Flow Is Necessary, But Not Sufficient

Hey everyone. Welcome back here to Techstrong TV. I want to introduce you to Phil Clark.

Phil is the former VP of engineering at Parchment and the founder and fractional leader of RYU Advisory and Media. He's going to explain all that to us in a minute. But for purposes today, another important thing, Phil, is Phil is a presenter at the upcoming Flowtopia, a virtual event, all things flow and value stream, and we're happy to have him in here to talk about it.

Phil, welcome to Techstrong TV. It's great to have you on here. Yeah, very happy to be here.

Thanks for the opportunity. Thank you. So Phil, I always like to share with our audience who are they listening to?

I gave them your titles and your background a little bit, but fill people in on who's Phil. Yeah, sure. I've said this many, many ways throughout my career.

But I started as a software engineer, so I'm one of those people who started as a software engineer in the late '90s. So over the last two and a half decades, I worked more into a player coach management role, which was very popular, the hands-on directors and leaders and managers, in the 2000s and 2010s. Most recently, I spent the latest part of my career, and I've worked for startups, very large global enterprises, so about eight years working with global enterprises, global teams, very large companies.

And then transitioning into finally was recruited after two years of a friend trying to get me over to a small Series C startup at the time that we ended up exiting it at $835 million valuation when we were purchased in 2024. So been through nine acquisitions where I was actively part of the diligence of all those acquisitions, and then we were finally acquired. And so that was a very exciting run to go from about a $5 million organization with 20 engineers, to over 100 engineers and about 150 million in revenue when we sold.

So it's quite a hyper-growth run. And with that, I helped when I came on board, when I transitioned over in about 2014, help them transition, from their monolith, from Waterfall to cloud, to Agile, Lean, DevOps, and a lot of the team structures and things like that. And in 2022, really late 2021, adopted value stream management to add on top of all of those practices that we're transitioning through.

I myself had to transform significantly as a leader. I was highly successful in the Waterfall, the ITIL eras, and the heavy practice, and the long tail practice. When it came into going into the digital transformation world and really helping our team, I like to say that we were about 70% successful at Parchment.

I look at us as a poster child for those that say Agile fails, self-managed, cross-functional teams fail. I disagree. I love to use us as a case study when I was there.

But, in 2021, 2022, I came across value stream management flow, and it really just was at the perfect timing because it helped visualize the work we were doing. Even though we had moved to Kanban from Waterfall, and we had our boards, it was just the right timing and natural fit. And then more recently, I was hyper-focused on efficiency and productivity of teams and outputs.

And in 2023, I started publishing articles, and I noticed 2024, I really shifted that focus to outcomes, which is a very popular topic right now. What is the value that your teams are bringing to the business? Can we tie the purpose of their work back to the impact that they're having on our customers and business?

And I can jump right into where I'm at today, which is just recently in 2026, I left that organization to pursue other avenues, and started an LLC called RYU Advisory and Media, and really that's to manage the publishing of my first book, which is coming out by the end of this month. It's two-year progress and that, man, that was a learning lesson all on its own. Yes, it is.

Yeah. I've played with it a few times and never gotten across the finish line. So I applaud you and good for you, man.

And yeah, so thank you for that. Very excited. I can't say I was excited about the process.

It was a strain on my wife and my family at the time, but very glad I did that. Wouldn't trade it for the world just for the experience. But I started this company to help with fractional leadership until I find my next role.

So given coming through the Waterfall era, and then all the struggles with change management, that's all life has been for 20 years, adopting DevOps and cloud and Agile and all that. I have never been more passionate about software delivery and helping teams successfully deliver software than I've ever been today. I love this kind of work.

It's highly challenging. There's been points and times where I probably could have had a heart attack, a lot of stress and things like that. But I ended up starting this organization so that I could do fractional leadership until I find that role.

I am looking again, I'll land hopefully by the end of the year back into a senior leadership role with another company that's more like a Series C to mid-market. I think, luckily I have the opportunity at this point in my career, to kind of avoid the global enterprise pressures that are there. Don't get me wrong, I think everybody should have the opportunity to lead at that level.

Again, these are experiences. I often say that DHL, where I worked, was my master's degree, Parchment was my PhD. These are experiences that you just can't, if you get the opportunity, don't fear ...

promotion and getting into new things. It will change the trajectory of your career. So that is me in a nutshell.

I love it. That was excellent, Phil, and good for you, man. Congratulations.

So I guess we should transition into Flowtopia. For those of you who've been following "The Bouncing Ball," we've been interviewing a bunch of the speakers from Flowtopia, including Helen Beal, the chief Flow flower, floweer of Flowtopia, and the event is a virtual event. It's live, I think, I want to say for 12 hours on June 24th.

Phil is of course one of-- We've probably interviewed six or so of the speakers at this point. Phil, I wanted to dive in on your particular topic, right? It's Flow is necessary but not sufficient: Why leadership choices determine whether flow and realization survive.

What exactly do you mean there? Well, this will be my third talk at Flowtopia. Previously, I served on the board for Value Stream Management Consortium, which was the prior organization to this.

And so this talk, you'll notice I always stay in the leadership track, because you will find unlimited opportunities to get the hands-on practical experience and tools that you need through Flowtopia, through all the different speakers. So I focus more on my experience of change management and leadership, right? What are the struggles and how does leadership, you hear grassroots, top-down, bottom-up.

I think both work, but ultimately I see failure after failure no matter how great your success is if leadership is not aligned, right? Because ultimately, I think in my book I call them the old, I'm the new dinosaur. We bring our biases from our previous successes, and it's very humbling.

And as I said, I had to transform as a leader to kind of relearn. And we're doing it now with AI again. We're no longer the experts.

We have to relearn what it means to operate, why do we want to operate in these areas, and not just check a box and say we're going to adopt agile, right? Why? What problem does it solve?

Does it really fit our need? So I will be talking about the leadership track, and flow is necessary because delivering software and team efficiency is just what organizations need, they want to see, right? So it's the classic productivity of software engineers who are highly expensive and even AI, right?

Tokens are expensive. What are we getting for our money in teams? So I focused the last four or five years on flow engineering and the efficiency of teams.

But it's not sufficient. Realization I've talked a lot about you have to have flow plus realization. Other words you'll hear in the industry called outcomes, right?

What is the value? Why are we working on it? If a team can go from producing five widgets to eight widgets during the same period with the same people and become more efficient, with output did we increase productivity?

Well, if it added no value to the customer, we overproduced. We actually wasted productivity. So I really heavily started tying to where all of our teams had to start documenting before they started work their anticipated outcome, right?

So they understood the purpose, what is the goal, and then at the end, close the loop, right? That's the difficult part. " Because sometimes you can get that information within weeks.

Sometimes it might take six months or longer, right, to actually get the impact of that. Plus then there's multiple departments that were part of that thing like that. So basically, but if leadership, my talk is a simple idea.

Work has to move, but it also has to matter. So agile, DevOps, value stream management, and now AI can improve flow, but leadership determines whether that flow becomes real business value. Yeah.

I hear you, Phil, and I realize it myself, especially as you say, as we enter this AI stage now, right? The fact of the matter, even here at TextShroom, we're going through an experimental stage with agentic AI. I've told everyone, I encourage the experimentation.

I encourage you using it. The only thing I ask is every week, give me a little report, what you're doing, what worked, what didn't work, and we'll figure it out. But I think that's a necessary step you got to go through before you get to the position of saying, "Okay, this works, or this doesn't work.

This is what we should do going forward. " And here at TextShroom, for instance, we're making these decisions every day based upon what our people are doing and seeing and what they're able to output and does it work for us, right? Does it move the needle?

It doesn't have to necessarily bring in more revenue. Does it save us money? Does it make something better that's meaningful?

And I think that this whole AI thing is really forcing us to kind of go through that experimental phase again, right? Where we're not quite up to the bottom line, if you will. But yet we're also hearing now, my God, look at the tokens they're using.

Look how much money this is costing. It's cheaper to have a human do it. It may be those are the lessons we're learning that, yes, this does pay to have a human do it.

No, this doesn't pay, this is not worthy of using an AI model on. Maybe I need something cheaper. I don't have to use the best model.

I can use a lower, slower model. These are all things I learn on the way to the market, if you will. And we've done it with other techno-- We did it with cloud.

We went through this, right? Hey, maybe this doesn't pay to put up in the cloud. Maybe we should keep it in the data center here.

So, it's not without precedent. This follows the kind of rules that we've seen before, right? Just a new technology.

Yeah, 100%. I think you hit the nail on the head, and I think we are lucky right now, because the age of AI currently is being subsidized, right? While token costs are being talked about and things like that- We're not really playing ...

we also have this huge opportunity to do exactly what you said. I love when people like yourself or my teammates used to come up to me and I said, "Well, how do you know that? Now, how do you know it's not working?

" It's because you're in the experimental phase. Mm-hmm. You're starting to use it, right?

So start learning and so, push really hard. There was a big push with colleagues, I do a lot of pro bono work and conversation in the industry, and it was AI first. We are AI first, right?

What does that mean? That was 2025, and we're going to be an AI first organization. We're going to adopt AI, as to where we took a different approach.

I just said we will be AI literate by 2026. " Or they'd come back and say, this is pre the newer frontier models, right? " I'm not making, but, "We're adopting these tools and they don't work as well as they say.

" And they said, "Because here's what I did," right? And they go through their retrospective with me, and I said, "That's beautiful. You're becoming AI literate," right?

So you don't have to listen to the hype. So I also think it's just right now is that period where unfortunately it's starting to change, but it's heavily subsidized, right? Tokens aren't going to cost as much before.

And I think we're shifting a little bit to where back into my world, we are going to have to start addressing ROI, right? What is the investment? Because now finally, we've been playing with these tools for a while.

Hopefully, we're AI literate. Those days are kind of, I would say, we're on the downside of that cycle to where now we have to learn how to manage, just like we manage-- I've had to manage human resource costs for most of my career, right? And justify- Mm-hmm ...

the expense of it. I have to do the same thing for agents and tokens and things like that. And to your point, when do we have some type of routing system so we use this lesser expensive model than that?

So I think we're getting out of the literacy phase, even though it's infancy, because I think there's just a lot of organizations who still aren't there yet, and getting into more the cost and ROI. So I, as a leader, how do I go to a board or how do I go to an executive group and justify my costs and what are we doing to curb those guys? Almost like you said, here we go again, cloud, right?

So it's just a thing of- FinOps ... yeah, FinOps, exactly. It's FinOps for AI.

Same thing. But here's the interesting point that I find fascinating, just at this stage, Phil, I look at this almost, it is an experiment. It is almost a case study.

And so what's happening is just when we're really going to have to start paying the true cost of tokens, right? It's not going to be subsidized because these companies are going public, and they need to be profitable. At the same time, the pressure is on to sort of make intelligence a utility, right?

Make it a commodity, and so that will hopefully bring down the cost of tokens. Mm-hmm. So that we need to find an equilibrium here, right?

How can we make AI truly affordable without being subsidized, right? And that's going to take breakthroughs in energy. It's going to take breakthroughs in cooling, and they're building $8 trillion in data centers.

Someone has to pay for this, right? So it's going to be interesting. But all of this, if you look at it through the lens of flow and how it flows within your organization and even at the broader level, it's fascinating.

Fascinating stuff. Yes. Anyway, we went astray of your talk there.

But, for those who say, "Well, I don't know where this, where rubber meets the road," that's where rubber meets the road. Yeah. Right?

That's why it's important. Anyway, Phil, go ahead. I was going to say, exciting times.

Bring it back to Flowtopia and flow and all that. AI, highly-- I'm a little bit biased because I think value stream management, just being able to map and making work visible as well as teams more efficient, AI has really stepped in and helped a lot of things I used to have to analyze myself. I can allow AI to do that now, so I can have real conversations with the teams on the bottlenecks and their solutions and things like that.

So I think AI is really stepping in and helping in that area. Also with Flowtopia, you are really this year compared to the previous years, always been a great event, but you're going to get a really broad range of opportunity of experts, to learn how to improve efficiency, outcomes, realization, and AI, right? How AI is impacting the world.

So I think if you can grab a ticket, I'm not paid to do this. I just think it's an awesome event, and I hope I can continue to contribute to that group for the next several years. I love it.

Thank you, Phil. Hey, one last thing, for people who maybe want to reach out or follow you or see what's the best way to do it? So two primary ways, LinkedIn /philclark.

And then the second way, I do have my own website that I've had out since 2019 called Rethink Your Understanding, which comes from originally my own transformation on my own leadership. com. And then at the end of the month, I should have my book, "Profitable Engineering: Transforming Technology into Business Partners," come out by the 30th of this month.

But I think LinkedIn and my website are the easiest ways. There's contact forms on there. Excellent.

Yeah. Phil, congratulations on the book. I'm sure you're going to have a smashing session at Flowtopia.

Thanks for coming here on techstrong TV, and good luck to you. Yeah. Thank you for the opportunity.

My pleasure. Phil Clark, speaker at Flowtopia, as well as RYU Advisory Media LLC. He's got his book coming out.

Check it out. We're going to take a break here on techstrong TV. We'll be back in a bit.