Data-Driven Podcast: Creating High-Impact Teams & Validating Niches

Featured Image: The Data-Driven Podcast — Creating High-Impact Teams & Validating Niches

Summary

In this episode, David Hershfeld, CEO of Tekyz, discusses his unique Launch First methodology, which emphasizes the importance of validating product-market fit before building a product. He shares insights on niche analysis, customer engagement, and the significance of generating revenue early in the startup process. The conversation also touches on the role of exceptional teams, metrics for success, and the integration of AI in software development. David reflects on learning from failures and the iterative nature of startups, providing valuable advice for aspiring entrepreneurs.

Chapters

00:00 Introduction to Launch First Methodology
02:45 The Importance of Proving Product-Market Fit
05:56 Niche Analysis and Customer Engagement
08:48 Revenue Generation Before Product Development
11:51 Challenges in Startup Methodologies
14:48 The Role of Validation in Startup Success
18:10 Lessons from Past Experiences
21:03 The Future of Startup Methodologies
26:44 Startup Lessons and Learning from Failure
29:03 The Role of AI in Software Development
32:20 Building Exceptional Teams for Quality Software
36:39 Learning from Mistakes and Overcoming Obstacles
39:27 Metrics for Success in Software Projects
52:00 Conclusion and Key Takeaways

Introduction to Launch First Methodology (00:00)

Welcome to Data Driven. Today, Frank and Andy are joined by David Hershfeld, CEO and founder of Techies Corporation, a firm specialising in helping startups and saving troubled projects with his unique Launch First methodology. David brings us into his world of startups, where, believe it or not, he advocates for getting customers before you even build the product. He explains how Launch First combines high-fidelity prototypes, niche targeting,

and a lean, marketing-first approach to de-risk software launches and prove real demand. Whether you’re an engineer-turned-entrepreneur or just curious about the art and science of startups, buckle up for an enlightening episode with insights on maximising impact and proving product-market fit with speed and precision. Let’s dive in.

Well, hello and welcome to data driven the podcast where we explore the emergent fields of data science, artificial intelligence, and of course data engineering with me. As always is Andy Leonard. How’s it going Andy? Pretty good Frank. How are you doing? I’m doing well doing well. I just trained a large language model using the instruct lab methodology. So it was interesting experience. Took about 27 hours to compute.

on somebody else’s hardware and somebody else’s dime. So, you know, I’m sure if I were, I actually did try to run it on my local hardware that I have. I actually have a new Mac from work, which has, it’s an M3, 128 gigs of RAM. And one of the reasons why Macs are popular in this space is they share the system memory with the GPU.

So I effectively have 100, you 120 gig of GPU, which is pretty interesting. It definitely makes things a lot faster. Some of the Olamis stuff, if you run locally, is really snappy. It’s kind of nice. But again, so somebody else’s dime. So, you know, it’s always good when you get high end compute from somebody else. How about you? Yeah, I’m just peddling around here. I’ve been doing a little writing.

Cool. And threatening to… I’m actually working on a series on my blog about data engineering frameworks and fabric. Nice. Fabric’s kind of new. right now, well, I have a check today and fabric’s moving at that pace where you can kind of say that. They’re adding new things daily. But all of the pieces and parts that I want to use are not quite there yet. They’re probably coming because…

There’s something similar in Azure Data Factory. And you know, I have some friends on the team and I’ve let them know, hey, this problem I’m trying to solve and this what I think I need. So they’ll get to it. It’ll get there. They should have had a launch first or generate revenue first methodology, which is a good segue into our guest today is David Hirschfeld, founder and CEO of Techies Corporation.

And they have what they call a launch first methodology where software founders can achieve product market fit easily and quickly without the typical risks associated with software startups. It is primarily a software development company focused on startups and they love recovery projects. You know, the projects that are in trouble and you need a special forces team to pull them out.

Welcome to the show David. We can talk about that and kind of what your launch first methodology is. Sure hi Frank. Hi Andy. Thanks for having me today. No problem. So looking at your LinkedIn profile, you’ve been a. You’ve been a founder and CEO for awhile. Yeah, so so seems like you know what you’re talking about in terms of helping startups.

What inspired you with this launch for? Well, first off, what is the launch first methodology and what inspired you to make it? OK, well. I’ve been doing. Working primarily with startups, but not limited to startups, but primarily with startups for the last 17 years. And I’ve had my own startups even prior to that. I started a software company and logistics.

inventory management route distribution as sort of a lark in early 90s with a friend of mine. We were working at Texas Instruments and we started the software company and despite everything, every effort on both our parts, we ended up growing it anyway to to 800 customers in 22 countries and sold it in 2000. So.

Of course, now I’m thinking I’m a software startup genius, right? Because I had a successful startup and but I didn’t realize what we did right and what we did wrong and how much of it was just timing. I was VP of products for the company that acquired us for the next couple of years and then cast about and started techies a few years later. Working mostly with startups and I have worked with well. Probably close to 80 startups at this point.

in lots of different business domains and most of them, a few of them really successful, but most of them, the vast majority fail. And they all fail for the same reason. They just wait way too long to prove product market fit. And the only way to prove product market fit is by getting customers to pay money for your software. Because it doesn’t matter what they say about how great it is and they’ll buy it when it comes out.

If you don’t get them paying for it, you don’t have any proof that they really met them. watching all these companies fail for the same reasons, it dawned on me over a period of, know, sort of how peeling the onion started to realize the things that they were missing. And the primary thing was going out and basically starting the marketing engine first before building the software product. I know that sounds kind of in reverse.

but that’s the only way you know if you’ve got a product that the market wants is to go out and market it successfully. So the way we do that is we go through a really involved niche analysis process, which is also my own methodology, the way we figure out what that niche is and who the early adopters and how to speak to them, what their top two or three root level problems that they need solved are, what that value is for solving those problems.

Then we create a very animated set of mockups. We call it a high fidelity prototype. It looks like a real product when you demo it, people think you’ve actually built the product. And then we basically create a simple marketing stack, go out and market the product with some high value offer like lifetime license or something that somebody will look at and say, that’s a really high value. I don’t really want to miss out on that offer.

And this is a big problem. know it costs me a lot. So I know I’m going to buy this software when it comes out. So sure, why not? I’ll go ahead and buy into it. Now I’ve got to guarantee that if you don’t come out with the software in a certain period of time, you’ll give me the option to get a refund. So that’s all fair, right? And if you can sell enough product at high enough closing ratio at this point that you can prove that your lifetime value of your customer is going to be high enough to offset what your cost of acquisition is, you’ve proven you’ve got product market fit and a market. Now,

and you’re generating revenue to help you fund the development of the software. And if you want to raise money, raising money is a whole lot easier because you can show demand and customer acquisition. that’s the idea, start to generate revenue first, then build a product. So this sounds like it intersects with lean marketing, lean startups, I should say. It does. It definitely borrows concepts from lean startup with the exception of you actually

start to sell your product and generate revenue with it, which lean startup doesn’t, right? That’s the one piece I think they’re missing. A lean startup is all about testing, test, test, test, test, right? You have an assumption, don’t put any belief system to it. Go out and actually test the assumption and see if it’s true. So this just takes that to the, little bit farther, you know, with the, okay, you think there’s a market for it, let’s go out and test and see if people will spend money for the product.

Red I can’t blanket on that guy’s name, but it’s called million dollar weekend. No. It sounds not exactly the same, but he would basically find a way to raise money before he bills anything. Kind of like, Hey, would you put down any offers offers? He’s the founder of APSUMO among other things. And one of these things, it’s not quite exactly. I think your approach is a little more refined. I think he’s

more seat of the pants type guy, but he’ll basically send an email out to his friends. Like, would you pay X amount of dollars per month for this? If you’re interested, I’ll give you a super discount. Now give me like, you know, like one 10th that price. And then he basically raises the money. And of course, I’m sure he does math in his head to make sure he’ll break, you know, like money on the lower discount. And then, you know, if enough people sign up like, well, it, there’s definitely a market there.

Do you want to transform your business with AI? Send us a message to schedule a call!

Technically, he’s making money right away, right? Right. So this kind of plays off the crowdfunding model, right? So let’s say I have a market, I think it appeals to the masses, then I’m going to go after a crowdfunding model. But that doesn’t work if you’re going after a niche market, a B2B niche market, right? Because there’s not a, because they’re just not going to see your marketing. So that’s much more of a either a webinar or direct sales type of approach to do the same sort of thing.

And as you know, crowdfunding works when you have a really elaborate crowdfunding campaign, right? We’ve got a demo that looks like you’ve built the product and you’re showing something really sophisticated and and you’re and making it sound big. So you do the same sort of thing with launch. First, you just do it in either a one on one or a webinar fashion and and for us the big discount is well, let’s say.

Let’s say it’s going to be $99 a month subscription for an entry level, right? So if you pay two years of that subscription subscription price in advance, I’ll give you a lifetime license. And why would anybody want to give away a lifetime license? Because because you’re going to give away a an infinitesimal fraction of a percent of your market selling hundreds of lifetime licenses and generating thousands of dollars for each one.

As opposed to going out and finding and giving away 10, 15, 20, 25 % equity to raise the same amount of money.

Interesting. So you’re kind of Robin Peter to not Robin Peter to pay a Paul, but you kind of have you can kind of pick your poison, so to speak. Yeah, terrible analogies. Andy, help me out here, man. Those are actually what I what I love about it is, is a I’m an engineer. Okay. And Frank can back me up on this. And that means I’m lousy at accounting. And I’m meh at business. There are engineers who are great, great engineers, great at business. I’m not one of them. But

What I like about it is something that it sounds going to sound simple. You feel free to laugh, but it dawned on me that when I do things this time of year, like my November sale instead of Black Friday, I put everything on sale and I’ll have products that I normally get anywhere from $299 for a day of training up to $899. I’ll set it on 99 bucks and the numbers work out where I make a whole lot more money.

this month than any other month of the year. And the thing that dawned on me that again, feel free to laugh is when somebody pays me for my content or my time, I’m getting paid. Right. That’s what I heard when you said that. You two year, a two year license, two year, you know, two years at the subscription rate equals a lifetime license. I love that idea and I’ve sold some products and I’m going to give you credit for that, but I’m going to try that out see if it works.

Yeah. Yeah. And then if it’s something that is solving a quantifiable problem and it’s something they know they’re going to need, then they’ll buy it. and you’re not really giving much away when you think about the amount of what percentage of the market you’re, you’re losing in the process, right? Right. Because that’s preventing you from having to give any equity away in, the startup raising money. Yeah, that’s perfect.

I love that idea a lot. Yeah, thanks. All right. I’m going. See you. I’m kidding. I’m just taking a minute to get some more ideas from you, David, but I’m glad we kind of cleared that up early because I do that. I’m not, I promise I’m not trying to, you know, not trying to steal anything from you. You don’t take it that way, but it’s a real world example for me because if I knew about your service, I probably would have been talking to you before now. So now others, if you’re listening, if you’re like me, an engineer,

who’s maybe meh at business and marketing and all of this stuff, but a great engineer, and you just got a great idea and you wanna see how it goes, talk to David’s company, Techies, is that correct? Yeah, T-E-K-Y-Z.com. We’ll put the link in the show notes for you and I’m gonna go check it out and see what else I can learn. I appreciate that, thank you, thank you. Yeah, it’s the hardest part of…

the whole methodology, because there’s a lot involved when you actually do this in a formal way. The hardest part is the niche analysis piece, because, and it’s a methodology I created that is metrics driven. A lot of the startup methodologies, and many of them are good, the biggest problem I always have with them is that they’re, each step is kind of siloed, and it’s subjective when you’ve done that step.

completely and correctly, then you move on to the next step. I’ll give you an example. Some of them I think are misplaced. For example, discovery interviews. Discovery interviews, you take your idea, you break it down to all the features and the benefits and the value, and now you go out to the market and you interview potential customers on this idea of

here’s a product I’m planning on building or I’m thinking of building and here are the features that we’re providing and this is the value we think it’s gonna get. Would this be value to you? What’s the cost to you for doing those things? And there’s a couple problems with that. One is people don’t usually wanna be honest with you when you’re asking them questions like that. Not because they’re dishonest, but just because they’re trying to help you out. They may say some instead are,

positive sounding or they may not want to reveal a weakness in their business in some way or maybe they’re or maybe they really believe that this is a great idea but because they’re not doing any cost benefit analysis or they don’t have to justify the purchase right that’s right and so you get a lot of miss and you’re usually startups are asking people in lots of different niches while they’re doing this discovery interview phase

Not realizing that each niche has their own issues that are specific, so you’re not laser focusing the questions and the discovery around the problems in that niche and launch first does this really differently. There’s we break it down by by teasing out the root level problem statements across lots of different niches and we build this matrix and then we score. OK for this.

Here’s all the list of problems that we have identified across all the niches. Which one apply to this niche and what’s how much is the impact to that niche for that problem? Then we score this matrix so that it starts to become obvious. Which are the top three problems that impact each niche right out of the list of all the problems on potentially solve it and then we do the same matrix again and we say, OK, what’s the true cost of each of these problems for each of these niches?

What’s the perception that this is a problem and what’s the actual cost associated with the problem? And then we basically use a reporting technique to bubble chart these things and identify the ideal second, two or three top niches and the two or three problems they need solved now. That’s because it’s a high value and they perceive it as a high impact problem. And then we do deep dive, right? And in the process of doing that is you’re realizing you don’t.

understand each niche and their impact, you have to go out and you just sort of randomly start doing calls to two or three stakeholders in each niche instead of a formal discovery phase. It just happens naturally. I really like that. The organic field to that is very appealing to me. Yeah. And when you’re done, you have evidence of what you should be doing because the bubble chart produces a, you know, it’s a scoring algorithm instead of it saying, I feel like I’m done.

You know we can move on to the next phase of this. Then we do a deep dive to figure out which is the ideal niche and at the very end of that whole process. Then we do what we call a validator set of validating interviews. We have all of these assumptions. All of this information was collected. We’ve made a decision. We believe we understand the value. We believe we understand the messaging because we want that messaging to to touch on the what they feel is the impact of those top two or three problems. Then we go out.

And we talked to 15 to 20 stakeholders in that niche and find out whether or not we captured this information correctly or not. If not, then we drop that niche and go back to the second niche that came up in our list. Because one of those top two or three or four niches, you’ll get a 30 or 40 % affinity to the validation to. They’ll say yes, that’s problem kills me. It costs even more than what you’re.

saying it does because you probably didn’t factor these two things into it. I would, I think you should charge more because this is, know what you’re saying. You’re going to come out with it, but people will pay a lot. You know, that’s kind of what you’re looking for from the state when you really nail it. Yeah. Cause they know they’re not at risk. So you want to hear them say stuff like that, right? It’s, it’s very cool. And I, you know, just kind of rolling back a little into what you said earlier about how people, you know, unless

unless they’ve made a purchase. That’s how you know they like what you’ve shown them. they’re at least willing to bring it into their organization and give it a shot. They’ve spent money. And I think that’s a huge chasm crossing event when someone actually gives you money, pays you for your product. And before that, all of the reasons you listed are just so accurate. mean, people being polite.

And they’re trying to be helpful. think I don’t think I’ve never encountered anybody being malicious about it. I don’t think that’s part of the, know, if they’re willing to take the time to answer an email or get on a call with you, there’s some interest, right? But there’s no guarantee you’re to close, you know, if you do that. Yeah. And very plucks down the money. It’s all exactly. Exactly. Yeah, that’s exactly.

Have you guys heard of the mom test? The book the mom test I have. OK, it’s I think it’s my favorite business book of all time. It’s if you the problem with asking your mom about your new business idea is she wants to make you feel good and so she’s going to say, darling, that’s wonderful. I think you’ll be so successful, right? So you’re not going to get an honest answer from her.

because she either because you have a lot of deep seated problems in your family or because she’s a loving mother and wants to take care of you. Either way, you’re in trouble. So it’s called the mom test because if you learn how to ask questions in the right way, then even your mother will give you honest answers, not realizing that she necessarily even what the answers are for. Right. And it’s just a great book. And the guys

I can’t think of who the author is right now, but he’s had several very successful startups and it’s just a brilliant book. It’s a fun listen to if you like listening to books. we love audio books. That’ll come up later. Yeah, for sure. Yeah, I found it. And Rob Fitzpatrick, is that his name? Yeah, that’s it. Right.

Yeah, so the hardest part of launch first is the niche analysis piece because it’s tedious to do to do it properly. So that’s our one of the AI tools that we’re building right now is to basically do that niche analysis for you filling in the matrices primarily. Yeah. It sounds fascinating and I you know, I definitely definitely going to look into the product and and.

may end up chatting with you more about it. I love to. Yeah, I like talking about it. Right. Well, yeah. Well, and you know, you’ve been there, done that. So and I loved your in your introduction. I loved how you talked about, know, you had this first go and it went well and you sold the company that the stats on that you’re probably more aware of them than I am the statistics on doing that the very first time out of the gate coming from, especially it’s kind of a handicap.

Discover how our expertise in AI development can revolutionize your operations and elevate your business. Contact us today to get started!

coming from a large corporate America institution like Texas Instruments and then jumping into that unless you’re like in marketing or something like that. It’s just, it’s a hard thing. It’s a number of obstacles you guys had to hit. And you mentioned timing. I think that’s a very astute observation and you can get everything else right. And either wait too long or like you said, don’t get out there early enough with something.

and the game’s over before you even step on the field. Well, luckily, we didn’t know this industry that well. My brother-in-law was in it, which is why we thought we would build some software for him and maybe create a, not SaaS at the time, Windows 3.1 was brand new. All the products at the time were all Unix-based and expensive and green screen. So we thought, okay, maybe this will be

That’s why I say we kind of accidentally backed into the whole thing. That’s fine. This would be our first shot at at trying to do a software company just because we always wanted to do that. And and so we got a very simple product out fairly quickly and it was the first Windows product in this industry and that was and that proved that there was product market fit. Because we put some ads out.

Interactive Infographic: How to Determine A Product Market Fit

and people slowly started to buy it and started to pick up. Then we went to a trade show and people were lined up outside our booth at this really cheap booth. I mean, it was basically made out of PVC and we went down to Kinko’s and got a banner that was pushed out in the middle so you could see it from both sides. And we were in the back corner of this trade show.

We had fire marshals telling us we had to disband people because they were blocking the aisles. Wow. Good problem to have. Yeah, right. But product market said we we knew that people wanted it. We didn’t know it was going to be like that. Yeah, right, right. But still, like, you know, it was a good kind of, you know, test the waters. And I think that, you know, even if you are in marketing, just the the. The jump from.

You know, kind of a big corporate or I mean, that that’s impressive. My hat’s off to you. I don’t have a hat on right now, but I love to tell you how brilliant I am, but we really got very lucky. That’s I’m sure there was I’ll bet you there was more brilliance than you’re letting on. We interview a lot of people and, know, it’s it’s when we when we encounter folks who are being humble about it, usually it’s there’s more more to that story. So.

Not that the people that aren’t being humble. OK, never mind. I think everybody we talked to is pretty humble Frank. I think well, I just just to put a nail into that literally a nail several years later. Actually, this was about nine years ago. We I’m involved with another company called Anzu. We do health care related software and products for clinical testing and my partners.

plastic surgeon and we’ve been doing that for 14 years now. About nine or eight years ago, one of our patents, we decided to turn that into a social media product. And I, of course, I had a very successful startup before and I thought I, and this was going to be huge and Instagram had just come out and it was called Kludge. I just, and I did tons of.

demos to people who said this would be life changing. But we didn’t go out and start to sell it and create a monetization model at all. I just thought it would take off virally. That didn’t happen. We spent a lot of money on that product in a couple of years. So that followed much more the classic startup failure process. That was before I figured out Launch First. Otherwise, I would have done it very differently.

So like I said, it’s not about being brilliant. Pardon me. Well, but I mean, sometimes that’s what it takes. And classic learning from your mistake. You you realize that, you know, it’s one thing about software. I tell friends all the time that if you do software for a living, if you’re touching software, it is a very unforgiving feedback cycle that you go through. I develop.

behind the Teams window here we’re using to record this. I’ve got C Sharp open, Visual Studio and a C Sharp app open and setting on a break point. And if I step through this break point a few lines, it’s gonna hit an error, you know, fall from the tri block into a catch block. It’s not gonna, you know, tell me, you know, if you’d have written it this way, or, you know, it’s not gonna be mom to me by any stretch. It’s gonna, you know, it’s gonna flash at me and say, you fail.

And my favorite error message, you know, object reference, not set to an instance of an object. It’s not going to, you know, it’s not going to be all handsy and nice. It’s just going to say this didn’t work. And that’s important. And if you do this, I’ll have 50 years in this, David and may. I started with machine code, know, machine code when I was 11. And if you get to a point, if you stick around that long,

where you divorce the emotion from success or failure. It’s just like, that didn’t work. And you don’t even think about it. You just go and do the next thing. Yeah. Yeah. And I hear you, you know, walking through that same path with startups. And I love it. I admire that. Yeah. Thank you. I appreciate that. And hats off to you, 51 years doing this. thank you. Yeah. So now, now.

I’ve got to ask you since if you don’t mind, I ask a question here? AI code generation and also code assessment. How are you integrating that into your practice? Right now, I’m experimenting with both. I haven’t really done anything outside of the equivalent of commercial off the shelf, the OpenAI and some of the free models. Frank,

mentioned, I don’t know if it was when the recording started or not, but Frank’s tinkering with, you know, with, with the tools through his, the company he works for and the company that owns them that, that allow you to train these models in fractions of the time it was taken a year or two ago. But that’s it. have a kind of a glancing passing. There’s been a number of tools around for a while that, that were automated and they will do things like

you know, test automation and they’ll tell you they’ll actually generate tests for you that’ll catch all of the, you know, all of the errors that you don’t think about when you’re writing the code, you’re writing the code to work. You don’t think about all of the other things you’re not thinking about. was a little bit of a tautology there, but I should get a shirt. Frank, make us a shirt. I’m not thinking about the things I’m not thinking about. I love that. That’s the quotable for the show.

No, you know, I full disclosure. I worked for Red Hat and you know they are owned by IBM and they came out with this. This notion of using. It’s called lab, but basically it’s large scale line map for chatbots is basically the idea where you can give it a very small subset of data. And use synthetic data to kind of boost your input and then create.

more robust models on the other side and read at all up is doing other things where you know it’s basically like GitHub copilot but for Ansible for OpenShift for you know whatever one of the things that they showed us. Let’s say year ago now so like I’m sure it’s not you know too proprietary but the notion of what if what if you had a you know a command line terminal that was also assisting you with AI right? So if you do a lot of stuff with command line.

I rely way too much on copy paste, right? So I’m already kind of augmenting memory with that, but also what if it kind of assisted you with creating this type of thing? I think that’s, and also in my personal kind of, I have my own GitHub copilot that I use quite a bit. It’s very useful, right? Not because, not because I need it, but.

There’s only 24 hours a day. I’m going to build more stuff if I augment myself with AI. gosh, doesn’t make sense. And plus their content creation is largely driven by automation and different fragments of AI. And you asked that question though for a reason. David, you got sorry we got. no, was good. Your no no no yeah, I have because we’re. Trying to leverage it wherever we can in our practice, right?

to speed up development, to reduce the costs for founders, to improve the quality of the code. One of the things that I always stand for, and it’s really important to me is being exceptional in terms of the team, not just any one person. Because you can have an exceptional person, but if the team doesn’t work in an exceptional way, then you’re not consistently producing.

The kind of quality that you can just be really proud of from release to release to release. think sports is a great example of that, right? can have one rock star and team, but if they don’t. If you don’t have everybody kind of working at the same level, more or less right, you’re not going to the championship, right? And that’s why on my website it even says, you know, hyper exceptional software development team because there are artifacts that you produce if you’re an exceptional team that you don’t produce. If you’re.

A good team or you know the typical kind of team. The artifacts in terms of the level of detail, the transparency. The way that iteration is documented in, you know in agonizing kind of detail and all and the where compliance is just a natural process because it’s baked in things like that. Which.

Lot of team like I had a call from a friend of mine at somebody I met at a conference recently. He’s also a developer in the Midwest. And he asked me about fix. This is a good example and he wants to be. You know exceptional, but he’s not sure the best way to go about some of these things, so he has this fixed by his body. He’s doing software for an embedded system and now he’s getting bugs from the from the client.

that he has to fix and they’re saying you have to fix this because it’s part of the contract and he’s saying this wasn’t in the requirements and so he said, well, how do you deal with this? And I said, well, number one, it should you have known about that requirement and if so that you would have either asked or clarified or just naturally taking care of it. For example, if I’m doing a login screen, there has to be a forgot password link, right? They know somebody shouldn’t have to put that into the requirement.

I mean, of course you’re going to detail all that stuff out if you’re a good team. At some point it’ll get into the crap anyway. So he said that I said that’s your acid test. If there’s no way you could have known that it should have done that, and that’s why the defect is there, then it’s on them. If you should have known it’s on you and that’s a negotiation between your client and and I said, how did you test it? He said what was very hard to test because it’s hardware that we don’t have access to. And I said, did you build a test harness?

And he didn’t know what that was. And I yeah, right? And I said, well, how do you test APIs? He said he said, well, we usually have access to the API. So I said OK, so any application ever built in the future. This is you need to build a test artist when there’s some external application. That’s a pretty straightforward thing for experienced developers. He’s younger. He’s a really smart guy and his team seems to be pretty good, but they don’t have the experience to know.

There’s this thing out there you don’t have access to it, so you simulate, right? Yeah, that’s an example of that’s not even really exceptional, but that’s a quality team. Any quality teams going to be doing that, and it’s all of those things that inequality team would be doing it if they’re doing all of those things, then now you’re reaching a level of exceptional exceptionalism. That’s kind of how we look at it. Yeah, and it’s and we’re never good enough.

at it. So we’re always trying to figure out how to improve them. Yep, that’s that’s all excellent, David. And I’ve run into those situations myself. I mean, you know, over the years doing, know, doing so much software development, it’s it’s a lot like learning. mean, it is it’s learning from your mistakes and you don’t know what you don’t know. Right. That’s the hardest. You know, the hardest part is the unknown unknowns. And, you know, it’s even harder than the, you know, the unintended negative consequences.

So, but that’s how you learn. And, you know, hopefully your friend will pick up on that and go with it. You know, it’s just a time thing. And again, it’s one of those mistakes where if you want, you can beat yourself up over it. But everybody’s made that mistake. Steve Jobs went through that, Wozniak did, Elon did, everybody. And it’s just, you got to be determined to not stop. know, obstacle has to be defined as something to overcome.

That’s how you keep, it’s not about, to me, I don’t think it’s not about making the bajillion dollars, although that would be nice if it happens. Hasn’t happened yet. But it’s more about not letting stuff stop you. If you stay in the game long enough, you’ll win.

Right. If you stay at the casino table long while that’s a terrible. No, that’s a bad example. There’s a very good, but mean, know, you’re not in the games, Frank, I’m telling you, you know, I got, I got this deal, on coffee and it clearly is not strong enough. So David’s been the perfect guest with us so far. So he’s called us on, on bad analogies and that’s all good. Things always kind of work out for the best.

Yeah, right. think so. But you do definitely you should, you know, mark that coffee and never get that one again. Yeah, seriously. It’s it’s messing with your chain, man. Just put this path and put a couple of shots of espresso in the next. I feel like maybe maybe that’s what it was. It’s it’s nothing fancy like it’s McDonald’s coffee. Right. And like, if you order McDonald’s with the app, you get like a deal. And one of the deals was, you know, coffee. And I don’t think they added the espresso because I did ask for it.

And it’s clearly, clearly it’s not working whether, maybe they switched it out for decaf or whatever. But. have to go buy another. Well, I mean, there are other, know, as we’re recording this, McDonald’s is recovering from an E. coli thing. maybe like, maybe, maybe coffee, like, you know, the biohazard that is decaf is not top of mind right now. Wow.

So, no, this has been very useful because I think one of the things that, you obviously have a lot of data to back up your statements here, right? Because you’ve been doing this for a while. So what does that look like? What sorts of metrics do you track, right? Aside from income. Well, income’s a big one. Metrics in terms of like, you mean on projects, right? you’re about projects, you’re talking about business, startups, whatever.

Yeah, when I originally asked, I startups, but I mean projects. I think I think it’s something everyone can relate to. What are the metrics there? OK, so like we track and and I don’t know of other teams that do this because it takes you have to really bake it into your system and new people that come on have to be inculcated in your processes and procedures, right? So we track, you know, we do time sheets. We try to do all of our projects on a time material basis because.

fixed price is difficult to manage because then you end up spending more time managing the customer and the change controls and the requirements, although you can’t get away from it completely. So we always track everybody’s time. On a daily basis, we have time sheets and our own software that we built for tracking all this stuff, not that people should use their own software.

build their own there lot of time tracking tools, but we’re meticulous about this. So everybody on the team records the time that they spend and what they spent it on every single day. And it’s not a lot of detail, but enough detail so that we know what’s going on. And then we break those down by client and by project at the end of every month when we’re doing our invoicing so that we know if we’re doing three or four projects for a client, even if there’s sort of internal projects all associated with the same app.

We know where that time went and how much each person that was involved in that piece spent on it. And then we attached that along with the detailed time sheets to every invoice so that clients, because I hate being asked, you know, when we had blended rates, which was about 10 years ago, and they’d say, wait, you know, it’s blended rate, but wasn’t it all just testing this month? inexpensive people were working all month. I said, yeah, but.

That makes up for it. I just said, you’re right. I’m just going to do it based on each resource will have their own billing rate and it’ll be on the invoice. And then they’d say, well, that seems like an awful lot of hours for that one person in this month. And it was like, my God, I can’t get rid of these questions. Here’s the time sheets. You can audit us now. And now I never get that question ever because clients can audit them. Right. This is about

stepping into being exceptional. But to do that, you have to have a discipline team that’s recording everything every month and what project it was on, right? Which is not hard to do, but it’s just a daily discipline. The same thing with documentation. So right now, we have a client that I guess we were too successful with. It was a fairly large healthcare client. They got a big commitment from Google.

And, and Google’s agreed to spend a million dollars on converting their software into the Google cloud, but they want them to use, but they have to use a Google premiere partner to do that. And so now it’s, we’re handing this over to a Google premiere partner. I mean, very large premiere partner. right. So we have to make the handoff. The client felt really bad. I’m an investor in their company. So I look at it clinically and this is totally the right thing to do, but.

For many reasons, but. But we’re still working with them and on the product over a period of time as it transitions. We have to hand the documentation to Google right and our status reports up to the last nine months so that they can kind of follow the progress and all that. And we’ve had compliment after compliment from this really top premier team about the documentation is really good. It’s so detailed and our client was it.

in front of Google was saying, know, just let you know they’re very organized. You’ll have everything you need and and then the client asked me what he should be watching for with this new Google Premier team to make sure that there aren’t any. Any issues that that might indicate that they’re not handling things as well as we did. So that was really great feedback. That’s the kind of thing like I. That’s what I want.

from all my clients. That’s very cool. Yeah. Well, do you have, I know we’re coming up on time here for the interview. We want to be respectful of your time. Where can people find out more about techies and you, David? Okay. My LinkedIn, you can go to, we’ll put the link in the, right? We can put our LinkedIn in, right.

And techies.com, is spelled T E K Y Z.com. And from there you can learn about launch first. If you look at the menu, you’ll see a launch first logo at the top of the page. And if you click that, that takes you over to the launch first page and you can learn about launch first and how that works. Excellent. Yeah. And another episode where we had such a great conversation, we totally forgot our canned questions. So let us know.

It was too awesome. That’s what it was. was too awesome. It’s fine. Well, you guys are fun to talk to. thank you. Thank you. And this is like the low caffeine version of Frank. So you can just imagine. But this is why I look forward to robots making coffee, because it would be far less likely to skip skimp on the caffeine and the extra espresso shot. It would even tell you, hey, this is decaf. Do you sure you want it? Yeah, we’d probably get safety rails. Everyone’s talking about AI safety.

Right. That’s a big one right there. This is a bit of a non sequitur, but my my nine year old, who’s one of his chores around the house is emptying the dishwasher. Say asked me the other day saying he want explaining how he wants to get a Tesla robot for Christmas. And I’m like, why? He’s like, so we can clean up the kitchen and maybe even cook for us.

I like, I like you too. I’m like, that’s impressive. They’re cleaning up. My wife’s an excellent cook. So I can actually smart. I’m sorry. You were saying about wiping an excellent cook. Yeah, that’s okay. That is smart. I agree. David, finish your thought. No, that was my thought is that pretty smart for your nine year old to put that together. I was, I was impressed. I was like, cause I’m like, gee, that sounds like, cause like, you know, I was like, goes, well, I’m sure it’ll only cost a couple hundred dollars. And I’m like, I’m pretty sure it’s more than that.

And I think the target price now is like between 20,000 and 30,000, which as it, it, know that sounds expensive, but you think about the getting a maid or like any kind of Butler or anything like that. If it, if it can do all the things it says it can do, which again, bit of a stretch, not that outrageous. Yeah, right. Because Butler is top on my list.

Well, it would nice. It would be nice. It’d be nice to have it like, you know, make this food or, you know, empty the dishwasher. I mean, from him, it’s completely self-serving, which you don’t know. Which is the origin of all genius, right? Like is a bit self-serving, right? I always say the best programmers I’ve ever worked with are the ones that just love programming and are really lazy. They’re lazy because they don’t if they realize they have to do the same thing twice.

they find some way of encapsulating that functionality. That’s it. Yeah. I call that efficiency. I mean lazy when I say efficient. Well, my wife is efficient, but she would not end detailed, but she doesn’t have that lazy quality. So she would just be just as happy doing something very detailed and doing it right and doing it over and over again, as she would to try to, you know, if it’s really boring, then she’ll say, we need to figure out a way around this.

You know, but it’s more the CPA mindset versus these brilliant programmer mindset. Right. Yeah. That’s a perfect job for people who think like that. Absolutely detail oriented. Yep. Yep. That’s super cool. David, I love this conversation. I got a lot out of it and your stuff and connect with you on LinkedIn. And if nothing else, maybe we can converse from time to time in the LinkedIn chat. Just keep up with what’s going on. yeah, I

I’d love to, you should write a book. I don’t know if you’ve thought of that. You know, it’s come up in my mind many times and I just keep thinking, you know, a book, there’s so many books. Are people really reading all these books? Right. So, yeah. So I was thinking of a It’s crazy now, but Frank and I are both published and it, I mean, you don’t need like the published credit or anything like that on your resume. You’re doing okay. But it’s not just that. It’s, you know, being able to come up with a concept like,

you know, like the mom test or something like that. And you strike me as one of those minds that, you know, could put some information wrappers around so much like startup and, and software. And, know, once you, once you combine a couple of disciplines like that, I think it was Scott Adams that said, you’re not adding, you’re multiplying at that point and value. So it sounds like you’ve got even more, you know, more in the mix than that stuff we didn’t get to unfortunately, but.

I think it’s worth looking into. just, I don’t know, I’d read it. I appreciate it, thank you. I would totally listen to it. Do you do audiobooks? Yeah, audiobooks. I love audiobooks. Do you have a recommendation? Well, the mom test would be, that’s the recommendation. Mom test, okay. All right.

And if you aren’t already an audible subscriber, you can get a free book on us. If you go to the data driven book dot com or the data driven book, the either the or the either we made it. We made it so the pronunciation didn’t matter. But you can check it out. You get a free book on us. We get, you know, if you sign up for a subscription, you do that. I just looked the other day and this is I’m at almost 900 audible titles. Which I know, right. And I only got an audible account in 2017.

wow, that’s lot in that short of time. Frank lives in the suburbs around DC. So if he goes to the store, know, he can listen to half I have a lot of quality time. I live in Farmville, Virginia, okay? The other day it was crazy here we had six. Six people backed up at the stoplight. Pandemonium. I get half a chapter, he gets half a book. Well, I’m not going to tell you where I live. okay, that’s okay.

San Diego. okay. How’s the traffic out there? It’s really not that bad. I moved here a year and a half ago from Scottsdale. The traffic was worse in Phoenix than it is here, which surprised me. Yeah. Wow. But the weather here is just spectacular. That’s what I hear about San Diego in particular. Much better than like, LA or even the Bay area. It’s just, it’s just this pristine, awesome.

I actually have never been, I totally wanna go. well if you do, then you gotta let me know. Yeah, definitely, definitely, cool. I was on a Zoom, this was about five or six months ago with a whole bunch of people from around the world, and somebody was from the south of France, and he said, I’m from the south of France, I think, I don’t remember where exactly, and he said, we have the best weather in the world. And so I said, no, you have the second best weather in the world, I’m from San Diego, we have the, anyway.

And so we said, let’s look it up. we looked it up online. San Diego has the most moderate climate of any major city in the world. So it’s official. And I love the South of France. When I visited there, was the only place there was like within like an hour was like, I need to retire here.

Coolness. Well, awesome. we’ll make sure to include a link to your profile, your website, and a lot of cool stuff going on there. like your marketing funnel page. Very well done, by the way. Well, thank you. I appreciate that. And so with that, we’ll let the nice British lady end the show. And that wraps up another insightful episode of Data Driven.

Huge thanks to David Hirschfeld for sharing his experience and unique perspective on startup success through the Launch First methodology. It’s not every day you hear a founder suggest building demand before the product even exists and doing it with such precision and data-driven strategy. If you’re ready to bring your startup dreams to life or simply want to hear more from David, be sure to check out Techies Corporation and explore what they offer for ambitious founders. And remember, if you enjoyed today’s discussion,

Make sure to subscribe, leave a review, and share with your fellow data aficionados. As always, thanks for tuning in, until next time, keep being data driven.

Discover how AI can revolutionize your operations and drive business growth and success. Get started today!

Key Takeaways:

  • Launch First, Build Later: Focus on generating revenue and proving product-market fit before investing heavily in development.
  • Niche Targeting: Identify your ideal niche market and understand their specific pain points and value propositions.
  • High-Fidelity Prototypes: Create a compelling prototype that demonstrates the value of your product without full development.
  • Data-Driven Niche Analysis: Use metrics-driven approaches to identify the top problems and their impact within each niche.
  • Exceptional Software Development Teams: Build a team that prioritizes quality, transparency, and continuous improvement.

Ready to revolutionize your startup journey? Visit Tekyz at tekyz.com and learn more about the Launch First methodology. Connect with David on LinkedIn and join the conversation about building successful startups.

Want to learn about the Data-Driven Podcast: http://link.tekyz.com/datadrivenmediumpod

David Hirschfeld, Tekyz Founder

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].


Data-Driven Podcast: Creating High-Impact Teams & Validating Niches was originally published in Tekyz Blog on Medium, where people are continuing the conversation by highlighting and responding to this story.