Andreas Welsch: 0:30

Today, we'll talk about scaling your AI programs via citizen development. And who better to talk about it than someone who's actually written a new book about just that. Ian Barkin. Hey Ian, thank you so much for joining.

Ian Barkin: 0:42

Andreas, thanks for having me. Awesome new intro, by the way. That was very cool. Thanks for letting me be part of that whole thing. And in a thrill to be here. I love the show. I love the work you've done over the last few years in educating our community. You are clearly an expert yourself and it's fun to be here. Happy 2025.

Andreas Welsch: 0:58

Happy 2025. I couldn't be more excited to have you on the show. And we've talked about this for a couple of months actually getting this lined up. For those of you who don't know you in the audience, maybe you can tell everybody a little bit about yourself, who you are and what you do.

Ian Barkin: 1:12

Yeah, my pleasure. I am a combination of things. I'm a BPO and an RPA guy. And so what that meant was the beginning of my career, I was helping enterprises tap into global talent, finding really good, highly skilled people all over the world. And about 12 years ago, that was luckily the place where we started to play with automation to augment and emulate some of the work that those people were doing. And so that sent me on a my second sort of career journey, which was around automation and how do you digitize tasks in enterprises, front, middle, and back office? It resulted in launching an RPA consultancy about 11 years ago. Awesome journey, so much fun, built a great family, had a team all over the world, worked with amazing clients and learned a ton. And and did that for a few years, sold it. And since then have been following random passions. I invest in AI startups and help advise them, hopefully with some skills that I've picked up along the way. I educate just as you do. I've got LinkedIn Learning courses, just as you do. I've written some books just as you have, and I just love being part of this community. Hopefully sharing useful information for folks who were starting their journey, who are expanding their journey and looking how to tap into what has been an amazing set of capabilities. That's just so much more than those few tools we started playing with 11, 12 years ago. It's a, it is a robust and sometimes very confusing toolkit we have available to us today.

Andreas Welsch: 2:49

Hey I must tell you, you've been a great inspiration for me. So with all the"just as you do", you actually went first and you've inspired me to follow that path.

Ian Barkin: 2:59

I appreciate that.

Andreas Welsch: 3:00

LinkedIn Learning courses, the Creator Accelerator Program, and just seeing what you do for the community and how you share your knowledge as well. It's fun stuff.

Ian Barkin: 3:08

It is fun stuff. And it's and there are skills that you collected through your professional career that are so invaluable and that you take the time yourself to share them in a way that's not just creating and influencing for sake of being a creator or influencer, but because you actually know it, you actually understand it, resonates in everything you do. And I'm thrilled if I could have been even the smallest of inspirations to you on that journey.

Andreas Welsch: 3:32

Absolutely, yeah. Now you've been on the show before, together with a few other guests, but this time, I got a surprise for you. The same surprise that I got for everybody. Okay. And so the little icebreaker, when I hit this buzzer over here you'll see a sentence and I'd like you to answer with the first thing that comes to mind and why, and to make it a little more interesting, you have 60 seconds for your answer. Probably, you've seen this by now.

Ian Barkin: 3:58

Because it's a speed round for slow people.

Andreas Welsch: 4:01

Are you ready for What's the BUZZ?

Ian Barkin: 4:03

I am. Let's do this.

Andreas Welsch: 4:05

Okay let's do this. If AI were a sports car, what would it be? 60 seconds on the clock.

Ian Barkin: 4:14

The Tesla Coupe. The small little sporty one that has yet to come back out again, but it would be that.

Andreas Welsch: 4:20

And why?

Ian Barkin: 4:21

Cause it's wicked fast. It is a lot of promise doesn't necessarily exist, yet. But a lot of the mechanics and the elements of it do exist. And now we're just we're waiting to see it. That sounds very negative about AI, but it is here in pieces and parts.

Andreas Welsch: 4:36

Wonderful. Sounds a little bit like a Tesla advertisement.

Ian Barkin: 4:39

Yeah, exactly. I do drive one, although I'm looking at various other things now. There's some cool new vehicles out there.

Andreas Welsch: 4:46

A little eco friendliness in the AI space would go a long way too.

Ian Barkin: 4:51

That's true. That's an interesting point. Yes. The Tesla doesn't consume all the energy that LLMs do.

Andreas Welsch: 4:56

Not the one of an entire power plant.

Ian Barkin: 5:00

That's right. Okay. How did I do that was is that the only one because that was faster than 60 seconds now we got a vamp and fill time.

Andreas Welsch: 5:07

So definitely 60 seconds is over and it was a wonderful answer. We're obviously not just here to talk about random questions and coming up with answers. This topic of citizen development is actually near and dear to my heart because I've spent a lot of time in IT and even 10, 15 years back, getting the technology and what it can do, as I mentioned in your intro, there's just been so much tech being layered on top of each other from RPA to machine learning, generative AI, yes, there are all these possibilities. And, about two years ago, we were saying you don't have to be an AI expert to work in AI. That's what your background is plus AI, and I think that's still true, but if I look at my feed, it seems that's truer than ever with everybody all of a sudden jumping on the bandwagon. But I'm curious, how are you seeing this play out in business, this whole topic of citizen development and what should people know about this going into it?

Ian Barkin: 6:07

Yeah so the reason I embarked on the research that went into the book with my coauthor, Tom Davenport, who is a legend in many spaces and has written 25 books. And so that was just a joy to be able to do that project with him. But the reason we embarked on it is because, as we said, 11, 12 years ago, there was this concept of automation that could emulate the work that people was doing. And so we embarked on that. We took a look at it. It was working. But it still required some rigor. It wasn't easy for a business analyst couldn't do it. And yet the narrative in the space was often, it's so easy, a business analyst can do it. And there was a lot of narrative around just having everyone have their own robot. And that was, from a marketing and a sales perspective, totally get it. Understood why that was the narrative, but from an execution perspective, it was harder than that. But then progressively over time, and you mentioned there were layers of higher complexity with intelligent document processing and machine learning and some other things, and process mining and discovery, etc. There became, there was still a commitment to that narrative of making it easy enough that a broader audience and a broader user base can actually capitalize on this set of capabilities. And so what happened is we started to use, or vendors started to use these terms low code and no code. And I thought that was funny because ultimately it's weird to name a software trend after the complexity of the tool. I joke it's like calling a a stick shift car harder to drive and a manual car easy drive. And like you can buy a harder drive or an easier drive car, what do you want? And so lower code or no code really was a way that we were classifying technology that required less coding, there's more visual, more building with puzzle pieces, drag and drop, etc. Unfortunately, even then, it still wasn't so easy that a business analyst could do it. But, in the course of our research, and actually, after we went to press, this stuff is getting easier and easier. I'm ultimately prompting, which is the ultimate in no coding, where you're just using human language to articulate what it is that you have dreamed up, that you'd to see turned into something. Sometimes it's an email, sometimes it's a poem, but sometimes it's code and applications and et cetera. That is happening now at a rate that we've never seen before. And so this low and no code, whether we use those terms or not are truly becoming possible. And so what that meant was domain experts, the people who understand the scope of the work that they do. If you're an HR and you understand employee onboarding, if you're in finance and accounting, and you are an expert in accounts payable or receivable or whatever else, those domain experts now more so than ever before in the history of work and the history of work with technology can turn those domain expertise, those skills that they have and the needs that they know that they have to get their job done more efficiently. They can start turning those into applications, integrations, and data models. And so that's what I think is so exciting about citizen development is there have always been citizens developing for them has been hard. It's getting easier and easier. And there's two reasons why that's exciting. One, because digital transformation requires a huge number of people to contribute, and there just aren't enough IT professionals to do that. And two, IT and business have always been an oil water dynamic that don't understand each other, don't really work well together. Not to say we don't, but if you have the actual end user more intimately involved in designing and executing that which they dream up and need, man, this is a renaissance of digitization across every enterprise of all sizes in every department, every nook and cranny. Citizen development for the win. That was not a short synopsis, but I'm clearly passionate about this stuff.

Andreas Welsch: 10:20

That's wonderful. I read a post by Shail Khiyara if you're leading an AI CoE, some kind of an innovation CoE becoming to bottleneck again as well, and I was so reminded of IT, because when I talk to IT decision makers, they tell me, yeah, there are all these things that we need to do and we have backlog of things. And even in my corporate career in IT, you said, yeah you can come back next year, maybe in six months, because we're really busy with other things. So I do see that promise. And I do see that value of citizen development with certain guardrails, certainly empowering your business, empowering your users to do the things that are important. Easy enough, scalable enough, yet less your security and your integrity of the business.

Ian Barkin: 11:05

Totally agree. By the way, it was that IT roadmap, right? Necessities, mother of invention. It was bottleneck was the mother of invention in the RPA days. IT just could not get to projects that business users found useful and valuable. And so they found a, an outlet, they found a way to work around IT. And that was RPA initially, and it's only getting easier. It's only getting more powerful. We assert in the book strongly a need for governance, to your point, and B, a need for collaboration because otherwise this becomes the grey IT rogue digital initiatives shadow that you do not want in any enterprise ever.

Andreas Welsch: 12:09

And by the way, it's great to see so many folks joining. Ramnath and Douglas and Stephen. And Martin. So thank you for being with us. If you have any questions, please put them in the chat as well. And we'll pick them up in a couple of minutes. I do want to make sure that this continues to be an interactive show. That's why we are live. So please keep your questions. And easy. All of that, and it's. It's Tuesday, just afternoon in the U. S., so keep them easy, keep them light. But yeah, as we're already saying that there's so much tech in a business layered on untapped potential. So why do we talk about citizen development? Why do we need to say, in addition to bottlenecks, what are things that domain experts in their respective business functions can all of a sudden do, or what do they need to do?

Ian Barkin: 12:56

Yeah. So I think there are tiers and levels that, there's that which is possible and that's, which is reasonable. And again, in the book we might touch on this later, but we've got several different frameworks and structures that we major on both because we saw it happening within the enterprises we interviewed. And also we knew it should happen in any enterprise that wants to embark on this. But back to the fundamental point, there's more work than there are IT professionals to to do that work. There is a greater level of expectation and marketing around the potential for, let's call it AI. And a lot under that umbrella, but a lot of people are hearing a lot of things about it, especially in this era of AI agents, where it seems like that is just, All of the news, all of the story, all of the update, all of the product release. And every single major enterprise now has agent as the concept that they're organizing around. The lens through which we look at the work that we do. is heavily AI ified and agentified and we want to see it used. And so that automatically creates a massive supply demand gap. And there are, which I think is a good thing in so much as, In the old days, you'd do sort of continuous improvement exercises and Kaizen and Lean exercises where you'd study Pareto's and whatever, I'm showing sort of my background, but you would identify areas of opportunity. This sort of is a shot in the arm for that initiative. Everybody almost can't help but look at their work through a lens of what can automation do for me. Which is really exciting. And the other thing that we found in the book and in the interviews that we had was the level of digital quotient, just the shared vocabulary is such now that a business person and an IT person that find themselves in a collaboration session or in a meeting room of any sort are often. Much closer together than they ever have been. They're speaking similar languages. They understand constraints. They're discussing the art of the possible. They know what the existing enterprise architecture enables or constraints, and I think that's been a huge asset just to evolve that discussion to a point where we can be brainstorming what is possible now, it still comes back to, do you have the capabilities? Do you have the guardrails and governance internally? And do you have the right tools to enable that? And that's up to an enterprise and often leadership and IT to decide on how to approach that.

Andreas Welsch: 15:41

Are you seeing that business is getting more tech savvy and literate or IT is getting more business? What did you find in it?

Ian Barkin: 15:51

It's a great question. I think it's more the former than the latter. I believe that business is becoming more tech savvy just because of, and this is not the last few weeks. This is the last decade that they're becoming more tech savvy. But I also think it's it is. As a necessity, it is requiring that IT become more business savvy because they are having those discussions with business users. And they're, they are, IT's role is to support an enterprise, right? Every company is now a software company, but the company first and foremost is a product company or a service company. They have customers, they have strategy, and supercharge that. And we argue in the book that IT has to find a way to do two things which is never easy, but one of them is the traditional enterprise IT role. Of making sure all of the traditional systems, the ERP systems and CRM systems, and now more and more the security and compliance and regulatory components of an enterprise are sound and stable and safe. They also have to build in this sort of guidance element, which is one of the frameworks we touch on which is around educating and nurturing and shepherding a lot of efforts that will happen whether they want them to or not. You don't have a choice, it's happening, you do have a choice about how you respond to it and how you enable it. You're never stopping it, I would argue. You can try, but good luck.

Andreas Welsch: 17:20

That's a big realization for technology driven departments. It's happening and you get to decide, how much you will be involved and how much control you can still exercise. If at all, right? It's all about guardrails not so much the stopping.

Ian Barkin: 17:37

The framework in the book is a 4G framework and the G's, the traditional G's everyone talks about and constantly brings up are governance and guardrails. And we added two more. One of them was guidance is that concept of it being and not just IT, but other elements of the organization, educating, supporting, sponsoring, coaching, et cetera. But the first G, so we added a first G, which is Genesis. And the purpose of that was, Genesis is the beginning of something. And the purpose of that G was ultimately to explore that concept that if you think You're spearheading this initiative in your organization. Quite frankly, you're wrong. It is already there. It is already started. So you are not starting this. The genesis of the the innovation and the pension to DIY or to make or to build or to iterate has always been in your best people. And they did it with Excel macros beforehand. They have more powerful versions of tools to do that in today. And so you just have to recognize that is happening everywhere in your organization. Hopefully, genuinely, hopefully you've got people who are. Who are those creators and builders and makers who who think big thoughts and solve big problems and try to make their operations more efficient. Now it's up to you to decide how you enable them.

Andreas Welsch: 19:02

So I'm curious, building on this and what you said earlier around exercising some kind of control or guardrails coming back to your 4Gs. And I see doug is in the audience today, Doug Shannon, and he said, Hey, how do you see the potential rise of shadow AI code coming into play and how do we verify that the code is safe, is sound is well written to me that kind of goes into the direction of how can you have some some control over this citizen development and how do you make sure that what citizen developers build is safe?

Ian Barkin: 19:33

So there's a lot of questions in there. Yes, so ultimately a simple answer is you need to have some form of platform, right? You need to have some set of tools that are vetted and approved by the enterprise that your citizens can work in. Because the last thing you want is, bring it from home. Just chaos where everybody finds their ends. And by the way, in the old days, there were, there was a small number of tools that they might be able to sneak into the enterprise and start playing with now. It seems like that list is infinite. So this gets harder. But you need to enable them with the tools and that they, the sort of the building. Set that they can use and you saw that you see that to some degree with things like copilot in the Microsoft platform that most enterprises have. There are others that they may have already had in house. And so some of the RPA platforms are well positioned. And if you've got UiPath wall to ceiling then that's a place where you start to do the brainstorming and look for ways to to digitize more broadly. So those are the sorts of platforms you need to lean on because you don't want to enable just a pure bring your own device or bring your own. Automation sort of mentality. And then to the point of how do you prevent shadow? Again, it's around those guardrails of you can't build anything. And in the book we discuss actually levels of remit. So there's those things that matter to me to oversimplify it. And then there's those things that touch or matter to we. And so if it's a me thing, it is an application that helps me do my job. It's probably on my desktop or is very localized. Anything that starts to branch into the we, which could be any sort of crossing the. boundaries into a different department or a client or a vendor or a partner or something, those start to become ones you want to govern more strongly and keep an eye on and, or bring in IT to support. Cause I might have a great idea. Like I am in accounts payable and I have just a really strong idea for how to make it better for everyone. But if it starts to get complex, I'm not the person to be developing that. So you bring in it to shepherd and to collaborate and to support those sorts of efforts or to take them over all together and to build them because it's a good idea. And then the maintenance thing too. The last thing you want is to have necessarily citizens maintaining their own and then, and some enterprises do go that way where you have to make sure that you're updating it every once in a while, or they kill it, you have to maintain it or it goes away. But in the same old sort of logic of in the old days, in like a mainframe, Bob would build a great tool, but. You kept Bob away from bus routes because you didn't want to lose him because he's the only one who knew how it works. You've got a, obviously a plan for that sort of situation as well. So in general, robust, mature citizen programs do involve IT for the governance the guard railing the remit elements, and then ongoing maintenance and support.

Andreas Welsch: 22:47

And I think it also partly answers Ramnath's question, who was asking, how far, to what extent should you let citizen developers build their things? What if it gets too broad, too many users? I think you, you nicely answered that there is some, some some good measurements that you need to say, this is what I let them, because it's me. This is what maybe we collaborate on together.

Ian Barkin: 23:10

Let's be clear. At no point in time are we suggesting that a citizen should replace SAP for you, right? We are not suggesting that citizens step in and retool your ERP systems and rewire how the enterprise is working. But there are, there's nearly endless list of opportunities. To transform and digitize and evolve every process within an enterprise that domain experts are closer to than it ever will be, and they can make efforts to improve them. And if it gets bigger than those very discreet solutions, then think, how lucky are we that we're able to have meaningful brainstorms with the end user and then be able to put it on a roadmap. Collaborate with them, figure out what really great looks like. Not only does it solve big problems, it also keeps your best people. And those are two wins in my book.

Andreas Welsch: 24:09

Awesome. Speaking of book, what's the book again where people can read all about this?

Ian Barkin: 24:15

Oh, that's man, you just asked me to plug my own.

Andreas Welsch: 24:18

This guy.

Ian Barkin: 24:19

All Hands on Tech. The AI Powered Citizen Revolution.

Andreas Welsch: 24:22

Awesome. So definitely check that out, folks. Now, Ian, we're getting close to the end of the show, and I was wondering if you can summarize the key three takeaways for our audience today, all about citizen development.

Ian Barkin: 24:31

Yeah. Again, I think one of the key takeaways is. This sort of ingenuity, this grassroots innovation is happening and you want it to be happening in every corner of your business. So then it's just up to you to determine how do you enable and shepherd it so it does it happens safely. It is a call to arms for IT to to explore how it can better collaborate with the business users because they really do have great ideas and they are constantly creatively trying to problem solve and build, but it makes a lot more sense for you to all work together. And I guess it's It's just a statement on the future, which is, as large language models and large action models and agentics, whatever we're calling it this week expand and their capabilities grow, this is a call to arms, as far, a call to action for constant and persistent education. Because it really is important that everyone in the organization know what these tools are capable of how they enable them and how they supercharge them. This is, I would argue, this should never be a discussion of, can you replace a human with a robot? This is more about, there's so much you don't get to that you can. If you're able to tap into these capabilities and I guess my, I'll add a fourth because darn it. Why not? The point, the purpose of all this isn't more applications, the purpose of this isn't more glue and integration between systems. The purpose of this is great, meaningful, valuable outcomes, and keep an eye on that, Because ultimately, that's the point of all of this, is to use the best tools available with the best thinking available to create the best outcomes, which create the best experiences that will enable the growth and success of the best businesses.

Andreas Welsch: 26:40

I love that. It all comes back to people empowering them, giving them the tools to their work and do meaningful work, not having to others to have time in their backlog to, to do this. And I think it's a super important and valuable message at this time, because I also see that we will, like you said, hear a lot more about agents and replacements and being able to do things that we haven't been able to do before in all of these great and scary ways. So giving people the tools to, help themselves, I think is an excellent call. I'm super excited that you've had the, or that you've taken the time to, to spend it with us today and talk more about.

Ian Barkin: 27:22

My pleasure. Yeah. Yeah, no, my pleasure. Anytime. I'll come on weekly if you'll have me, this is fun. Yep. If anybody is listening and wants to learn about agents, Andreas's courses on LinkedIn learning are bar none. You got to check them out because honestly, back to that raising a digital quotient and understanding what is actually part of this new vocabulary of the future of work. His courses are invaluable. Assign it to everybody in your company. You won't regret it.

Andreas Welsch: 27:51

Thank you. And likewise about intelligent automation and yours, you've been doing this for a couple of years. Good amount of time too. Wonderful folks. If you're still with us right now, join us next week for another episode of What's the BUZZ. We'll have Nico Bitzer, CEO of Bots and People, on and we'll talk about how you can upscale your workforce and raise their AI literacy.