This podcast episode features Ryan Ferris interviewing David Hirschfeld, founder of Tekyz, a software development company. The discussion centers around Hirschfeld’s extensive experience with startups, his company’s approach to software development, and his “Launch First” methodology, which challenges traditional startup practices. He challenges the conventional MVP approach, emphasizing early validation, revenue generation, and the power of AI-driven efficiency. Learn how Hirschfeld’s strategies help startups succeed faster and wiser.
The Value Maker Podcast: A Conversation with David Hirschfeld
This podcast episode features Ryan Ferris interviewing David Hirschfeld, founder of Tekyz, a software development company. The discussion centers around Hirschfeld’s extensive experience with startups, his company’s unique approach to software development, and his “Launch First” methodology, which challenges traditional startup practices.
From Mainframes to Startups
Hirschfeld’s journey began in the late 1980s with IBM mainframe sales. He transitioned into software development after realizing he could create more effective client presentations. This led to roles at Texas Instruments, where he ran projects for major corporations like Intel and Motorola. He then co-founded a logistics startup, growing it to 800 customers in 22 countries before selling it in 2000. This experience and observing the successes and failures of over 90 startups through Tekyz informed his understanding of what drives startup success.
The Tekyz Approach: Transparency and Trust
Tekyz primarily works with non-technical founders, often serving as their outsourced development team for extended periods. Hirschfeld emphasizes their commitment to transparency and trust, highlighting their practice of attaching detailed timesheets to every invoice. This meticulous tracking, he argues, eliminates potential trust issues and allows clients to focus on the value being delivered. Tekyz also avoids working with competing clients to maintain trust and avoid conflicts of interest.
The Launch First Methodology: Pre-Launch Sales over MVPs
Hirschfeld introduces his “Launch First” methodology, prioritizing pre-launch sales over the traditional Minimum Viable Product (MVP) approach. He advocates for creating highly polished animated design prototypes that convincingly showcase the product’s functionality. These prototypes are then used to secure early buy-in from customers through attractive offers like lifetime licenses. This approach, he argues, generates early revenue, validates product-market fit, and allows for quick and inexpensive pivots if necessary. He criticizes the common practice of free betas and delayed sales efforts, often leading to wasted time and resources.
Identifying the Early Adopter
A crucial element of “Launch First” is identifying the ideal early adopter. Hirschfeld emphasizes the importance of initially focusing on a single niche and understanding the specific problems that resonate with them. He recommends a metrics-driven approach involving scoring problem statements based on their perceived impact and actual cost to the stakeholder. This process helps identify the niche with the highest potential for early adoption and informs the product’s value proposition and messaging.
The Impact of AI on Software Development
The conversation shifts to the impact of AI on software development. Hirschfeld discusses his company’s increasing use of AI tools like Cursor, ChatGPT, and OpenAI 4.0 to accelerate growth. He shares an example of building a referral partner portal using AI, estimating a dramatic reduction in development time. While acknowledging the challenges of UI generation, he expresses optimism about the potential of AI to reshape software development teams and project approaches.
Key Takeaways
This podcast episode provided valuable insights into David Hirschfeld’s entrepreneurial journey and his unique approach to software development. The discussion highlighted the importance of transparency, trust, and a metrics-driven approach to building successful startups. Hirschfeld’s “Launch First” methodology, emphasizing pre-launch sales and early adopter identification, offers a compelling alternative to traditional startup practices. Finally, his experiences with AI-powered development tools suggest a significant shift in the future of software development.
Transcript
[0:00–0:15] Ryan Ferris: Welcome to the Value Maker Podcast. My name is Ryan Ferris. I’m your host, supporter of Heather Bicknell. As we meet with the people that make the digital workspace work, it’s about the technology, it’s about improving people’s work experience and finding the return on value.
[0:16–0:27] Ryan Ferris: Thanks for joining us on the show. Please like and subscribe and share with your colleagues. It helps us grow the podcast and we appreciate any help you can give us. David, it’s the Value Maker Podcast. You want to introduce yourself, please?
[0:27–0:37] David Hirschfeld: Yes, hi, I’m David Hirschfeld. I founded Tekyz, my current company, almost 18 years ago. We do software development.
[0:37–0:39] David Hirschfeld: Would you like me to go into my history?
[0:40–0:44] Ryan Ferris: Yeah, tell me a little bit about what you guys do in your history.
[0:44–1:05] David Hirschfeld: Yeah, yeah, yeah. So how I came to where I am now. So I started out literally in the late 80s in the software world doing IBM mainframe sales and then ended up teaching myself development because it always took me too long to get a systems engineer scheduled for a presentation.
[1:06–1:29] David Hirschfeld: And they weren’t as good as I felt they should be, so they weren’t really connecting with clients. So I ended up teaching myself the products and doing all my own demos. Then I ended up getting hired at Texas Instruments, and I did the same thing there with their big case, which is computer-aided systems engineering products for several years, doing all my own demos and building prototypes.
[1:30–1:46] David Hirschfeld: And then I ended up running projects for leading Texas Instruments and ran projects for big companies like Intel, Motorola, Allied Signal, Arizona Public Service, until I started my own startup in the early 90s, logistics, route distribution, inventory management.
[1:47–2:00] David Hirschfeld: And I did that while I was running projects for these other companies for a few years until our company grew to where my partner and I had to go full-time because we had a bunch of employees and grew it despite every effort.
[2:00–2:18] David Hirschfeld: And despite every effort on both of ours, we grew it to 800 customers in 22 countries and sold it in 2000 to a publicly traded Toronto firm. And I was VP of products for that company for the next several years until I left there, cast about for a couple of years, and then started Tekyz.
[2:19–2:29] David Hirschfeld: We’ve worked with a lot of different size companies in a lot of different industries, but primarily there’s been a lot of startups, over 90 startups during those last 18 years.
[2:30–2:41] David Hirschfeld: And I thought I knew what made startups successful because I had a successful one. And I actually was probably more lucked in anything else in that first one because there were many startups I thought should be successful that failed.
[2:42–2:57] David Hirschfeld: And a couple, one in particular, my most successful that I didn’t think would make it. And I started to learn the patterns of why the few that were successful were successful and why the vast majority that failed, why they failed.
[2:57–2:59] David Hirschfeld: Anyway, that’s led me to where I am now.
[3:00–3:10] Ryan Ferris: So a couple questions. So funnily enough, I was trying to pronounce Tekyz a few times because Tekyz for me in South African means sneakers.
[3:11–3:12] David Hirschfeld: Oh, that’s funny.
[3:12–3:16] Ryan Ferris: And it’s T-E-K-K-I-E-S. Okay.
[3:16–3:26] David Hirschfeld: And then obviously, in the U.S., Tekyz typically means a bunch of people that are technical sitting around talking about. Right.
[3:26–3:26] David Hirschfeld: Yeah.
[3:27–3:38] Ryan Ferris: And that’s what I thought you were trying to get with the combination of letters. So that was the first thing. The other thing that I was interested in is you’ve come together with this methodology, Launch First.
[3:39–4:06] Ryan Ferris: And obviously, we’re in your history with some of the stuff you mentioned already. what I was trying to wonder about was did you always envision being at this point sort of if you look back to where you started out did you always think you’d be involved in startups and building software quickly or was it basically a you kept falling into the same pattern and then you recognized the pattern and you went oh actually this is the pattern
[4:06–4:18] David Hirschfeld: a little of both so I knew I was just going to do custom software projects when we started Tekyz because that’s what it was all about. I figured along the way I would also do some of my own startups.
[4:19–4:32] David Hirschfeld: Okay, yeah. Right. But the startup community has always been really interesting to me. I mean, I started out in enterprise, right, but I find startup community a lot more interesting.
[4:32–4:43] David Hirschfeld: When they’re really trying to solve a problem that’s close to them, as opposed to I’ve got a cool idea for the next social network. Those aren’t as interesting to me. I know those are going to fail.
[4:43–4:57] David Hirschfeld: and one in a thousand of those will find a path to revenue. But when somebody’s trying to solve a real-world problem, especially when it’s a sophisticated problem, a complex problem to solve, that really kind of gets my energy going.
[4:57–5:16] David Hirschfeld: And it doesn’t really matter even the industry that it’s in because it’s the problem-solving process, whether it’s a workflow automation issue or whether it’s a fuzzy logic issue or whether it’s a whatever it is, I know that it brings value to the stakeholders in that industry in a way that they didn’t have that value before.
[5:17–5:29] Ryan Ferris: And are you guys getting involved in augmenting a team? Are you taking over from a team or are you just being a net new, we’ll do this thing to prove it works for you as a resource project?
[5:30–5:48] David Hirschfeld: Yeah, so we’re technically working with non-technical founders, sometimes technical founders, but usually non-technical founders who either are creating something new or they’ve built something and launched it, but now they really need a bigger, more galvanized team to do it.
[5:48–6:06] David Hirschfeld: They might have had one developer working on it before, and they really need a team because, you know, there’s a lot of pitfalls that you have if you don’t have all the different people involved with their unique expertise, you know, the idea of team, of running projects.
[6:06–6:17] David Hirschfeld: Anyway, so — but, yeah, so we are usually the outsourced team, if you want to think of it like that. And very often we remain that for that client for many years.
[6:18–6:29] Ryan Ferris: Okay. That was another question. Do you try and share it over to, you know, sort of help them? Do you deliver an MVP? They might raise some money, then they bring their own team in, and you hand it over, or do you just carry on?
[6:29–6:41] David Hirschfeld: It depends on the client’s needs. It depends on the investor’s requirements if they need to bring things in-house, right? But, I mean, we want to continue on, all right?
[6:41–6:57] David Hirschfeld: We get very invested, personally invested in these projects. We have an awful lot of intellectual property around the product. I mean, we never, like one of the rules we have is we’ll never take on two clients that are competitive in any way with each other.
[6:57–7:29] David Hirschfeld: so if we’re working with one client even if they’re small and some big client comes along but it’s a clear competitor to the small client we just walk away from that because it’s too much of a conflict of interest and that rarely ever happens it has happened a couple times and we had to say no to projects for that reason but we don’t want to be in a position where now our clients aren’t going to trust us because we’re going to but they’re telling us, they’re going to find out that we’re working on something else that’s in a competitive
[7:29–7:54] David Hirschfeld: situation with them. And then we have to say, no, no, no, we’ve got a Chinese wall between us. And that becomes a trust issue. Anything that creates a potential trust issue with our clients, we just don’t go there. I’ll give you an example for that. And this is something most software companies don’t do. When we started out, we used to do only fixed price.
[7:54–8:09] David Hirschfeld: and we stopped doing mostly fixed price. We still do, but we do much, it’s a small percentage of our projects because it’s restrictive for our clients because then we have to hold the scope based on what we’re bidding.
[8:09–8:21] David Hirschfeld: The clients never understand all of the requirements that they actually need until we really dig into the design of the product. So then we went to time and materials and we had a blended rate.
[8:21–8:31] David Hirschfeld: and then clients were uncomfortable with the blended rate when we got to the end of the project and most of the work was testing in QA and they felt like they were paying too much for that.
[8:31–8:42] David Hirschfeld: And even though they paid a lot less for the front, it didn’t matter, right? So I said, okay, fine. So then we started doing individual rates for the specialty of the person who was on the project, right?
[8:42–8:57] David Hirschfeld: And then they said, that’s great, but why did this person work 180 hours this month? That seems like an awful lot of hours. What did they do? And it was like, then we have to go. And I said, okay, forget that. So we are really meticulous about timesheets, about tracking that.
[8:58–9:13] David Hirschfeld: And we do it in a lot. We have a lot of automation, so it doesn’t put a burden on the developers. But we get every single day, we know what they worked on at a high level, you know, as well as at the detail because our project tracking, they’re updating it.
[9:13–9:46] David Hirschfeld: And we attach those timesheets to every invoice every month for every client. We’re the only company I know that does that. I’m sure we’re not the only one, but it’s a level of transparency, and I did that because I don’t want that client to have a trust issue with me because there’s no reason to, and we’re meticulous about what we do, and procedures and protocols and tracking, which is also kind of you can see that when we show clients how we do status reports, our weekly status reports.
[9:46–9:57] David Hirschfeld: the level of detail is way beyond what you see in most shops. I just don’t want there ever to be a question about are they doing their job? Am I being billed for the work they’re doing?
[9:57–10:02] David Hirschfeld: You know, those sorts of things. So I don’t like the way that feels. Yeah. Pardon me?
[10:03–10:11] Ryan Ferris: It’s the wrong focus. I mean, if everyone’s focusing about that, then you’re not focusing on are you getting value for what you’re doing? You’re delivering what it should be doing.
[10:11–10:35] David Hirschfeld: Right. And I talk with a lot of other people that run software development firms. Most of them just sort of skirt those issues. They figure, you know, a client will ask that, and they’ll just sort of skirt the issue instead of — because it’s a lot of — it takes a lot of discipline and protocol and culture to be able to track things in that detailed manner every single week, every single day, every single month.
[10:36–10:50] David Hirschfeld: So, you know, and if you go to my website, at the very top, it’s kind of my marquee, is hyper-exceptional software development team. But I don’t expect anybody to believe me because I say that, right?
[10:50–11:02] David Hirschfeld: So I have, if you’re really an exceptional team, then you have all this evidence that are just artifacts you produce naturally because you’re doing things at an exceptional level that typical software teams don’t do.
[11:03–11:31] David Hirschfeld: The way we do estimates, the way we, and I’m not, you know, I feel now I’m pitching myself and it wasn’t why I wanted to be interviewed by you, But it’s important because I think, you know, whether you’re a startup thinking about creating a new product or whether you’re an existing company looking at it really kind of changing the nature of your team or you’re trying to build a product and you’re struggling with a team, those are some of our favorite projects.
[11:31–11:44] David Hirschfeld: And you realize that it’s taken way too long. It’s not going to get any better. it really matters that if, let’s say you do switch to another team, that it’s a dramatic shift, right, from what you’re experiencing.
[11:45–11:51] David Hirschfeld: And so that’s partly why it’s some of our favorite projects because people realize really quickly the difference.
[11:52–12:06] Ryan Ferris: Yeah, and look, you know, part of this is to pitch yourself. You know, it’s the value you bring to your customers. And I think, you know, I mean, I do have a friend who runs a dev shop, but they do also have meticulous timesheeting and it’s by the line and all the rest.
[12:06–12:26] Ryan Ferris: In fact, sometimes you would be frustrated with the change. Like, I’ve spent so much time building a timesheet management system, I shouldn have sold this as a product you know because it becomes such a thing But I think you know as we move down this route of automation more and more intelligent automation you know generated code and all the rest of it, that’s even more sophisticated than it was 10, 20 years ago.
[12:26–12:39] Ryan Ferris: Oh, yeah. Understanding the metrics around what we’re actually saving and being able to exponentially do, people can’t grasp. If you’ve never coded, people can’t grasp.
[12:39–12:49] Ryan Ferris: If you’ve never coded. Right, yeah. They never understood that sometimes to fix a problem that you’ve coded could take three days or could take five minutes. Right. Because of whatever.
[12:51–12:51] Ryan Ferris: And as we start generating…
[12:51–12:55] David Hirschfeld: That’s just the nature of the problem, right, that you’re fixing.
[12:55–13:05] Ryan Ferris: Yeah. Yeah, exactly. I mean, you could have a timing issue between, you know, two APIs that are calling each other and there’s a thing and you’ve got to factor it in and you’ve got to handle retries and whatever.
[13:05–13:19] Ryan Ferris: I mean, that could be… I’m sorry. But the thing that I think will happen is as we start generating more and more code that’s going to talk to more and more other things that have been generated, you’re almost going to be obfuscated by what’s been done.
[13:20–13:30] Ryan Ferris: And the time to troubleshoot that and to figure that out, it will be important, too, to know why you’re spending so much time fixing the problem. Right. So I think being meticulous is important.
[13:30–13:45] Ryan Ferris: I think it’s an important part of the fundamentals. Oh, yeah. It’s critical. And looking at your launch first methodology, and I’m primarily looking at some of the diagrams, and I love your animations, by the way.
[13:45–13:48] Ryan Ferris: Thank you. I think that’s awesome. What did you do those in?
[13:49–13:56] David Hirschfeld: Those are animated vector graphics. I think they were done in Flash, but they’re Lottie files is what we created.
[13:56–14:01] Ryan Ferris: Oh, they’re Lotties. Okay, yeah. Yeah, they’re Lotties, yeah. Yeah, I love them. Not Flash.
[14:01–14:14] David Hirschfeld: I mean, Fire, Flash. I’m going back like 15 years now. Yeah, right. Yeah, but they’re Lottie files, so they’re animated vector graphics.
[14:14–14:21] Ryan Ferris: Okay, no, they’re brilliant. I mean, I can actually have them on my side screen and just watch them. Like a fish swimming in an aquarium.
[14:22–14:22] David Hirschfeld: Oh, thank you.
[14:23–14:34] Ryan Ferris: Anyway, so the thing that I love about your diagrams is you’ve got this flow of time, which would be based on ideation, you know, whatever it is. But then you’ve got the barrier of viability.
[14:35–14:46] Ryan Ferris: And it’s very similar to what I have with the value methodology that we use, which is probably very similar, where there’s always going to be these balancing points or levers to what you’re doing.
[14:47–14:58] Ryan Ferris: And at some point you’ve crossed that barrier that it’s ready for it to show its value. So if one-to-one was — it cost you one and it delivered one, then that’s not going to be a good return on your project.
[14:58–15:05] Ryan Ferris: That’s a failure. That’s a failure. Yeah. Right. Yeah. Because you always want the return to be a little bit higher, you know, 2X, 5X, whatever it is.
[15:05–15:13] David Hirschfeld: We use 3X as sort of the golden rule. You need to be a three-to-one cost of acquisition versus lifetime value.
[15:13–15:18] Ryan Ferris: Yes. And your lifetime value, what do you normally put that as?
[15:20–15:34] David Hirschfeld: I mean, that’s part of the calculation. It depends on the product, the customer. And then you’re doing estimates on what that lifetime value would be when you get a customer, right? It may be short initially just because you don’t have product solution fit, even if you’ve got product market fit, right?
[15:34–15:45] David Hirschfeld: But you can usually estimate what that lifetime value is for a customer within a reasonable range. And so when you get the customer, how much did it cost you to get them?
[15:45–15:53] David Hirschfeld: And if you don’t have a three-to-one ratio, you don’t have enough energy, enough profit to be able to then put energy back into the system to scale. Yes.
[15:54–16:18] Ryan Ferris: Okay, yeah. That’s awesome because that’s the whole loop, the feedback loop thing. So when I looked at this and I looked at sort of the way you change things, so you’ve got your revenue bar at the bottom, and you’ve got your expenses going up and your risk at the top, because I think a lot about this is risk mitigation.
[16:19–16:32] Ryan Ferris: How are you educating your customer? your customer being the software founder yeah so maybe your customer is how do you start the conversation that they get to trust you
[16:32–16:55] David Hirschfeld: ok so it’s usually not a trust issue usually just because I’ve been doing this a long time people recognize not that I know how I’m going to be the right way to launch their business but that they’re at least that I have a valid position from which to that I’m coming.
[16:55–17:07] David Hirschfeld: Well, let me, let me say it a different way. Most founders, when they come to me with the rare exception are wearing the black robe, right? They have a vision. They’ve got this product idea.
[17:08–17:21] David Hirschfeld: There was probably started out with a problem that they were trying to solve. Right. And then they created this, this automation or product concept that they know is going to solve that problem. And they believe that everybody’s going to need it.
[17:22–17:34] David Hirschfeld: and want it, right? And that’s like the kiss of death for a founder. So I gently peel that black robe off, right? Because it’s basically right now it’s a religion.
[17:34–17:45] David Hirschfeld: They believe it. There’s no proof. And I slowly kind of wrap them in a white coat and say, you’re now a clinician. You know, if you believe it, then you need to prove it.
[17:45–18:00] David Hirschfeld: And whatever assumptions you’ve made are nothing more than assumptions until you have metrics and validation and enough metrics and validation that you can predict that it’s predictable as you’re out in the market and you grow.
[18:00–18:20] David Hirschfeld: So, and the only way you can prove that you’ve got product market fit and product market fit means people want to buy you, pay you for your product at that three, at least at that three to one ratio or higher, then the only way you can prove that is that you’re selling your product to customers and they’re paying you money.
[18:21–18:42] David Hirschfeld: And the biggest mistake founders make is that they wait way too long to ask customers to pay them money. Usually they’ll get their MVP done in six months to a year, you know, depending on the level of sophistication of the product and their ability to stay focused on delivery of that product.
[18:42–19:14] David Hirschfeld: and then they put it out for free betas, which I’m not a fan of free betas. And then they’re getting feedback, very often misleading feedback, because these are not invested customers, and they’re making changes and making shifts trying to get to product solution fit before they have product market fit, and they spend another year, year and a half, sometimes anywhere from 18 months to three years before they actually are ready to drive a stake in the ground and do a formal process of launch and marketing, which is crazy.
[19:14–19:24] David Hirschfeld: And that’s when they find out they have no idea who their customer is. People aren’t buying it. They don’t know how to message. It’s the wrong product.
[19:25–19:35] David Hirschfeld: It has the wrong features. The value proposition isn’t — they can’t articulate it in a way that anybody cares. And now they’re in panic mode because they’ve spent all this money, all this time.
[19:35–19:46] David Hirschfeld: They’ve got this big roadmap of features, but nobody’s paying for the product now. and they’re trying to pivot and figure out the right message, and then this is another one, two, three years slog down the hill until they fail.
[19:47–20:00] David Hirschfeld: And I’ve seen this happen over and over and over and over again. And they could have easily figured this out three, four months into this by doing a prelaunch sale model, which is what Launch First did.
[20:00–20:14] David Hirschfeld: So Launch First, we say forget MVPs. MVPs are once you’re sure you know who your customer is and that they’ll buy it, then you can build your MVP.
[20:15–20:37] David Hirschfeld: Before that, you need to have a really elaborated animated design that looks like a real product that you can put in front of customers and ask them to buy in early. And if they won’t buy in early, and you do that by offering them some kind of high-value proposition, And if you’re solving a big enough problem that costs enough to that customer that they know when this product comes out, they’re going to have to get it.
[20:38–20:50] David Hirschfeld: And you’ve shown them a demo that’s realistic enough that it looks like you’ve already built the product, even though it’s just a very animated design mock-up, because you don’t want them to ask the question, how do I know you can build this?
[20:51–21:06] David Hirschfeld: And if the prototype, the design prototype you’re showing them is realistic enough, that question never happens. So then the question is, can you offer them a high enough value opportunity to buy in early and the cost of the problem you’re solving is big enough?
[21:07–21:18] David Hirschfeld: If not, you’re not going to probably sell the product when it comes out. So you do things like lifetime licenses. You know, people go lifetime license, then I won’t ever get any more money from that customer.
[21:18–21:30] David Hirschfeld: Well, yeah, you probably will. But let’s say you’re selling lifetime licenses for some big high-value B2B thing. and you’re charging $1,500 for it.
[21:30–21:50] David Hirschfeld: And you can sell 100 of these, right, because the value proposition is so big. You raised $150,000, which would fund your development, and you’re giving away an immeasurable fraction of a percent of your market, typically, right, instead of 10%, 15%, 20% equity to an investor.
[21:51–22:06] David Hirschfeld: And all that time you spend trying to recruit that investor, not any of that moves your business forward. Whereas these pre-launch sales, you’re basically launching a sales and marketing engine around your business and learning how to sell it, learning how to talk to customers.
[22:06–22:17] David Hirschfeld: And if nobody’s buying it, you pivot quick and cheap because you’re getting feedback why nobody wants it, right? You’re not hitting the right market, the right product. You’re not going after the right high-value opportunity.
[22:17–22:28] David Hirschfeld: And you can pivot two or three or four times in the space of, you know, a few months. and if you still can’t find any way to revenue, then you can say to yourself, I fail fast and cheap.
[22:29–22:43] David Hirschfeld: And for anybody who’s a startup founder, they should pray for fast and cheap failure if they can’t find a path to success because the alternative is incredibly expensive and a big portion of your life gets tossed.
[22:45–23:13] Ryan Ferris: Yeah, I’m nodding because this is exactly what I’ve gone through with value. we’ve built a product we’re at that point now where we still quite haven’t closed the loop on a few things and I was thinking about lifetime licenses I was thinking well let’s let me go and give it away for whatever the number is for lifetime and I do think you will get paid by those same customers in the future because they’re buying the core product and you can keep adding on stuff it’s how HubSpot works to a large extent
[23:13–23:28] David Hirschfeld: but even if they don’t you’re still generating revenue and now you’re getting customers using it who are invested in the product. And so you can focus on product solution fit at that point if they’re buying it.
[23:29–23:30] David Hirschfeld: So, yeah.
[23:32–23:34] Ryan Ferris: It’s a fantastic way to do it.
[23:35–23:50] David Hirschfeld: Now, one of the key success factors of this is that you nail down who that early adopter is and what’s the one or two or three at most problems that you’re solving for them.
[23:51–24:06] David Hirschfeld: And because the early adopter, if you think about it, when you’re building a product, usually you have all this broad number of niches that you can market it to. Yeah. But you can’t market to, you know, 15 or 20 different niches when you’re new, right?
[24:06–24:20] David Hirschfeld: You can market to one. And this is another huge mistake that a lot of startups make is they try to — they’re too broad. And so I spend a lot of time with founders on what is the root problems that you’re solving.
[24:21–24:50] David Hirschfeld: And before we do that, we have to say, who are all the niches that you can market to? And if you sitting in that stakeholder standing in that stakeholder shoes what is how do you articulate the problem so that they feel threatened by it All right It not just yeah I have trouble with this issue but because I have trouble with this issue, it’s causing me this and I’m, and that may affect my career or my, my, or I’ll lose the competitors or whatever, whoever the stakeholder is.
[24:51–25:01] David Hirschfeld: What is this? How did, why did they feel threatened by that problem? So you have to ask that question. Why does this matter? Like three times, four times to get to the root level. And why does that matter?
[25:01–25:15] David Hirschfeld: Because they won’t pay attention to you if they don’t think you understand what they’re struggling with, right, what really keeps them up at night. But if you nail that really well, then they’ll lean forward and they’ll say, I hate that.
[25:16–25:27] David Hirschfeld: Can you actually fix that for me, right? And now you’ve got their attention. And once you’ve got that, then the solution should just be the natural mitigation of that, you know, the problem mitigation.
[25:28–25:42] David Hirschfeld: So who is that? And how you figure this out is a founder, not a founder, a stakeholder that’s struggling with a problem has, there’s two factors that you need to measure.
[25:42–25:56] David Hirschfeld: One is how much do they feel that problem impacts them personally, right? That is like they feel threatened. What’s that actual threat? And the second one is how much does that problem actually cost them?
[25:56–26:27] David Hirschfeld: those are independent numbers because you can have somebody with a really big cost but everybody in our industry deals with that same problem you know my boss isn’t going to care if i try to solve that because it just doesn’t matter and even if i show him these numbers it’s going to go that’s not that’s not as important as growing our business or whatever right big cost but no personal no perceived impact or low perceived impact and the opposite is possible god i hate that it drives me nuts I waste so much of my day on that.
[26:27–26:41] David Hirschfeld: But then when you quantify the value, it’s like, yeah, but it’s only costing me, you know, I can’t really justify much money to buy it. And so you can’t charge much. So what you’re looking for is both those numbers are really high for the root level problems that person’s struggling with.
[26:41–26:53] David Hirschfeld: And when you find that, then you know how to talk to them about the problem. You know how to charge for it. You know how to explain the value proposition, right? And you know how to reach them, who that person is.
[26:53–26:57] David Hirschfeld: when you got that, that’s when you test yeah
[26:57–27:11] Ryan Ferris: I totally get it because the way that I would normally cost for that would be take a third of that price that you come up with and that’s roughly your indicator indicator price maybe a third
[27:11–27:24] David Hirschfeld: maybe a fifth, maybe a tenth it’s like it’s the number that gets that person to start to go, huh I don’t know that this is really good at right It’s whatever that number is, right?
[27:25–27:25] : Yeah.
[27:27–27:28] David Hirschfeld: Yeah, exactly.
[27:29–27:38] Ryan Ferris: And I guess the question is those one to three things that you mentioned, the one to three problems, how do you determine — you might have ten problems on the list.
[27:38–28:12] David Hirschfeld: Right. Well, you usually do ten to 12 to 13 problems on your list because of all the different niches that you might market this to, right? So you do a scoring. Okay. So, like, a lot of methodologies have this concept of discovery interviews, but it’s siloed and subjective. So, like, you put your interview questions together, right, and then you have a discovery session, right, where you go out and you call people that — and usually it’s not very niche-specific. And so you’re calling people across niches,
[28:12–28:44] David Hirschfeld: and you’re getting mixed feedback and very often being taken in the wrong direction in terms of trying to figure out what matters. Also, you’re not focusing on their problems in these discovery interviews. You’re focusing on some problems and your solution. So what I say is that whole concept throughout the window, you got to figure out who the niches are. Once you do that, you figure out all the root problems. Then some of these you may know, some of the niches you may know really well because you may have worked in those niches, right? Others you don’t know
[28:44–29:17] David Hirschfeld: anything about. The ones you know, you can kind of score. And I have a whole scoring methodology for how we do this. It’s all metrics driven. You score each of the problem statements for that niche, you know, in terms of impact. One sheet is impact only. And then, but if there’s a niche that you’re not that familiar with, or maybe a little familiar, but not completely, then you call one or two stakeholders in that niche and you ask them about the problems. You focus on the problems. What are the problems that you struggle with? Here’s some problems that we’ve identified
[29:17–29:50] David Hirschfeld: in your industry. How much do you perceive that those impact you on each of the problems? Are there other problems that we missed related to this, right? And it’s niche specific. So, right, just one or two is all you need in that niche. So, you end up doing a lot of iterative discovery. You do that on impact and you do that on a separate sheet where we just do cost, same thing. And then you bubble chart those things together. You take the top three problems from a cost perspective for each niche, and they’re not the same problem. And then you go to
[29:50–30:08] David Hirschfeld: the score sheet and pull those problems and sum up the score on that side. And now you’re going to get a bubble chart that takes each niche and floats it up to the upper right or lower left or different places in a four-slice quadrant. The ones in the upper right are your likely early adopters.
[30:08–30:20] David Hirschfeld: and then we do deep dives on the top two or three to figure out which one really is because there may be reasons why one that floats way up to the upper right is a bad early adopter.
[30:20–30:30] David Hirschfeld: Like might be CEOs of Fortune 500, you know, the sales cycle is really long, the cost to reach them is super high, you know, that drops them out as an early adopter, things like that.
[30:30–30:34] David Hirschfeld: And others, they get promoted higher. So, yeah.
[30:35–30:48] Ryan Ferris: I love it. It’s very similar to what I do with value. And even to the point of quadrants, although we’ve gone from four to six quadrants now, but that’s just because we found there was a need for a middle one or a middle column.
[30:49–31:08] Ryan Ferris: And what I love about what you’re doing is you’re prioritizing, and you’re getting the sequence right inherently because you can start seeing where there’s weight. And it’s moving away from opinion to evidence-based decisions, which is always hard because everyone has an opinion, but no one really wants to trust the evidence.
[31:08–31:13] Ryan Ferris: Right But you can get around that when you have more of it
[31:13–31:32] David Hirschfeld: It’s much easier to trust the evidence When you’re documenting the metrics And then you can chart this And then all of a sudden it’s like Oh, I wasn’t really thinking that was the niche That we should be going after But now it’s starting to make a lot more sense Yeah
[31:32–31:42] Ryan Ferris: And what sort of capture methods do you do for this? Is this, do you have your own tools you guys have built, or are you using tools in the industry?
[31:42–31:54] David Hirschfeld: This is the hardest part because this is a very tedious process for founders to go through. They don’t want to. They want to get to the product because that’s, and they think if they build it, it’ll be easier to sell, and it just doesn’t work like that.
[31:55–32:09] David Hirschfeld: So, right. I mean, and we know there’s so many examples of this in, like, in the B2B SaaS, I do this. You either do the sales in a webinar manner or a one-to-one manner, depending on your reach and the value.
[32:09–32:23] David Hirschfeld: Because these can be very expensive pre-launch sales. Or, you know, it could be like, let’s say, in a clinical trial. This is one of the ones we’ve done, clinical trial software, where these were $250,000 pre-launch sales.
[32:24–32:39] David Hirschfeld: And they would get free implementation, which usually costs half a million dollars. for, and so to them that was a huge value, and it was a new pharmaceutical product for a small boutique pharmaceutical maker, that sort of thing.
[32:39–32:59] David Hirschfeld: We sold two licenses at $250,000 each. Pre-launch, there was no software yet, but we had a design prototype that was just killer. So they thought they knew what they were going to get, and they didn’t need the software for like nine months, but they had to make those commitments early, because that’s how clinical trials work.
[32:59–33:14] David Hirschfeld: And then another one was real estate investment portfolio manager product, which we sold 23 licenses in 60 days and generated $68,000 in those 68.
[33:14–33:28] David Hirschfeld: And now we had 23 clients. One of those sales was like $21,000. I had a company that was creating a hedge fund. They hadn’t created it yet, but they were creating a hedge fund that was going to be buying up all these apartment units, and they bought into it.
[33:28–33:55] David Hirschfeld: So, and again, we’re giving away literally an immeasurable fraction of a percent of the market. So it’s great, and we’re bringing revenue forward. A third one was for aerospace parts distributors maintaining the history of products in assemblies, and those assemblies become products in other assemblies, and you have to have all these certification certificates all the way from a nut all the way up to a 747.
[33:55–34:09] David Hirschfeld: You know, so there’s millions of these certificates, and the parentage of these certificates has to be maintained. And it’s a big deal, and it’s very hard to do, and there was no system to do this previously.
[34:09–34:31] David Hirschfeld: And sometimes something would happen, like there would be a fuel leak in an F-15, and because there was a shipment that didn’t have all the certificates, the history of certificates, they literally had to spend like $100 million to pull the fuel systems out of like 30 of these F-15s.
[34:31–34:45] David Hirschfeld: That’s the kind of impact that this kind of thing. And that’s a big one. Some are very, very small. But anyway, so we sold this system, three of them. We did three demos, sold all three for $15,000 each.
[34:46–34:56] David Hirschfeld: And so, I mean, it can be anything, right? Yeah. It depends on the — if it’s a mass product, then crowdfunding kind of approach like Kickstarters and things like that is better.
[34:57–35:21] David Hirschfeld: If it’s a really inexpensive product and you’re trying to get mass market adoption, prove that there’s a mass market for it. But for high-value SaaS products, then this is like the only way to — and I can’t tell you how many of the 90-plus clients that I’ve worked with failed because they were two years in before they tried doing any real serious sales.
[35:24–35:43] Ryan Ferris: Yeah, no, I can imagine. I mean, you know, because then you’re doing desperate sales, and that never goes well with a customer. Just a couple of questions. Let’s go back to the start of two things. So your approach with your customers, are you guys using your own frameworks that you’ve built to build this quickly flow in?
[35:43–35:51] Ryan Ferris: or are you starting everything pretty much fresh so you’re not diluting or sharing IP from other projects across different customers?
[35:52–36:07] David Hirschfeld: We don’t share IP from other customers. We reuse common components, right, that we might build and enhance in a customer, but they’re not IP of any kind. But for the most part, we’re building most everything from scratch.
[36:07–36:22] David Hirschfeld: We have components that we’ve built that are common that we just built out and created a lot of flexibility in them, and we reuse those. In particular, like in authentication, permission management, role-based roles and permissions, all that stuff.
[36:22–36:45] David Hirschfeld: You know, these are things that are fairly large efforts, but we’ve built them many times. And so we have really good componentry around this. So it doesn’t, you know, we can snap them in and customize them for particular clients and tenant management for distributed systems that distribute role and management down and level.
[36:46–36:58] David Hirschfeld: Like I a corporation and I have distributors and my distributors have distributors and each one wants their own portal and it just kind of right where it kind of rolling downhill Those sorts of things.
[36:58–36:59] David Hirschfeld: Yeah, those frameworks.
[37:00–37:07] Ryan Ferris: So, what are your thoughts on the sort of low-code, no-code world? You’ve got your bubble, then units, those guys.
[37:10–37:21] David Hirschfeld: There’s a product that we use. It’s an offshore product called Continio that we have used in a few projects, and it’s really pretty amazing, and you can build anything with it.
[37:21–37:45] David Hirschfeld: So it’s not like Bubble where you’re really kind of limited in the kind of framework and design and functionality. You can build anything with it. But with now, and we’ve done a few projects with that, which we really like, But we’re finding that with AI now, you know, the need for the low-code stuff is much less required than it was.
[37:46–38:10] David Hirschfeld: So we’ve been doing a lot. This has been a big thing for us in the last six months, and in particular in the last few weeks, is using AI to help us develop systems in code. And then now we’ve been benchmarking the difference in terms of when we use AI and how we do it and documenting from a protocol perspective as a team how we accomplish this and generate the code.
[38:10–38:24] David Hirschfeld: And the benchmarks we’re getting are pretty amazing. So one of the things that we’re building a bunch of products internally right now. One of them is the niche analysis tool I was talking about because founders get really fatigued doing that.
[38:25–38:38] David Hirschfeld: So we’re building an AI model for doing that niche analysis and basically building up the scoring same methodology, but just where AI is deriving all this for you in a feedback iterative process.
[38:40–38:55] David Hirschfeld: We’re building and we’re using AI to develop that system, and it will be an AI system. Also, another one, which is kind of funny, is because I’m out in the market, I have a lot of referral partners, right?
[38:55–39:14] David Hirschfeld: That’s one of the reasons why I started podcasting was to basically find people in my world that I can refer projects to, that can refer projects to me, and then managing and tracking all the referral partners and the referral fees that they generate and all that becomes a pain.
[39:14–39:34] David Hirschfeld: It’s fine with five, three or four or five, but when you have 10 or 15 or 20, then all of a sudden it’s really hard because you don’t want somebody to refer a project to you and you forget who referred the project, or you don’t remember what the agreement was, or you’re not accruing referral fees once that client starts to become a paying client, right?
[39:34–39:48] David Hirschfeld: So we’re building a referral partner portal. It’s not an AI model, but we’re using AI to build the tool. I mean, we’re using Cogen AI tools to build it, and from scratch, all of it.
[39:48–40:06] David Hirschfeld: They add the UI, the back end, and we require that it’s built in a very scalable manner. So everything is containerized and microservices and serverless, which is how we build all our systems so that you can, you know, when we launch an MVP, it scales day one.
[40:06–40:21] David Hirschfeld: We didn’t used to do that because the overhead was too much, but now it’s just part of our DNA. But we’re building all that from AI, and it’s amazing how quickly we’re able to assemble this product and build it.
[40:22–40:35] David Hirschfeld: So we’re benchmarking this, and now I’m having to go to all my clients and update all of our estimates to our clients in terms of what it’s going to — some of the ones that we have proposals out to. I never had the benchmark, so I was always very conservative before.
[40:36–40:41] David Hirschfeld: I knew we were getting benefits, but I couldn’t really quantify it. Now we can quantify it. It’s been pretty striking.
[40:42–40:47] Ryan Ferris: So give me a feel. Is it 10 times faster? Is it 5 times faster? 100 times faster?
[40:47–41:14] David Hirschfeld: Okay, okay, okay. So with this app, and we’re still working out some of the bugs in the workflow and how this is done, in particular on the UI side, although we’re getting much better at how we control that and get AI to basically do the — I don’t know that we’re getting as much of a benefit on the UI part, in terms of just creating the screens and designs, is we have a really, really exceptional designer who’s very fast.
[41:15–41:39] David Hirschfeld: If that weren’t the case, then, yeah, we’re getting fantastic. Then we’d be getting great benefits. But on the back end side, so we developed the, for our MVP, we developed all of the server code, all the business rules and back end code in this serverless microservices containerized architecture, probably six days of effort.
[41:39–41:53] David Hirschfeld: The front end is probably taking us three or four days just because we’re fighting with the UI generation tools. In fact, that one I took over on myself because I want this UI to look a certain way.
[41:53–42:10] David Hirschfeld: And then it’ll take another day, day and a half to basically implement all the APIs for the front end to the back end. So we expect, we started this mid-last week, we expect on Monday we’ll have an MVP to start to test, Monday or Tuesday.
[42:11–42:23] David Hirschfeld: Wow. I mean, it’s like, what? Yeah, and it’s not a super sophisticated application, but it’s not a simple one either. I mean, it has a decent amount of functionality in the MVP. I can add referral partners.
[42:23–42:37] David Hirschfeld: I can add referrals. They can log in. They get their own dashboard. They can add referrals. They can see the status of the referrals as we update the status. They can see how much has been accrued, and then they can become a referral partner.
[42:38–42:58] David Hirschfeld: We don’t have — this won’t include the SaaS front end to it yet, but they can become a referral. I think they can sign up for their own account, and now they’re building their own referral partner network as well, right, and maybe giving referrals to some other account admin, right, so it can kind of grow on itself.
[42:59–43:01] David Hirschfeld: Anyway, so we’ll have — But now — And the reason — Yeah.
[43:02–43:03] Ryan Ferris: Yeah, I’m just curious.
[43:03–43:10] David Hirschfeld: The reason I did this, I couldn’t find a good product on the market that did this that wasn’t built for, like, big channel partners.
[43:11–43:11] : Yeah.
[43:11–43:23] Ryan Ferris: Yeah. Yeah, yeah. Exactly. I mean, because go back to the sort of AI thing. Are you using a well-known product, like a Copilot, for example, a March Copilot, or have you guys gone and built your own one using the libraries?
[43:24–43:43] David Hirschfeld: We’re using Cursor, which is an editor, a really good editor, that allows you to embed AI. You know, you have the AI tools, and the AI tools have the context of your application, so you can ask it questions and ask it to build something on top of something else, and it knows what that thing is.
[43:44–43:54] David Hirschfeld: Okay. Right? It’s really good. We use ChatGPT for generating a lot of the code, or we use OpenAI’s 4.0, you know, in cursor.
[43:55–44:09] David Hirschfeld: We do use both. We were using a product called Veo for generating the React front end, but it’s hard to guide. So I’m using Bolt, which is an application builder that’s UI.
[44:09–44:23] David Hirschfeld: I’ve used Bolt and another one called Replit, R-E-P-L-I-T, to try to build the entire app myself, just AI 100%, you know, build this app that does this functionality and then guide it through the process.
[44:23–44:37] David Hirschfeld: And there they write crappy code. and they get stuck in these mental loops that you can’t like, you know, it was working and now breaks, and it has the same error message on eight iterations.
[44:37–44:47] David Hirschfeld: You’re trying to fix it, and it won’t fix it. It’s just too frustrating. Plus, the code is terrible that they do. But Bolt does a nice job creating the UI.
[44:48–45:01] David Hirschfeld: So I’m using Bolt to basically create a click-through UI, and it will build the whole application, but I’m telling it, do not put any logic in it, just create the screens and the navigation.
[45:02–45:16] David Hirschfeld: And it’s creating a nice UI. It still takes a bit of guiding along and fixing, and then it starts to get off track. So then I’ve got to go do a cut and paste of something I want it to look like this, and I’d say make that look like this, and then it’ll start to get it right.
[45:17–45:26] David Hirschfeld: But, you know, I’m redoing the UI because we were trying to do it all with this VO product, VO.dev, by the way, V0.dev.
[45:26–45:27] : Yeah.
[45:27–45:39] David Hirschfeld: And my developer was getting really frustrated using it, so I said, let me just take this on. So I took it on yesterday. I’ll be done with the UI probably today and give it back to him.
[45:39–45:55] David Hirschfeld: Then I’m going to take that UI, try to get it, Figma, one of the Figma AI tools to generate it in Figma. Then I can take the Figma and put that in the V-O tool, and it will build the React front end, the total React front end that matched that Figma perfectly.
[45:56–46:07] David Hirschfeld: So this is kind of the process. And probably three, four days to get a completely React front end UI that’s well-designed, that supports all the functionality we want.
[46:07–46:19] David Hirschfeld: That plus the back end plus wiring the two together and pushing it into production. and also, you know, serverless using Lambda functions or Cloud Functions or whatever your serverless brand is.
[46:19–46:38] David Hirschfeld: This is — anyway, that was eye-opening for me that we can — we’re going to actually get a production application out. Then we’ll do the SaaS front-end, which will probably take another, you know, week because then you’ve got roles and permissions, sign-up and payment processing, but not a month, a week, right?
[46:38–46:52] David Hirschfeld: Now we have total control. This is our app. There’s nothing unknown about what we want to accomplish. It’s not AI in terms of having any kind of, like, AI intelligence that we’re trying to get a workflow in.
[46:52–47:03] David Hirschfeld: It’s very nuts and bolts referral partner management. But it’s a really pretty application. It’s got dashboard widgets and all the functionality you would expect it to have at a very basic level.
[47:04–47:23] David Hirschfeld: Later, I’ll integrate it with QuickBooks so that as we start to get payments from clients and it automatically feeds, you know, sends that payment transaction to the thing and then calculates fees so that with notifications to the referral partner, right, all the stuff that you would want in an app like that.
[47:23–47:38] Ryan Ferris: So we built value in Bubble and I wasn’t impressed in how we built it, et cetera, so we’re rebuilding it now. And a lot of the tools you’ve mentioned, the guy that I’m talking to is using a lot of them, and I’m very curious to see how it would go.
[47:39–47:50] Ryan Ferris: And maybe we can do this as a part two conversation. Oh, I’d love to. Sorry, I mean, because I’ve got enough for the first part. Because the part two for me is, you know, how do you see your teams changing?
[47:50–47:55] Ryan Ferris: How do you see your project approach changing now that you’ve got the same ability?
[47:56–48:05] David Hirschfeld: Dramatically. Dramatically. So the whole team structure is going to change dramatically. Yeah. Yeah. That would be a very good discussion, yeah. Yeah.
[48:05–48:11] Ryan Ferris: So if you don’t mind, we’ll just put this up if you want to just let people know how to get a hold of you. Oh, yeah.
[48:11–48:23] David Hirschfeld: Thank you. Yeah. Yeah. You can get me at Tekyz.com. If you’re watching the video, you can see Tekyz spelled over my head. If you’re listening to the podcast, then it’s T-E-K-Y-Z, Tekyz.com.
[48:24–48:43] David Hirschfeld: So you go there. You can get a hold of me by doing the contact us or reach me, david at Tekyz.com. Though, please don’t put the email in the podcast links, but anybody listening to this that made it to the end, please email me directly at david at techuse.com.
[48:43–48:48] Ryan Ferris: Will do. Perfect. Thanks so much. Great. Thanks, David. I look forward to chatting with you again.
[48:49–48:50] David Hirschfeld: Thank you, Ryan.
[48:50–49:03] Ryan Ferris: I appreciate it. Thank you for listening to our episode. I hope you enjoyed it. Please share it with your friends and colleagues. Help us grow our network. Please also check out our website at www.value.work and our product that’s on there.
[49:03–49:05] Ryan Ferris: It’s something we’ve worked really hard on and we hope to benefit too.
David Hirschfeld founded Tekyz, a company dedicated to transforming business software development. With over 30 years of experience, his journey began with a physics degree from UCLA and a successful sales career at Computer Associates. After launching and selling his first software company in 2000, David found his passion for empowering entrepreneurs.
He developed the Launch 1st™ methodology, which focuses on generating revenue before coding begins. This helps startups gain traction while minimizing risks. With a commitment to innovation and collaboration, David leads Tekyz in providing AI-powered development and SaaS solutions, making a meaningful impact in the tech world.
Tekyz is set to launch two new AI applications: one for automating the Launch 1st Methodology Niche Analysis and Estimiz, an AI-based project estimation tool. Outside of work, David enjoys golfing and woodworking.
You can learn more about David Hirschfeld and Tekyz by following his LinkedIn profile — David Hirschfeld LinkedIn Profile.
For more information about Tekyz’s services and how they can help you harness the power of AI in healthcare, visit tekyz.com or contact the founder directly at [email protected].
From Concepts to Code: The Tekyz Approach to Software Development with David Hirschfeld was originally published in Tekyz Blog on Medium, where people are continuing the conversation by highlighting and responding to this story.