyway, it is what it is. We have a Slack board that's growing sort of steadily. Anyone's welcome to join. The link is in our Substack page and in the episode description notes, it's Slack. So it's easy to come into. It's way better than the competition that I won't name. If you want to continue the competitions on our show or deep dive in anything else, you can either find us on Substack or find us on Slack.

Tell us how you're thinking and feeling and all of your, your deep dark thoughts. I also want to mention that if you like our show, please give us five stars on Apple Podcasts or Spotify or YouTube or wherever you listen to the show. I'm sure there's other platforms out there. I just not cool enough to know what they are.

In our show description, we do have links to our BuyASAPir portal and our new new merchandise store, which is not really new anymore. We also have a link to donate to Wikimedia, which, um, if you want to sort of at least retain the coolest corner of the internet. Uh, for free speech, you can just give them a few bucks.

It doesn't take much. Two or three bucks is fine. And then you can also buy us a beer because beer is good. And so [00:20:00] is keeping, um, freedom of press alive. So as I said last week, the focus was on actively preserving autonomy and IIT, not just reactively protecting it, but actively, actually building systems and processes that intentionally preserve it.

So this week we're tackling something that's fundamentally connected to that autonomy preservation, essentially the next step in the process, which is how do we create and maintain technical diversity without unleashing complete chaos. So here's the thing about technical diversity. I'm just going to lay the context and I'll shut up.

It's not just about having different tools in the toolkit. It's about maintaining your ability to make strategic technology choices. Because if you can't effectively manage diverse technologies, you're going to end up locked into whatever's easiest, which is a zero decision, not what's best for your organization.

Another way to think about, every time you standardize on a single vendor solution, you're trading some level of autonomy for convenience. Sometimes that trade makes perfect sense. It's actually a one decision instead of a zero decision. But when you lose the ability to effectively manage diverse [00:21:00] technologies, you've essentially handed over your technology destiny to whoever's got the most complete suite of integrated solutions and that's integrated in air quotes.

And something that you and I experience a lot, well, you, the three of us, experience a lot, which is that a vendor's idea of integration generally means replacing half the tools with their ecosystem. That's not an integration. That's a simulation. So I'll pause there. That's the baseline for tonight. Okay.

Technical diversity with the minimal amount of chaos. Any thoughts?

Michael Crispin: Paradoxical to some extent, right? It's hard. It is hard both sides of the scale, a lot to think about. 

Nate McBride: We'll talk about the paradox because it depends on when you started in the company. Yeah. How far, how far downstream have you started since the company's beginning? What's your role? How much, how much autonomy do you start with?[00:22:00] 

Um, all things that will contribute to the chaos level effectively. 

Kevin Dushney: Is this a more, is this the varsity version of 

Nate McBride: Shadow IT? Kind of. We're going to get to that too. I mean, it's a, that's a great point. I didn't really think about it from that angle, but yeah, it's, it's like boss level, shadow it almost. So here's where it gets interesting.

The challenges in the technology is the organizational capability to handle diversity in technology. So I've worked at enough companies now, as have the two of you and experienced in each, what I call like the pendulum swing of standardization. I couldn't think of a better phrase, but it works. So phase one, total chaos.

Every department has their own tools, nothing talks to anybody else. And IT is trying to keep the lights on, aka day one. So when I walk into a company, that's what I'm seeing. I'm seeing phase one. And it's kind of like, it's chaos, but it somehow works, which is [00:23:00] the wonderful joy of it. Uh, but they've usually called in a new head of IT because it's just starting to completely be on the edge of collapse.

So then you have phase two. So what do we all do as IT leaders? Well, we start standardizing. So phase two is the great standardization. Someone decides, usually the head of IT, enough is enough. Could be somebody else though, who's hired you. It forces everything into a single vendor ecosystem. For about six months, everything seems great.

You've moved all of your shit into one thing and login. My places, my stuff is right here. It's great. Then, uh, maybe towards the end of your first year. The reality check happens where people realize they've traded flexibility for standardization. They, they don't have the autonomy they had. And so they push back on you as a result of that.

And shadow it Kevin to your point starts emerging because the standardized tools you've provided to reduce chaos, uh, don't meet all their needs anymore. [00:24:00] And then you have phase four, which is the overcorrection, which is. You know, we need to be more flexible. And so suddenly now every new tool request gets approved in the name of agility and you're right back to the beginning.

Michael Crispin: Yeah. 

Nate McBride: Yeah. Loosely coupled architecture. Loosely coupled architecture. Right. And in the worst cases, that new world of chaos gets layer on top of your. World of not chaos. And so you have two distinct sort of parallel threads. One is completely a mess and the other. Well, if it worked great, it would be no problem.

Um, But the key, the key, the key to this message tonight is breaking this cycle, and which is what we're going to talk about. So instead of swinging between chaos and rigid standardization and going back and forth, and then this IT leader leaves and the next one comes in, they decide to do it all over again a different way, and so on and so forth, you have to build sustainable, sustainable diversity.

And this means [00:25:00] having clear boundaries between systems, so you know where one tool's domain ends and another begins. And we're going to come back to this point in big context in a moment. Uh, well defined integration patterns, not just like technical APIs, but actual data and process flows. And it's not a.

This is not a varsity skill. I mean, anyone with a lucid chart license or even a free gliffy license can go ahead and draw out a process, like how data moves a strong governance that enables rather than restricts, which is what governance should do. And then the organizational capability to manage multiple technologies effectively.

And that's just basic, you know, project management, governance, prioritization, governance, spending, governance, budgeting. So. Let's go back to number one, clear boundaries between systems. So I was actually walking with a good friend of ours this morning, Kate Bosman. We were talking about this and we were talking about the context of Airtable.

Airtable is a wonderful tool. And if [00:26:00] you said to me, hey, you know what we need? We need to get a database cloud solution. Boom, Airtable comes to my mind. That's what I'm going to buy. But then Airtable, the moment it comes in as a database solution, if you don't specifically say it's only going to be a database solution, it immediately becomes something else.

Because of what else it can do. Yeah. This is where it's a Swiss army knife. So, yeah. Right. So, so this is where the boundary question comes in. How would you establish a boundary around something that you know? Or potentially come to learn later, can do more.

Um, now it could be, again, this is what we talked about this morning, but it could be phased. Like, okay, we're going to bring an air table just for database, and we're going to do that for a year. Then, in year two, we're going to start using it for project management. Then, in year three, automation, right?

Sure. [00:27:00] You're setting clear boundaries, and you're setting expectations. But, um, back to the shadow it part, well, as soon as somebody in the organization learns that they can write a automation, then you're, it doesn't matter what the fuck your plan is, they're going to jump to phase four immediately. Yep.

Yeah. So how do you guys do that? How do you guys create boundaries between systems? I mean, are, are you doing it? And if you are, how are you doing today?

Michael Crispin: Largely for any system that's highly configurable, that's usually based on a use case. Um, at least to some extent, if there's a need for a highly configurable system, it's having a clear use case that you're going to set up against it, I think is a big, is a big driver as to why you might use something that Has a lot of customized customization, configurability in terms of instead of an out of the box solution, right?

See that that you're gonna put a boundary around an existing system. It's [00:28:00] you've got to have an intended use for it, and you sort of stick to that. You give it a maybe it's a one year road map, especially if it's got more capability. I think largely that the old school SharePoint model of we start with team sites and we don't do anything else, and then we move on to.

Applications or we move on to like, um, you know, site collections that are specific based on use case. And then we move on to full blown applications and and customized workflow. It's like, because you're building along the skills, you're hiring the people to support it. You're putting the guardrails in place as to what each level of service that's being created to support the application.

is and how it's defined. Um, yeah, you're having standards. You're trying to build a standard against it. But I think your approach, Nate, you're saying, like, especially with air table, like to me, just saying it's a database is giving it a very, very wide net. I would say I'm using it as an I. T. Database. [00:29:00] Uh, and we're not going to use it for anything else.

We need to master it in IT. It's going to be our CMDB platform, and that's it. And as we build scales, scale it out, as an integration tool, maybe that's phase two or phase three. But I also think these tools, you need to learn about them. And with Airtable, you learn about something like that so quickly. And you see all the Uh, opportunities and I don't necessarily believe in putting a boundary around some of those tools when there might be small to back to the use case discussion use cases that you could get value out of a tool like that right away as long as you document it and you have the skills in house to support it and you can support it in a business continuity context after you're long gone or the person supporting it is long gone that it can run itself.

Yeah, or it can be supported by someone who can pick it up quick. It's well, 

Kevin Dushney: well documented. 

Michael Crispin: Yeah. So [00:30:00] yeah, use case, um, people and skills, huge piece. Um, and then I would say just the, the guardrails and the governance. How do you put it around the application? How do you use it? What's it used for? But I'm not saying you can't break the guardrail.

I'm saying like, look, if you find a use case that you don't want to wait two years or a year, just because that's your plan. Um, if there's some big value proposition that you feel confident your team can support, because it was kind of both. 

Kevin Dushney: I think, yeah, I think your earlier point about governance being an enabler.

Um, this is true in this case, right? So it's, it's guardrails around the use of the tool using Airtable as an example. But, you know, on one hand, you want to provide those guardrails and have a roadmap based approach, but you're going to have some sharp creatives in your organization. Do you want to attenuate that, you know, desire to explore the tool and come up with really creative ideas within the guardrails [00:31:00] that you're not training your general population on?

I, that's an interesting question 'cause I, I'd like to see much like ai, some people are just gonna, you put co um, co-pilot or chat GBT on their desktop. Maybe somebody uses it once a month, like Agile 10 times a day and automating and creating scheduled tasks and, you know, API integrations and things like that and gonna jump right to the advanced level.

So I think the governance is, is huge part of it. 'cause it, it could enable you to allow. More creativity and more flexibility in the model. And if you feel like you're running afoul of the guidelines or guardrails, come talk to me, the GC, whomever's, you know, whatever the team or person is, that's responsible for that piece of governance and have a conversation.

Michael Crispin: I'm just thinking more of that, almost like two things to get put two things together, like an agile governance approach where things are more iterative and you're sort of building it, [00:32:00] building it as you go in some respects. It's so you can move with the pace of how quickly you might learn capabilities within a tool.

Kevin Dushney: Yeah. Cause if the person isn't bringing on undue risk or exposing IP, for example, then, you know, I would think through the governance lens of like, what's the harm letting people experiment. Right. I worry more like shadow it about tool duplication. You know, people buy four or five of the same thing and you walk in and you're like, why do we have all these, you know, this is more of bigger words than biotech, small biotech, but multiple document management systems and I like smart sheet.

I like project. And you're just. Tool sprawl to me is more of a problem and, and that consolidation is a, is a hard, hard sledding, right? Because people have their favorite tools, but if you want to really start integrating a tool or, you know, a platform like Airtable into a workflow, you can't have multiple.

[00:33:00] Otherwise, you know, this, this section of the company is using this and they're over here and, you know, data silos emerge immediately. 

Michael Crispin: We talked a little bit last week about having exit strategies around applications and planning ahead on that. I think that's a key to standardization is if you give them a year or two years, even, um, if you can, uh, to, to sort of show, do the storyboard, if you will.

And they're gonna, it's always going to be a 50 or 60, 40 people don't want to move. So it's just how do you put it up front what you're trying to achieve and not get into this circle of sort of chaos or turkey talking about where we end up adding something stapling something onto the side just because our standardization effort standardization effort failed 

Kevin Dushney: or 

Michael Crispin: committed 

Kevin Dushney: to this damn thing.

Nate McBride: Also, if we take so I mentioned the four parts of the. Sustainable diversity cycle, again, clear boundaries, well defined integration patterns, strong [00:34:00] governance and organizational capability. And if we, if we were to think of the ideal scenario, what I came up with, when I think about the ideal way to control chaos, it's not fun.

I'm going to give you a sort of a trailer preview of what I'm about to say. It's not fun, but a company would only get the minimal amount of software. to get the job done, ever. Not one piece of software more than what was needed to do the job. And that meant, a platform that could do more than one thing, would be considered over something that was, say, only able to do one thing.

Uh, number two, the company would create an evolution map. And I love this idea, and I have tried to do this in Lucidchart. I've actually done it in Lucidchart. Uh, it's crude. But an evolution map of every platform that not only includes what it is primarily responsible responsible to do But what else it can do and then mapping out those possibilities So I did the last did this [00:35:00] at Ohana.

So we're now talking like four years ago Yeah, but it was cool because I was like, okay, here's all the shit we do. Here's why we bought it or you know here's how the Here's what was told to me about why we have it, but then here's what I can do. And then you look at that second list of what things can do, and you start seeing all the Capabilities that are the same.

Like when you just make a, if you just drew this on a whiteboard, by the way, it doesn't have to even be a fancy map. If you make an evolution map anywhere of your world, you'll see just where the redundancies are at Kevin's point. And then the third and last is as the company evolved. either in growth, or executive change, or in its strategy, any way that it grew, which is to say changed, these evolutionary triggers would come into effect.

Okay, we bought Airtable for a database. Now we've evolved to X, so now we're going to also use it for automation. Now we've evolved to X, we're going to do X, Y, Z with [00:36:00] it. These evolution triggers would come into effect, and the company would move to the next iteration of the platform or technology based on that catalyst.

When this happens, we will now Open up this functionality. So I think those three elements, if they were to be done, would create, I'm not saying they would work perfectly, and there'd be a lot of pushback on this because they're very boring and they suck, but if they worked. They would create a very, very technically homogenous and forward looking environment.

But that's the simplest possible way to say that's very, very difficult to do any one of the three of those things. I mean, writing an evolutionary map is not hard. But then once you do it and you expose yourself to this knowledge, the very next thing that comes is extraordinarily hard. Which is you either forget about it or you do something about it.

So, if it sounds [00:37:00] simple, Uh, sarcastically, it's not, but this is where the autonomy preservation piece becomes critical because back to autonomy preservation from last week, maintaining technical diversity isn't about technologies. It's about making, maintaining your organization's ability to make independent technology decisions.

Like we said before, so. Let's talk about skills development for a minute. If your team can only effectively manage one vendor's ecosystem, you've lost autonomy, regardless of what your contracts say. I mean, you just lost it. You need what I call technological technological polyglots and a polyglot, a polyglot, much like a polymath is someone who can work effectively across different platforms, jack of all trades.

We like these people. I think all three of us are big, big fans of, of Jack of all trades people. We hire them and understand the principles behind them, not just specific tools, not only what it does, but why it does what it does and where that's important in the organization. These polyglots, and I think each one of us has at least one, [00:38:00] are the people that get it.

And even if they're not in IT, if they're in your org, they just get it. They understand, oh yeah, this thing can like, you know, chop lettuce and cut through shoes. Makes perfect sense to me. I get how it affects the business. So this brings me to something I've been, I've been experimenting with, and I think both of you have too, which is this concept of diversity where architecture.

So, and these are just ways to frame things, but instead of trying to standardize everything, we explicitly design our systems to handle technological diversity. So I buy platforms. And of course, this goes back to my very early point about introducing chaos. I do introduce chaos when I do this, but I buy platforms for their extensibility, not for their singular focus, whenever I can avoid it.

So I try never to buy a platform that can only do one thing. Now sometimes the business says, [00:39:00] yeah, we just need a platform that makes a diagram of this particular molecule. That's it. It's all it can do. And it's 10, 000. I'm like, it can't do anything else at all. They're like, no, it's all it can do. It's the best thing for it.

I'm like, well, shit. All right. But every chance I get, I will buy for the biggest bang for the buck. Do you guys do that? You guys do that, right? 

Kevin Dushney: Uh, I mean, it depends on what it is. You know, what's the budget, what's the roadmap look like? Are you buying a point tool? You're going to throw away, or are you making a big investment in a longer term commitment?

So, I don't know. I think there's a bunch of variables that. I think 

Nate McBride: one of those variables is the resources you have 

Kevin Dushney: if you have no great point, right? If you have a very small team, um, and also you don't have the people on the other side. I mean, I make this point all the time. Hey, my team might be ready to go.

But if you can't commit in your area of the business, the time it takes [00:40:00] to complete the project, which is why often as vendors, at swag yet. Hey, how many hours per week? So it's kind of like, how much per team member do you need in the business and an IT. And if the math doesn't work, this is heading towards a failed project.

Right. So you either need to commit more or come back later when you have more time.

Michael Crispin: Yeah. I think, uh, sometimes the tools that do a lot of different things require a lot of configuration management and a lot of discipline, or change management. Yeah. Change management. And it just, that's, that's one of the things you're going to weigh into it is, 

Nate McBride: you may 

Michael Crispin: be able to do a lot of different things, but.

Yeah. Can you, can you show how it works? Can you document it? Can you, it's almost like it often lends to a build versus buy because a lot of, a lot of tools don't do one or two things really well. And then they tack on a bunch of [00:41:00] other things, um, to, to try and build a value proposition, 

Kevin Dushney: including AI, 

Michael Crispin: including AI, all of a sudden, Hey, you 

Kevin Dushney: want AI here?

It is, and you can't turn it off or we're going to make it really difficult for you to do so. 

Michael Crispin: I think there is this central place, you know, of these, I even call, I call them more platform as a service offerings, you know, more that software as a service tools like air table or building their own ecosystem of tools.

And like, uh, zoom is, you know, trying to become more like Salesforce. And, you know, there's just this emerging from a software as a service type offering to a platform as a service. Is, is, uh, where, what gets you back into, do I invest in office 365 and the entire Azure stack and the AI stack there? Do I, the, the, the, the sort of best of breed, I guess, model where if you want to call it, that is, is shrinking because a lot of the great tools of the [00:42:00] past have now built platforms and you can buy one platform for 17 a month and get all this extra stuff.

I said, it's a tough decision to make. Um, Consolidation play, you know, um, from a support people perspective and continuity perspective, it's sometimes a hard argument to make. Buying like you're saying, they buy like one thing that does one thing really well, it's getting harder and harder to justify that approach because.

Kevin Dushney: Yeah. I mean, I see, I don't know how I don't follow, you know, Google as close, but on the Microsoft side, you know, there's so many add ons now it makes your head spin for security insider risk management. Um, you know, Intune and now Intune can go assess the performance of all your machines and tell you which ones are dogs in your fleet, but it's a charge, but so that's, you know, you're getting all these capabilities on one platform, but at the end of the day, you also have to look at.

You know, what's your cost per user per [00:43:00] month? And, you know, that starts to spiral once you start adding all this stuff up together, 

Michael Crispin: then there's the eggs in one basket thing, right? You're everything's in Microsoft or your, your core database is just snowflake or just air. This is the paradox. Yeah. So it's trying to like, like we've spoken another, uh, podcast.

I think it's. There's just this middle road you need to weave and build out based on your expertise and the tool sets and the marketplace and your peers around you and learning from them and being able to listen and hiring the right people that you kind of mold your own way. And, um, I do think that kind of an agile and flexible approach is important with a lot of these.

Decisions because your business conditions can change in weeks or days in life sciences and it's kind of 

Nate McBride: But Mike, I agree with everything you just said, but you, you completely, in your little list that you just gave, you skipped over [00:44:00] governance and evolution and planning. You went right to the experience and the usability.

Like these are all again, key factors, but I think we all, we all fall prone to that when you talk about middle of the road weaving. That we're going to rely more on experience. I do. So I'm going to just say this. I'm kind 

Michael Crispin: of coining it agile governance. Like, yeah, more, more than, more than hard facts. Yes.

Nate McBride: Like I, I'm going to go by my, my gut instinct on what's worked and what hasn't over all these decades, rather than saying, well, what's the percentage likelihood that this will succeed by math? 

Michael Crispin: Yeah. And I think the governance component is, uh, don't want to be too controversial, but if you put too much governance around things, not only is it sometimes become the guardrails, and it's not enablement, uh, as it should be, it becomes That's uncontroversial.

Nate McBride: No, controversial 

Michael Crispin: in the [00:45:00] respect that the optics sometimes are you spent all this time building committees and governance, and what has it done? I mean, it's, it's hard to see it. It's kind of like cyber security until you get breached or something happens like, Oh, yeah, we gotta gotta do this. It's like you've got to be able to show why you need to put people on boards and advisory boards and and have documentation and make people train on it and make people do it.

Um, to, to, it's, it's difficult to do that. Yeah. In a small company. When everyone's trying to go a million miles an hour. 

Nate McBride: It's the governance for governance sake issue. Yeah. And oh, we're just going to do governance. That's, that's not, that's missing the point. 

Kevin Dushney: Yeah, 

Nate McBride: I hear you. Um, so 

Michael Crispin: definitely struggle with that.

I think a lot of us do is we're trying to get some semblance of governance in place or just even some, some rules and policies, just a basic policy. Sometimes it's very hard. For your [00:46:00] peers in the business. I think it's how you sell the 

Kevin Dushney: governance too if you're saying trying to say look we're trying to enable you with this tool but we're also trying to keep the company safe and yeah you know as much as fun we poke at AI it's a perfect example.

I want to get, I want to get these tools in your hands but it's not a free for all and We need to make sure that data is secure and, you know, we give you basic training and, but it's like saying, here's the internet, I'm going to use it differently than the two of you are in the same way. I wish is a bit odd in this case, but the governance I think is, is at least fairly clear where.

Yeah, we're concerned about IP or, you know, you look at deep seek as an example and, you know, try to understand, you know, where's your data going and do you trust, you know, like any cloud provider, do you trust your data with that entity? It's got to be baked into your governance, right? [00:47:00] 

Nate McBride: Yeah, yeah, it does.

And I, I wish, uh, just as a, uh, an aside, I use Zapier pretty much daily. A lot of my automations and my company run on is Zapier. And I use Zapier pretty linearly, which is I use it for automations now. I'd be lying if I didn't say that I've considered use cases for Zapier tables. I've considered use cases for Zapier APIs, like I've considered these things.

But I've always essentially at the end said, you know what, I already, I already have this functionality. I don't need to do this, but I'm going to, I'm going to know it and understand it, but I'm not going to actually implement this because it just adds more shit to the pile. But it's just a triggering system for you, right?

I wish, I wish somebody, and I wish it was me, had starting about two years ago. Created a visual diagram map of everything that Zapier has done to their product because even as of last week, late last week, they've now added the ability through Zapier to create AI bots, much [00:48:00] like Botpress does. Yes. 

Michael Crispin: With 

Nate McBride: basic, with basic language skills, it's all you need.

You can create a, now a bot in Zapier that connects to any platform that you have. They, they figure this one out and they've deployed it. And I'm like. It's unbelievable. Every time I log into Zapier now, there's another new thing. And I have to, in terms of this evolutionary paradox of these platforms, again, I pick Zapier to do one thing for me.

But if I was to go ahead and do an evolutionary map of all my platforms, I think Zapier would win. I think it would have the most capability of any platform that I have. And it costs almost nothing to Mike's point. It's got the most horsepower. 

Michael Crispin: It could do. It's incredibly extensible, right? I mean, that's a incredibly the difference between that and some of the other and we're talking almost about middleware here, right?

And that's almost a category. So I know. Yeah. 

Nate McBride: Yeah. Well, [00:49:00] it goes, it goes back to that point I made about like, yeah, if I want to make an API or if I want to connect a to be great for me, I've done that. But I should know and have it mapped out how every single thing travels in my organization. Like how?

Yeah. So I put Zapier in the middle of four automations. Where, where's it documented that I did that? Oh, it's in Zapier. I have a recipe there. No, that's not, that's not, that's not equitable. 

Michael Crispin: Well, I think I, I'd argue that that's one of the powers of Zapier and, and make is that the recipe, unlike building something in Informatica, right, is It's right there.

Anybody can read that recipe and see what it's doing. Whereas if you build an integration in Fivetran or in Informatica and you've got all these Now, if you're building code, there is ways to certainly use code and snippets and JavaScript and automations and Python inside of Zapier, but a lot of what people spend all this [00:50:00] money on Boomi and other tools for is moving data around.

Yeah, it's And in shuttling data from 

Kevin Dushney: system a to system b 

Michael Crispin: exactly and you're and you're paying for a a bucket, uh, a overall license bucket type spend as opposed to Hey, i'm going to do 5 000 transactions a month and this is what it costs. 

Kevin Dushney: Yeah We 

Michael Crispin: pay more for that than we need Oh, that's another dimension 

Kevin Dushney: and that, you know, may not make the value proposition strikingly better, but what's the ecosystem, like the partner ecosystem look like, right?

So we use, we use Ricardo, which is pretty easy to write recipes, but there's also a pretty decent ecosystem of partners that built me these recipes to connect EDP to fidelity to this. And sure, there you go. 

Michael Crispin: Ricardo is a strong enterprise ecosystem. In fact, 

Nate McBride: Ricardo, Ricardo has partnerships that nobody else can have because Ricardo has worked on single source relationships [00:51:00] and blocked out the competition.

Yeah. So they have things that only Ricardo can do for certain enterprise platforms that you, I mean, there are certain HR platforms that will only work with Ricardo. They will not work with Zapier. 

Kevin Dushney: Yeah, 

Nate McBride: because they've worked, they've worked out agreements and that's your point, like partner strength where Zapier is just like, and that's the problem 

Kevin Dushney: where if you say Zapier is my tool and it's like, shit, now I can't integrate it with my HR system.

Are you going to change the HR system or change the integration or add an integration platform? 

Nate McBride: Well, and there's your chaos. Exactly. Right back to chaos. So let's just talk more about this. Let's talk more about this middle area that Mike was talking about. So, if we're, if we're explicitly designing our systems to handle technological diversity, and this means, and this is going to be a little redundant to what we just talked about, but it means that we're building integration layers that expect and support multiple tools, rather than trying to force everything into one pattern.

We're maintaining clear data ownership, which we're going to come back to in a second, because that's a big one. I mean, we all know this [00:52:00] one. Uh, and we're maintaining clear data ownership and governance, regardless of which tools are processing the data. We're creating architectural patterns that allow for tool replacement.

And this goes back to what Mike referred to earlier with last week's exit strategy discussion. We should probably touch on that one more time. And then we're developing skills and processes that treat diversity as a feature, not as a bug. Now, I'm going to caveat that by saying there are fucking bugs.

And they're bugs, right? They're just straight up bugs. Like, the thing doesn't work. Not everything is an enhancement. But if something works a very weird way and you're like, Oh, that sucks. That's a bug. Maybe, or maybe you're just not thinking of it the way that the vendor wants you to think about it. Um, and I'll go back to Zapier for that example.

I want Zapier to do a certain thing, a certain way. Right. And it won't cause it's going to do it. It's way drives me bananas, but I can't get my brain to adopt its method. I keep trying to like force it, do something else. [00:53:00] So we take those things and we put them into practice. You have. You have to go back to these integration boundaries, which is okay.

This thing's going to do this and it's only going to integrate with that. I'm not going to let it do anything else. And there are, there are organizational agreements about where one system's responsibility ends and where another's begins. And I'll give you an example that's happening right now. So. Um, we're, we have a relatively new lab platform, uh, that we're, that we're goofing around with, it actually was acquired by a big ELN company.

You might know about, uh, that's in like popular in the life sciences industry. Anyway, they, they acquired this new thing right after we bought it and. We were supposed to be using it for a particular area. We, when we explicitly defined its role, but now that it's been acquired by a much bigger fish, that's got more responsibilities, the question came up like, well, should we have it do more?

And almost. [00:54:00] And almost 100 percent agreement that no, this is great. Like we're only going to have it do that one thing. That's what we bought it for. Now, again, this goes, my, you know, my position is always to buy for as many things as it can do, but even this case, it was very, very clear from the onset. We needed to do a certain thing.

And so we're going to continue to limit it to that. Now, evolutionarily a year from now, we may take a different look, but by maintaining these clear boundaries, we will have the ability to swap out systems. And this, we've talked about extensibility. And I'll, I'll stop in a minute to get your thoughts because I want to, I want to circle back again to another thing we talked about, but, um, I love the fact, and, and this, this is both chaos and it's autonomy preservation simultaneously.

I love the ability that I can swap out things very, very easily in my, my stack. Not everything. Like I couldn't swap out Okta tomorrow, but I could swap out Okta if I [00:55:00] had to. 

Michael Crispin: Yeah. 

Nate McBride: I could swap out, I mean it would be a little bit more painful than say swapping out something else but every single thing in my stack can be relatively easily swapped out.

And I think in many cases, the business wouldn't even know or would have little guilt care. That I think is the better alternative to chaos that would be created by trying to replace something that wasn't, you weren't able to rip out. I think honestly, between those two cases, it would be more chaotic to try and remove an ecosystem that was embedded.

For instance, Back to that ELN platform. If I was to try and rip out that ELN platform and replace it, it would be two years of absolute madness and pain. And anger, people would not like it and resentment and all kinds of shit would happen versus replacing a smaller or a different tool that wasn't so deeply integrated.[00:56:00] 

And I think it would be, it would be much less chaos to have it just so I could just pull it out. So I'll pause there and see what you guys think about that. But I just want to, I do want to come back to the data ownership piece right after this. Do you guys have any thoughts on that point? 

Michael Crispin: Well, I think if it's a core business system and you're.

You've got to have a really good reason as to why you want to swap that out. And usually if you have a good enough But wouldn't that 

Nate McBride: be in your exit strategy, like we talked about last week? 

Michael Crispin: Yeah, I mean, if you haven't, you certainly want to have some exit strategy in case something happens to the company or something along those lines, um, but I think you've got to sort of weigh in your portfolio of applications which ones are going to be more resident and create the most disruption if you change them.

What you're getting back. Are you saving money? You're probably not because you're causing countless amount of hours of frustration for people working, so [00:57:00] you might save a little bit of money, but pulling it back, it's going to have some sort of way to measure that. And I think everyone has probably their own approach to doing that.

But, um, I'd say those business applications is really weaved in pretty good. It's difficult to pull them out without a really good reason. It could be a cyber security reason. It could be a data integrity reason. It could be a company's going out of business is one that I, I know of, uh, you know, it's pretty popular, uh, laboratory.

System that had code that was written 26 years ago and the developer is not well, the one developer. So, you know, there's always, look, if you have that data and that information, you can make a pretty good case as to why you need to change because you're responsible for the integrity of the data. Yeah.

Yeah. But you've got to have some sort of [00:58:00] way of quantifying that and being able to explain it if you're going to pull it out. Whereas I think some of the more enterprise systems, you may very well have a presentation layer that's on top of so many of your systems that changing things in the middle or on the back is a lot easier.

More easily interchangeable as long as you can keep the data intact. Um, you know, whether it's cost reasons, scale reasons, better, better AI, whatever it might be that's happening underneath. Um, as time goes on, presentation layer is going to get more available and similar, I think. Um, so to your point Nate, the stuff underneath, the more that your architecture is like Lego blocks that you can shift around and move around, you'll be able to be more.

Agile, right? And 

Kevin Dushney: totally 

Michael Crispin: talking to machines. It's going to large it research 

Kevin Dushney: organization that has a term for this, [00:59:00] which, which, 

Michael Crispin: which, what's the term? Oh, can we use it? Composable architecture? Oh, 

Kevin Dushney: yes. And the Lego blocks are the Lego blocks are exactly what triggered it for me. Cause that's, that's how it's been explained.

Michael Crispin: Well, it's came from, um. Like, uh, an enterprise architecture. Yeah, like the, the IBM enterprise architecture was a composite diagram of business services. Yeah, well, that is architecture. And that's where they, I think they, you can do what Nate 

Kevin Dushney: saying is, you know, plug things in and out without. Major disruptions to the business and yeah, you know, maybe that's even a place to start a Hey, what's the inventory of all my platforms tools integration out and which ones are easily swapped out with the least amount of or change management and what are the heavy hitters that this is a long slog to go through.

And you could also, you know, do an evaluation to say is the juice worth the squeeze here to change that versus [01:00:00] something simple like, you know, email security filter. Yeah, might be a couple of notes you need to send out but who do you really care if you're using minecast or some other tool. No. That's that's an easy one.

But if you want to say, Oh, I'm going to change, you know, Cooper tomorrow. Or imminently, you know, geez, if that's integrated into the, the way the company works for the last four years, that's a, that's a much harder decision. Unless it's driven by some other force like the CFOs, he says, look, you know, I don't want to pay for this anymore.

It's way too expensive. Let's, let's use the one integrated into NetSuite, even though it stinks, it's cheap. There could be a lot of drivers here, but I think your point, Nate, the stuff you can swap out easy is. It's the way to do it and get as many of those on the ledger as possible. And, you know, the things that are really hard to change, I think you're going to know about.

It's really more of the, the enterprise platforms, like even on Google and Microsoft or vice [01:01:00] versa. Switch is a big undertaking and God help you get both. If you remember the. One of biotech that we're both aware of that tried to continue to try running both simultaneously and the poor soul that had the head of it job there.

Oh my God. Yeah, 

Nate McBride: that's 

Kevin Dushney: a, 

Nate McBride: that's like a worst case scenario. I think. Yeah, 

Kevin Dushney: it's the only time I've ever heard of that scenario, but it's probably not the only case, but I'm just saying it's just like the worst nightmare of trying to stitch that all together. 

Nate McBride: So I want, I want to come back to, um, the point about, we were making before about the, um, clear data ownership, and I don't want to spend too much time on this, but I just want to kind of like, you know, we've talked about data ownership ad infinitum on this podcast in season one and beginning of season two, and data ownership is a very, very tricky thing, right?

To put it, [01:02:00] put it bluntly. So, yeah. Data ownership, though, does impact everything else we're doing in technical diversity without chaos. So, every single time that a new thing comes in, whether IT brings it in or the business brings it in, whoever brings it in, they're creating new data. 

Kevin Dushney: Nate, do you think of, sorry to interrupt, but do you think of data owner as, you know, synonymous or analogous to business system owner?

Or do you see it as a sort of event of 

Nate McBride: I don't. And that's a great question because they can be the same, but oftentimes I see the business system owner as somebody who's in charge of making strategic decisions for the function or that area. I see the data owner as the person who's responsible for making sure that all the data in that thing is clean, operational, effective, uh, measurable.

Yeah. Again, they may be the same person depending on the N of X. Issue with that function, but [01:03:00] 

Kevin Dushney: you know, I may be, I may be thinking through like the GXP lens there, where like quality is the data in the business system owner or rim and things like that. 

Nate McBride: Well, but the data owner, it can be, there'd be a person, it can be a function.

I don't think it should be. I think it should be. All attempts should be made to deamb, well, to make it as least ambiguous as possible. Just sort of have a name next to a role which is designed around that, that, that particular issue. But data ownership is funny because if you want an extensible plug and play architecture, which we do.

I mean, I think we're all in agreement that, that middle of the road capability where you can have a thing that when push comes to shove, you need to pull a block out, replace it with a new block, you can do that. Great. There's data though. There's data that was there, and now there's going to be a new thing, so the data's got to go in the new thing.

And maybe it's a clean transaction, maybe it's not, but, [01:04:00] um. You're creating, you're creating a data problem. And so we look at it from a platform lens. Oh, we're just going to rip out this platform for this new platform. And no problem. It owns the platforms, but there's a whole bunch of data in there too.

And I just throwing this out there, but whose responsibility is that data? I mean, if you were to. Slap a label on it. There's a data ownership element to this extensibility, you know. We don't just get away scot free by ripping things out and replacing them. 

Michael Crispin: Yeah, you need to have clear data owners that are, I would argue, are not IT people.

Um, they should be the people who know the data the best, and know who need access to the data, know the compliance rules against the data, um, and even to some extent, Controlled access controls, um, to the data in some, some scenarios or maybe all scenarios because ultimately I think as a CIO, your responsibility is to make [01:05:00] sure that they are working within the guardrails of, of whatever data management policy you've put in place or, or overarching oversight that you've put in place with data governance or what, what not.

Um, but because there's so much data and there's so much actionable data in a company. I think you'd be at a big single, single point of failure to have expectation that maybe it's an IT person unless you have a large IT department, I guess, uh, you know, kind of business technologists within your group.

Um, I think there's a data ownership component that's really the people who produce or manage or bring in the data should, should be accountable for it as long as you're providing them the right tools and framework. That supports it. Um, it's a partnership at the end of the day, but they're ultimately responsible for the data and accountable for the data.

Can they, can they say, well, you put no DLP in here and we [01:06:00] lost the data? Well, there's certain policies. They should certainly confidentiality policies and other things that people should be following. You should, you should be trying to enable that as an I. T. leader or cyber security leader to make their job is easy and they can't, they can't make human errors or mistakes.

So that's your job kind of from an oversight and, and, uh, protection perspective. 

Kevin Dushney: I agree. Hey, you know, 1 interesting way to drive a data ownership exercise is to undertake. A data retention policy, and we just did this and you're basically canvassing all groups. Okay, what data do you own? Do you manage, you know, what are the requirements around keeping this, whether it's just business practice or a compliance need.

And, you know, it's not all that much of a fun exercise, but it's informative and you learn a lot around, well, [01:07:00] geez, I don't know who owns the data and you got to hunt that down or yes, I own this and. I can talk to you about that. And ultimately you've got someone in each group that's responsible for answering the questions about that data.

It's a 

Michael Crispin: map 

Kevin Dushney: is huge. 

Michael Crispin: Yeah. Yes. You've talked about that. And Kevin, you just mentioned it, right? It's the data map is, is the core artifact that drives the discussion. You don't know who to go to about a set of data. 

Kevin Dushney: You're, you're sort of fumbling around in the dark, right? 

Nate McBride: It's such a beautiful document when it's done.

I mean, it's such a wonderful thing to have. And it's useful for such a long period of time, especially, you know, as companies transition from the early entrepreneurial phase into the, the rules phase, into the commercial phase, they just turn over so many people, each of those sort of pivot moments in their, in their growth.

If you just, if you just don't know, it just, it just, the [01:08:00] debt mounts so fast. And I would, I would argue that. The less, the less control you have over your data, regardless of what's happened to it over time, the more chaos you're going to have just by default. 

Michael Crispin: Yes, absolutely. That might be a cool jort or an episode of just to go through maybe a.

A data map mock, like just, uh, just showing like what that looks like, what does a data map look like? I know it probably has their own approach, but maybe we could come up with a few simple tips on how to do that. Because I know that that that's a big struggle, not just the actual artifact creation and what it should look like.

So it's legible and understood for people outside of, uh, sort of data architects or it leaders. Um, but also the conversations. Need to have to draw some of that out. If you're doing like a risk assessment, you're often asking some of the same questions. And it's, um, it might be [01:09:00] interesting thing to cover it.

Maybe we can do a screen share or something, 

Kevin Dushney: you know, or an AI assessment. I mean, these are all very current topics, you know, for, for me specifically, where, you know, people want to jump right to the end game and say, why can't we put a on AI on top of this? Like, well, guys, you know, you're not calling.

These compounds, the same thing, controlled vocabulary is not controlled and people are using different synonyms. So not only can I do AI effectively, I can even build you a decent dashboard from this, right, without cleaning this up. So that's, I think that's another strong reason to have a data owner.

Because that person ultimately be responsible, at least helping you to define, you know, data governance, right? That it's better that we take a whole section of like research, for example. This is, this is how we call data as a nomenclature. Otherwise, you know, it's hard enough to look up and down a program and ask questions.

Nevermind a cross. Yeah, [01:10:00] exactly what we're suffering with right now is, you know, a, we don't have enough compounds to make predictions and the quality of the data around it is, you know, it's dubious at best in terms of even, you know, natural language queries. So, you know, having the owner and an early engagement, I think in the, especially in the AI age is, is even more critical.

So you set yourself on the right path and get that data governance involving those owners. 

Michael Crispin: And that the common language, like you said, the data dictionary, Exactly. Common terms. 

Kevin Dushney: What's our ontology for all our compound naming and omics data? And because if you want to pull it all together and do a report, visual or, you know, ask questions, it's got to tie out, you need an architect for it, otherwise it breaks immediately.

Nate McBride: So there's, there's two more things I want to cover tonight. Um, I mean, we could go, we could go, I think this could be, this could be a freaking multiple [01:11:00] part episode and we could go for hours, but I want to cover something we mentioned earlier, which is the evolutionary triggers. Um, and then I want to talk about metrics, uh, everyone's favorite is topic, but first the evolutionary triggers.

So I mentioned just three now. Um, you know, the pivot from entrepreneurial to, to, uh, like what I call the adulting phase, and that's where generally it comes in. So a company started, everyone's got a credit card. They're buying all the shit. They're doing everything Someday. One day someone realizes, holy crap, we've got a IPO this year, or we've got this big launch, or we got this big change.

We've gotta move to a new facility. Yeah, we gotta hire a head of it. That's end of phase one. That's the end of the glory days. 'cause now the rules come, the bosses come, the adults come in and they start setting all the rules and process and governance. You're ruining all the fun. Right. And all the people in phase one are like, F them.

We're out of here. We're going to go find someone else somewhere else. We can do phase one again. And they do, they go as a 

Kevin Dushney: big company. It's too big. I'm like, it's a big unit. [01:12:00] 

Nate McBride: No. Well, then what happens is a couple of years later. Uh, the company is like, Holy crap, we have a product. We've got to sell it. So they go and hire all these wonderful commercial people and marketing people and analytics people.

And they're all like, well, we don't give a shit about you face to people. We just got to sell stuff. That's why you brought us here. So don't give us process and governance. We just got to do shit. Like we got to make money. And so they're the ones who are going to start making you really hope that during that phase two, you got all of your shit together, but.

These are three grossly blurry evolutionary moments. 

Kevin Dushney: Yeah. 

Nate McBride: But when we think about evolutionary triggers in general, and again, we're doing this as we move towards keeping technical diversity but reducing chaos. When you think about evolutionary triggers and that idea, what are [01:13:00] some other evolutionary triggers that come to mind?

Kevin Dushney: Um, I mean, you need the classics one, you know, the classic ones in biotech anyway, the R to D to C, right? Or in my experience, you know, huge inflection or pivot points, right? And, um, like you said, research is everyone's enamored with that and that developed, you know, development people come and waves and then it's commercial.

And what about us? So, um, but I think, you know, appropriate in this, you know, so called biotech winter is downsizing, losing staff. Um, you know, you've built some, you've built a bunch of things that you can no longer support, or it's more difficult to support, you know, that could be another evolutionary trigger.

Are you prepared for that? Or, you know, can you scale down? Are you trapped under licenses? So you have very complex workflows that 

Nate McBride: Can we talk about that for a moment? There's like, there's a [01:14:00] bunch of 

Kevin Dushney: examples, I think, that fall 

Nate McBride: into that bucket. Well, Mike, before you get into your triggers, I want to talk about what Kevin just said real quick, which is how, so to a degree, I think we all prepare, or those IT leaders that take the time prepare for an M& A, an M& A event.

And if we don't have a written down, we at least kind of know in our minds how it would go down in either direction, whether we were the buyer or we were the buyee. But in terms of a downsizing, do you ever, do you ever scenario play that? And do you, in terms of an evolutionary trigger, because it's so massive of an evolutionary trigger in terms of its ability to create chaos.

If you don't scenario play it, or even if you do, how do you, how do you account for that change? I mean, you're just going to be left with all this stuff. How do you account for that change? 

Kevin Dushney: I mean, I don't, I don't specifically scenario plan for it because, you know, my experience [01:15:00] is, you know, build, build, build until you hit that proverbial brick wall and then pivot.

Bye. It's, it doesn't mean you can't do things to help insulate like, you know, now this is easy in this day and age, but using SAS models instead of building out, you know, infrastructure on prem, which is, you know, no longer a thing, but even adding business downturn language into some of your agreements with some of the bigger players, if you can get it.

Yeah. Hey, I need to cut my licenses in half. I already, I know I committed to three years, but. We're no longer 400 people, we're a hundred or 200. So I think if you, if you're forward looking in those regards, you can at least save the company some money. Not sure if it helps you from a, you know, continuing to support all the platforms you've built, you know, and I don't have a good answer for that really.

I guess it depends on how big the riff is and what the impact is on the company and then what the expectations are. Is it keep the lights on for six months to a year? Or is [01:16:00] it, well, no, IT needs to keep innovating. Well, like, wait a second, you know, I lost half my team. I can't do that. So that that's, I think that's hard to plan for.

Michael Crispin: So the down downsizing kind of planning scenario is, you know, if you've got a, if you've got a platform or a technology or a certain skillset. Within the group, and this goes back to what you originally talked about, Nate, about having a tool that doesn't multiple things or an ecosystem or a platform.

Nate McBride: Yeah, 

Michael Crispin: it's having some sort of plan to to fold in to the commodity stack. So what I mean by that is. You know, often companies, they get huge and they can't, they're not going to buy everyone a zoom license. They're going to buy a team, they're going to use teams and they're going to have certain people have zoom licenses or they're going to, um, they're not going to have the expensive notebook program for every person.

Now they're going to have the basic one. Um, and, you know, going through some downsizing in my past has always been [01:17:00] kind of thinking about what's my second place platform and I already own it probably. So how am I going to transition certain things? over to not the ultimate tool that I think is providing the most value, but the good enough tool that is going to get us most of the way there so that I can just run, run, run, run, and not focus on growth.

And that was part, part of the approach on that was just really saying, we're getting 60 percent now instead of. 80 or 90 percent and setting that expectation up front from a business continuity perspective, even, um, in part of your business continuity planning is, is built into, uh, should be, should take into account downsizing or, you know, a lot of other unfortunate things, but the flip side is that that comes back to the Lego piece discussion too, if you, if you've got a good data source for some of those traditional IT [01:18:00] capabilities, you can kind of shift that over to the cheaper platform, um, Um, to, to prepare you for that and to have a, a basic strategy or just knowing that you can operate or execute with the right sourcing to do that in a short period of time.

Um, that evolutionary trigger ultimately can go the other way, which is we're, hey, we just got a ton of funding. You know, what are we gonna do now? You know what? How should we be doing things more optimally? Not that we should go to a shopping spree. 

Nate McBride: Right, right. 

Michael Crispin: Now, here's the opportunity to do some of the things that set us up for the next three years.

And we're gonna make the right investments. And that can be a trigger if you're hearing, you know, hey, we're gonna funding or we're going. Uh, well obviously we're going public, there would be like all these other compliance related things we need to make sure are happening anyway, but like what, what, what type of, um, opportunities, good things, uh, breakthroughs, inventions, things that can happen within your company, how do you plan for those?[01:19:00] 

Uh, and support those and be along for the ride so you're not lagging behind trying to move as fast as the business is. Remember you used to be IT that was ahead and now it's like in our industry these things can happen so fast and you're just trying to react to positive news. Um, so there's that side of the coin too but you're talking about the downsizing and I think that well I mean you can do talk about 

Nate McBride: other Other evolutionary triggers too.

I mean, it's just one trigger. Yeah. 

Michael Crispin: Yeah. No, we were focused on that. And it's just, I think that there's sort of a, um, a shift type model where you just go to maybe the, uh, a different tier of service and you have a strategy of how you're going to get there. That's usually what you're being asked to do anyway, is yeah, these, what can you stop doing?

Um, 

Nate McBride: yeah, I mean, I think a lot of times it's 

Michael Crispin: hard to answer that question 

Nate McBride: when we think about. technical diversity. I mean, we have to no matter what happens. [01:20:00] Um, and we talked about the ideal before, which is you have to have a plan at all times to evolve to the next level. Now, what the next level is can be one of 10, 000 things, obviously, but you should at least have an awareness of what is maybe most likely.

What are the most likely things to happen? And then how would I evolve with the least amount of chaos while still preserving autonomy and not being forced, say, into an ecosystem where I've just conceded autonomy or not being forced into making a decision that's actually the wrong one for my company, um, as opposed to, say, one that gives you the best potential for maybe you might come back or maybe you might downsize again.

You don't know. That I think is what ties us back to the, to you. The idea of this preservation effect and, and keeping a certain sense of technical diversity. I mean, I think the bottom line is technical diversity. It's, it's only [01:21:00] so critical that you, you have it to a certain point. And I, I don't know what that point is.

Like, I can't say you should be X percent technically diverse in your company. There's no, of course, one number, but um, You should maintain a certain amount of technical diversity that allows you to adapt. And whatever that number is, whatever that amount is for the company, it requires all the other things we've talked about.

The last Point I wanted to make tonight was about measuring these things and these diversity health indicators, which is okay Let's suppose that there's not a magic number for how diverse a company should be technologically, but that there is a way for a company Itself to measure how how diverse it could be or how diverse it is and I had some Examples here one was technology replacement costs which we talked about costs.

I don't think Dollar values are a component of chaos [01:22:00] and or technical diversity. I think dollar values are a capital means of measuring, you know, money in, money out, just basic math, but There's, there is an element of complexity that comes to contractual obligations, or there's a measure of complexity that comes to negotiating the difference between a team's plan, a team plan versus a business plan versus an enterprise plan.

Like you have to do the, the, the math. Like if I go to a business plan, I get this function, but I really wish I had this function, which is the enterprise plan. Okay. So now I have to think about what happens if I go all the way past this business plan, which is actually the only one I need. Like, this becomes an issue.

So technology replacement cost, and I'm talking about the total cost, then integration complexity, and you can score that. So to go ahead and, you know, do this thing on a score of, say, one to ten, how complicated is it? And that complexity includes some probably subjective measure of how to [01:23:00] rate how much the organization would be impacted.

And I often think about this in the opposite direction, which is how much change management am I going to have to do to do this thing? If it's like a shit ton of change management, you know, starting three months out, tons of emails, trainings, the whole thing, that's my score. Like, it's get to 10. It's highly complex.

If the change management is, Hey, everybody. Uh, here's a Slack message. We're, we're moving to this tomorrow. It's probably a one, um, this, the skill coverage. So back to like the Zapiers and the air tables of the world, how many things does this potentially replace now and in the longterm, and then lastly, the vendor dependency index, uh, it's all I could call it, but it sounds like such a cool thing, but I really it's, uh, by doing this, did I just sell out?

You know, like [01:24:00] how, how locked into my, to this vendor. And there are some vendors, once you get to the enterprise platform level, that you got to sell out like ERP you're selling out. Like there's no, not that. So, I mean, if we think about those four metrics and I'm, I'd be curious to hear what you guys think about other metrics, but those are the four that when I think about sort of how to measure all this, that's where, that's where it begins.

Kevin Dushney: Yeah. I mean, I think if you scenario plan, the likely, you know, inflection or change points in your org, you could, you know, do an assessment on how hard would it be to quickly scale up or down, um, you know, and have some, have some plans for, you know, Hey, I have a person on my team that does this now, if I lose that person is, can I go to, you know, uh, uh, OPEX, right.

And spend some dollars with a consultant to. [01:25:00] backfill because I need that to keep that capability alive or do you just shut it down? So I think, I think you could do some, you know, some brainstorming or mind maps on, you know, what's the scenario? How likely is it? How entrenched are we? How complicated are my Systems and integrations and how hard would it be to maintain this if this happened and that's even a conversation you can have That you know bring up to like the enterprise risk management level.

Michael Crispin: Yeah. Yeah It's a business continuity, right? That's a piece of the pie and um, yeah, it's 

Kevin Dushney: or my Earlier about there's one developer If that guy quits, like, what are you doing? Like, I guess you could continue to run it for a little bit, as long as it's on, but now the development's dead and you'd have to go immediately find a different platform [01:26:00] and migrate all the data and, and train everyone.

I mean, it's an unlikely scenario, but also advice to not get in bed with. You know, an app that's written by one person. Yeah. 

Michael Crispin: Talked about the Jack of all trades type of person or, um, skill set, right. That's so important. It's almost making sure you've got platforms and tools that, uh, also fill the same void.

So like when you're going through, unfortunately going through a riff, a lot of the times you find that the people. That you need to keep the closest to the ones that can do a lot of different things, even if they don't know how to do them. 

Kevin Dushney: Yeah. 

Michael Crispin: And that you have a strategy in which to shift certain people into place.

Um, if something happens and kind of your org plan and similarly with technology, it's the same thing. It's if there's, you've got [01:27:00] fault tolerance around capabilities within the platform and, you know, that technology cost, you can focus on building that into, I'm actually gonna be able to erase this cost or I'm gonna spend for one year, I'm gonna spend this much money to bring things back.

To this level, uh, to, to a level on this, this platform, the consolidated platform, without reducing the autonomy that people have so that we can ultimately create a solution that works where they can still build and get their work done. Um, but measuring that is more just about, I, I think a lot of us are, we're, I already talked about doing it, not just the data map, but the.

The, your, your, your, your portfolio of systems, your, your, your mm-hmm IT strategy. A lot of this is baked in. May not be, you may not be scoring each one of the systems. Certainly. Um, but I think that's, that's, that's an important thing. And you [01:28:00] talked about the complexity score, right? Um, and that's, you know, the, the custom integrations.

I, I think the fewer of those you have, the better. So obviously if you've got automation integration, data transformation type tools, those are the ones that you. Probably want to score the highest, um, because you had to customize them. You've had to configure them heavily. If you default to more of an out of the box approach, it's just going to save you.

Uh, time and time again, because through all the change and churn and people in your organization, uh, across the company out of the box, always stands, uh, configuration, customization kills, and that helps drive that integration complexity score. Um, and then you can almost treat it like a risk 

Kevin Dushney: assessment, right?

If you're a lab company, you know, as an example, we've built so many. Data [01:29:00] workflows in Amazon, right? Some of them clusters, super complicated, and they're reasonably well doc, you know, documented. But for someone to just sit down and take that over, that's not a trivial matter. And, um, you know, but so you could rank that as a, as a red.

For example, our AWS environment is very complicated. It's well run, but it's complex. Yes. And it would be very difficult to cross train or transition that over to, you know, uh, even a consulting firm would take time. In the meantime, you might be down, or if something breaks, we don't know how to fix it. You know, so I, that could be another approach is, is looking at it through the lens of the, again, the inventory and then rank.

How difficult would it be to You know, fall back to your point, Mike, to a backup platform that's simpler. Lift and shift to something else or outsource or cross train people [01:30:00] internally. I mean, there's a number of levers you could pull on, but. You know, you can at least sort of map out where the high risk areas from a risk standpoint.

Michael Crispin: That skills and vendor sort of dependency coverage piece, right? They go together. You're talking about AWS, you might have a vendor or a specific internal employee or employees that are managing all that. 

Kevin Dushney: Yeah. 

Michael Crispin: All in all, the sort of approach that you're laying out in terms of the four areas, just to map that out, makes you think, makes you think about where you're, where you're going to be, to your point, Kevin, from a risk management perspective, which is really an enterprise risk.

Um, 

Nate McBride: yes, 

Michael Crispin: management tell you a risk management exercise because not only is it technology integration, complexity, skill and vendor, just looking at the list here is that's true of almost every department and across the whole [01:31:00] business. Not just your stack. And because more and more technologies are embedded in the business, we're integrating with partners, um, with CROs, CMOs, potentially, um, since it's just of enterprise risk model in a lot of ways, I think it gets you to think in that context.

Nate McBride: Yeah, it's a component of it, I think, so if we think back to episode four, where Mike and I spent a considerable amount of time talking about value, and now it was, it was, it was in the context of the value of identity, but we got into the same discussion. I mean, there's, there's a parallel thread here. I mean, this whole point is we're talking about autonomy and, um, How do you measure something?

Like, how do you provide any kind of quantifiable measure to anything that's in technology? And here we're talking about, I mean, we, we can't help ourselves when we keep thinking about technology in terms of a value term. So how much does it cost to do something? How much does it cost not to [01:32:00] do it? Um, what's the impact to the organizational, the organization's resources if we do this thing or don't do this thing?

I mean, we're always putting things in, in terms of these, um, These value based concepts. And I just, I just think it's fascinating that even now we're talking about metrics, um, you know, I mentioned, I mentioned a metric that had to do with value, but we're talking still, like if we are able, if we're going to try and create technical diversity.

and prevent chaos or minimize chaos. We do have to have a very, very good idea of the value of everything that we have. Like, there's no question about it. I don't think you can go into a situation where you're gonna have any kind of chaos management and not know how much every single thing is worth against every other thing that you have.

Because I can't say to anybody, Oh, this thing is worth less than that, and therefore I can remove it. Without having already measured that thing or have [01:33:00] some sort of rubric of value around that thing. Um, it's a big, it's a big meaty thing. I mean, we talked about this at the beginning of the episode, but honestly, to, to try and create technical diversity without chaos, you have to have a lot of components in place.

For sure. You have to have an ability, as Mike has stated earlier, to see down the middle of the road and, you know, has, as Kevin has come back to multiple times, like we're, we have to be able to approach this from a, how many other people are going to be impacted by this idea? How many, how much does this transform the organization to do a thing or not do a thing?

These are not bad ideas, but How many of them answer the question of, well, will this create more chaos or reduce it, at least in the way that we're talking about it? 

Michael Crispin: One of the, I think, I think [01:34:00] I'd add is just a little bit out of bounds of what we're talking about is just to take even a more minimalist approach on some things.

If you're with a small company, you know, it's It's not that you're reducing the diversity or the autonomy. In some ways you're enabling it. If you, if you pick a few core platforms and you try to work within the boundaries of the tools that you're picking and to try and keep things as simple as possible until those triggers you're planning for show themselves.

Um, I just think there's Certainly me as a technologist, you get excited about things that, that come along and you want, you see through them and think about how they can provide this. Like you said, there's a value and this, uh, excitement that they could, whereas I think a lot of times in a small company that people are just would love you to tell [01:35:00] them what they need to do almost in some respects when they need to do it.

And they really don't care if it's the best tool. They just want to know the process. Yes. And if, and know where to go to look for it and if we can minimize the, and they have no autonomy because they're following a process and they're doing it the way you're kind of prescribing, right? But I think there's a certain workforce right now that's like they've got, they're so stressed out about their deliverables or the pace of the business or their lack of resources that the, the need for them to.

Be able to think outside the box and change the way they work is perhaps outside of their priority. And that's like it taking in a mini minimalist approach on some of these things, you know, in a, in, in a smaller context, you know, in terms of being a smaller company that's in phase one or even phase two is, uh, certainly where I am right now is, is taking that approach to [01:36:00] almost a less is more approach and focusing on the relationships and focusing on the trust and focusing on the results.

Um. So I'm trying to take that approach, but at the same time keeping in mind that as the company grows, there's going to be a need for more flexibility and outreach and autonomy. And you want to build, you 

Nate McBride: want, you want to build as an I. T. leader. I mean, especially where your realm is, is understood. You want to be as, as in front of that as possible.

Michael Crispin: Yes. And like you said, those event triggers are a big part of your strategy and your, and your long term planning. Um, and even though it's not a super light formalized, uh, I once was working on a poster and I'm looking at, I'm like, I'm going to put this in my bedroom wall, not my office wall, because it's a great work of art, uh, what I've created, but I don't think anyone else is going to, is going to, is going to [01:37:00] look at this.

I have an awesome one I just created. I think it's so cool. But it's, it's like, it's always changing. Then it's like, Oh my God, I'm going to edit that box. I'm going to change that box. And, um, but it's not so much the artifact that you create. It's the exercise that you go through to discover and learn.

That's the most important thing about those artifacts for me, like you talked about risk, risk exercise, uh, Kevin, like you talked about going through risk, risk analysis and, and your AI discovery, right? Your data, data governance, discovery, and inventory process. The paper that we write up at the end is, is good to have, so we can refer back to it.

But what we all learned through that, including our peers, uh, I think it's what you apply to the next, the next chapter and we can get, it's, it's awesome to be able to do these things and, you know, talk, talk about applying them and we're, I think we're making a lot of progress, you know, as, uh, all three of us, you know, as we're doing a lot of [01:38:00] these, especially having to come up aligned against AI.

It's making us think outside the box and why we, I think because we're leaders, we have that autonomy. You want to be able to give that also to our direct reports and to our peers in the business. That's trying to find that middle world balance. Like that. I, I, the minimum, minimal approach, you know, there's more and more goes on.

I'm like, God, I wish I just lessons learned from the past is keep it simple. 

Kevin Dushney: That's the ideal. Very easy to overcomplicate, especially.

Michael Crispin: I can give you 1 other example, Nate, you mentioned about, um, what was, what was we were talking about? Oh, the, the, um, moving from not just from a reduction in force perspective, but from a needing to move to another system is if a system you've implemented. Fails. And you've got an air table or you've [01:39:00] got a, uh, you know, we can use SharePoint or other configurable tools, right, that have good configuration capabilities underneath to literally save the day.

You know, we're in a situation right now where we've invested in a product and we very well may just shift it over to an air table. Because it will get us past the next 6 to 12 months. It'll work great. And hey, maybe it grows to be something bigger, but we made an investment in a tool that has the flexibility to absorb any sort of risk that we have, maybe making an investment, uh, in, in a tool for purpose tool.

So having that is purposeful. I was just going to say that 

Kevin Dushney: you're going in wide open versus. You know reacting and then you've got to scramble to do it. You thought about this the time you said well This may not be our forever tool, but we're going to start here and reevaluate in 12 months We might throw it away.

We might 

Michael Crispin: keep 

Kevin Dushney: it. 

Michael Crispin: [01:40:00] Yeah How do we keep going without having to stop?

Nate McBride: You create How do we just take it? 

Michael Crispin: Just take it. 

Nate McBride: You create technical diversity without chaos. There's your answer right there. There it is. That's how you keep going. Well, um That was awesome. I don't think we're going to do an archetype tonight, because I just think we want to end on a high note. Not go into the hoarder mentality of the archetype I had, so we'll save that for next week, but um, I just want to remind everybody that if you like our show, please give us all the stars on all the things.

Um, and don't forget, links to our description. Uh, we have links in our description to all the things that we do. Donate to Wikimedia. If you've got an extra couple of bucks, donate to the ACLU. Also don't be a dick, uh, to people and especially don't be a dick to the hardworking IT folks in your company. Be cool, get paid back in spades.

Be nice to animals, have them spayed or neutered and [01:41:00] be nice to old people. Um, they were once like us and, uh, which I mean, we're old. So that's why we get it. Uh, any final thoughts, gentlemen? Well, first of all, Kevin, thank you so much for coming on. 

Kevin Dushney: Yeah, no, my pleasure. Thanks for having me. I hope that you will come back 

Nate McBride: again this season.

Kevin Dushney: Absolutely. I always enjoy these, so. 

Michael Crispin: I feel like we need to love burritos. 

Nate McBride: Yes. 

Michael Crispin: Gotta do it up. 

Nate McBride: Once my yard is defrosted enough that you can walk on it without needing snowshoes, I know. I'm gonna have you guys both back out to the barn. 

Kevin Dushney: I need to do one of the, I need to do one of the on the scene ones to like the next, like, you know, bio I.

T. world, bio I. T. world. Yeah. 

Nate McBride: And then we should, we should go to that place, Mike, and that where we had that little corner table. What was it called? Patio Mc Flanagans or something. Oh, yeah, 

Michael Crispin: yeah, yeah. Patty's in Newton. 

Nate McBride: That's a cool place. 

Michael Crispin: That's [01:42:00] an awesome place for this. But we need 

Nate McBride: better lapel mics.

'cause I think we sounded like shit last time we went there. 

Michael Crispin: We did a, we need to, uh, get a production company in there and the whole thing with lights and everything. 

Nate McBride: We fired 'em all. I mean, we can't, we we, that's 

Michael Crispin: right. We just had to, it's like the those 

Nate McBride: squared YouTuber mics. I see everyone wearing 

Kevin Dushney: that pretty well.

Yeah. 

Nate McBride: Alright, we're gonna, we'll invest some money this season, we haven't really spent a whole lot because you guys haven't been coming for dinners and drinks. And so, the budget's, the budget's looking pretty good. We're only, like, maybe 7, 000 in the hole. Um, I mean, I think we got a few more dollars we can spend.

Michael Crispin: I think we should use all the money to fly down to the Dolphin and do a broadcast for the bar. 

Nate McBride: Well, I think we're supposed to. All 

Michael Crispin: these families are there, going to Epcot Center. That would be fantastic. I'll sit around drinking pitchers of margaritas. 

Nate McBride: No, we got to go to that pool bar that's got the hundred ounce yard, uh, beers.

Um, [01:43:00] whatever that place was called. That was like the brain eraser bar or something. I forget. Um, yeah, no, we, we, we, we got to go on location, Mike, before by YT world. We got to get our road chops back. It's been a while. 

Kevin Dushney: Yeah, 

Nate McBride: because that's what in April, 

Kevin Dushney: right? 

Nate McBride: Yeah. Yeah, we got to sharpen things up a little bit.

Get our acts back together. That's going to be great. What's that? I know I'm working on a panel. I'm working on it. I, I feel pretty confident it's going to happen and the vendor is who's going to do it for us. I think you, you know, but, um, I think they're going to try and like make it so that we talk nice about the future of their products.

And maybe we'll just give them like a one second plug and then we'll get into the real discussion. Maybe we'll see how it goes. I mean, once we got the mics in front of us, we're on leash as far as I'm concerned. Um. Anyway, any closing thoughts on this? Next week, by the way, well, before you say that, next week, just so we're all on the same level, [01:44:00] Kevin, you're welcome to come back any episode, by the way.

Yes. Next week, we're, we're establishing clear boundaries and principles, which is to have a path, stay on the path, even if you're, you have overwhelming zero choices. This is going to be a good one, because there's nothing I love more than establishing clear boundaries and principles and then throwing them all away to get my, get my, my thing done.

So this will be a good one. 

Michael Crispin: Establishing boundaries. 

Nate McBride: Establishing boundaries. Having a safe word? Temporary boundaries. Temporary boundaries. Well, IT leaders I think should feel inclined to throw away any boundary or guideline they set up for other people when it best suits them. Um, and then reestablish it after they've accomplished their goal.

So. Or if it's 

Kevin Dushney: impeding progress, maybe it changes 

Nate McBride: it. Or if it's impeding progress, just simply modify [01:45:00] the policy temporarily. Exactly. Uh, so any final thoughts?

Kevin Dushney: I don't think so looking forward to continuing the conversation, but I thought that this is pretty good 

Nate McBride: This is good. We left a lot on the table as we usually do We we created more questions than we answered, but you know what? I think it was awesome. And I think honestly Uh, there's a certain amount of chaos that's needed in order to do our jobs and we, we all kind of like are okay with a certain amount and there's no way to have, it would be boring without it, but the more that we can do and all the possible things you need an army of people to get them all done to create technical diversity in the truest sense, but we can at least all approach it from our own little angles and.

Do what we can do. And Mike, I got two jorts down that we're gonna have to do. We're gonna have to do a datamap jort. We'll have to do a measuring value jort, because I just, I just decided we're gonna do that. Um, we have to come back to measuring value again, [01:46:00] because I think that between what we talked about in episode four and what we talked about tonight, there is a whole nother measuring value thing that we have to do.

And I got some notes. 

Michael Crispin: Okay. Yeah, let's do it. 

Nate McBride: All right. Cool, cool. Great 

Michael Crispin: discussion tonight. A lot of good points. Good topic. 

Nate McBride: Yeah, I'm gonna, I'm gonna just do this, we're gonna fade away into the music now. 

Michael Crispin: I love it. So right now the 

Nate McBride: music is gonna start playing in the background. Alright guys. Night. 

Kevin Dushney: See that there's a third AI bot here.

Somehow it's a, yeah, we're 

Nate McBride: gonna, well, we'll find out tomorrow morning an album. Go of two people into one bot . I wonder if it will do that. 

Kevin Dushney: I don't know. . We'll find out. 

Nate McBride: All right, we'll find out tomorrow. See, it does. All right guys. Have a good night guys. 

Trance: Night flashing screens that glow so bright in the matrix.

We take flight holding truth within our side. [01:47:00] Calculus of IT, we value autonomy. The calculus of IT, we value autonomy. Through the cyber paths we

glide. In the circuits we confide. No restraints, no need to hide. In the system, we reside.[01:48:00] 

Leave our fate, in the data seas we skate. Zeroes, ones that can't abate. We control it, it's innate. Binary whispers in the night. Flashing screens that glow so bright. In the matrix [01:49:00] we take flight. Leave. Holding truth within our soul. The calculus of I. T. Revalued on me. The calculus of I. T. Revalued on 

me.