top of page

163 | PAVEL SCHILLER | HOW TO AVOID EXPENSIVE DEVELOPMENT MISTAKES




"Information + thinking = action = results. Seek information, it will take you the furthest.

Pavel Schiller | CEO @ ASN PLUS s.r.o.


I have a few companies from both sides of the barricade among my clients. One would like to upgrade their pro21st century and equipped them with intelligent electronics. It doesn't matter if it's an electric motor, a modern fireplace or an airsoft gun. The other companies then develop and often sell just such things.


It's awful fun to stand on both sides of the barricade like that and listen to how the other is ... Fill in the rude word according to your mood and vocabulary.


Well, since there are more and more companies that would like to add some form of intelligence to their products or a new progressive business model based on accompanying services (and here intelligent electronics is a great thing), I decided to shorten their dead ends and save them a few bumps. I invited Pavel Schiller, CEO of ASN PLUS s.r.o., a custom electronics development company, to walk us through the process of an ideal collaboration. I've been researching ...


🔸 How to choose the right electronics development partner?

🔸 How long does a development project take?

🔸 What should the initial project brief look like?

🔸 How to manage the project budget within the given limits?

🔸 When is it better to build an internal development team?




 

HOW TO AVOID COSTLY DEVELOPMENT MISTAKES (INTERVIEW TRANSCRIPT)


Martin Hurych

Hello. I'm Martin Hurych and this is another Ignition. If you like Zazheh or you liked any of the episodes, so give a like or ideally a subscription. If you give a subscription, we'll both profit from it. You won't miss any more episodes and you'll help me navigate the social media algorithms even better, which will allow me to invite even more great guests like today's. Today's guest is the owner and head of the development company ASN Plus, Pavel Schiller, hi.


Pavel Schiller

Hi, Martin.


How important is adrenaline for Pavel?


Martin Hurych

Before we get into the company and developing bespoke electronics for customers, what did you arrive in, did you arrive in a go-kart or on a digger?


Pavel Schiller

I came by car.


Martin Hurych

Reading your hobbies and your comments, how much of a little boy is in you and how important is adrenaline to you?


Pavel Schiller

Adrenaline is important pretty much all the time. I've always enjoyed going to challenges and that's some adrenaline too. Now I don't get as much of it, or it's not the physical one, maybe more the mental one sometimes.

I definitely have that kid in me and I don't know how much it will ever go away, but I can feel the aging or the wisdom too. I think even with declining physical fitness, you don't just jump around anymore.


How much of an adrenaline rush is having ASN Plus?


Martin Hurych

How much of an adrenaline rush is having ASN Plus?


Pavel Schiller

For me it was quite a big adrenaline rush, I was used to those sports that you just have to work hard for 2 years according to some good practices and you are almost immediately on top or you have a feeling of success. In this business it's more of a marathon, so it takes longer and it takes more endurance. It's especially true in our business, which is electronics development, because Covid jumped in, so for a while customers stopped us because they were saving, they didn't know what was going to happen, then there was the war where the same phenomenon happened and then there were no chips. So there was a lot of adrenaline and you had a couple of times when the plane was going up against the wall.

We have now taken off and are facing more challenges around sustainability, growth and customer filtration.


Martin Hurych

So come tell us what you're doing.


Pavel Schiller

We do customized development of special electronics, related software and even those ecosystems on top of that. It's special because there's no point in developing a product that can be bought on the market, so there has to be a good motive to develop it. At the same time, it's not the mainstream stuff that Apple and Samsung and other companies like that can develop, or Bosch and Siemens of the more industrial stuff. We're mostly doing something special that doesn't exist and those markets are relatively small, but globally they're mostly big.


Martin Hurych

How many of you are there?


Pavel Schiller

There are 20 of us.


How to choose the right electronics development partner?


Martin Hurych

The reason you're here is this. I have observed for a relatively long time that a lot of people are trying to add some intelligence to their traditional products. Anything that is tangible and not on mobile is now considered backwards and unfashionable. But at the same time, I'm noticing that a lot of industries, specifically those coming out of the more traditional engineering fields, have development centers that can only do the mechanics. A lot of times people my age are afraid of the electronics, but they know they need it and they need to develop it externally somewhere or recruit their own people. So I thought I'd break it down with you, if I decide to go down that path, what's in store for me and what I should look out for. I have a company that wants to add intelligence to some of their stuff, like a fireplace, which is what you're doing right now. I have this crazy idea, I've never done this in my life and the first thing is I'm probably going to be looking for and picking a partner. What to watch out for at this stage so I don't fall for it?


Pavel Schiller

If I already know there's no boxed product I can buy for it, or I want to develop something of my own to differentiate myself, I would approach it more as a choice if I were building my own development team. To want to somehow offer it and price it at that point, you're not prepared for that and i t 's going to hold you back, and then at the same time you're probably putting yourself in the role of a supplier and a buyer. The collaboration will then be more about some haggling and negotiation than you focusing on the product or the project. Most of the time when someone starts out, the stories are that they get burned once or twice. Either they find an inappropriate partner, supplier, or they don't make the mindset that it's not like buying laser-burned sheet metal anymore, but it has more unknowns and variables to do some development.


Martin Hurych

I understand setting my mind like that, however, you must be going through some selection procedures. It would seem unnatural to me, if this is my first time, not to reach out to several companies. When you look at the other side of the barricade, what should I do to ideally not get burned once or twice before I finally get it and what should I look out for in a potential partner right from the start?


Pavel Schiller

You need to get a feel for these few companies. The important thing is to say that it's a lot of work, so you should allocate some capacity for it yourself, so I would not advise to choose more than 3 companies. Then because you might have 10, but you won't spend enough quality time with any of them to evaluate it. So feel free to investigate 10 companies, but expect to have the volume and capacity to do so.

After that, references are key and looking at what the company has done or how much it has done and then probably the human approach because it's about communication. A lot of technical companies can develop good technical stuff but you're not going to get along because you're not able to communicate and I think that's absolutely crucial. That's what I would put first, to sit down humanly, communicatively and understand each other because you're going to be working together for a long time, it's going to be intense and you have to enjoy it together and trust has to be maintained. That's where it always breaks down.


How long does a development project take?


Martin Hurych

So what do I imagine at this point in time to be some period of time that we will spend together on one project?


Pavel Schiller

It can be from six months to 3 years. Usually by then the products are already on the market and maintained, so the collaboration is usually a few years, but the most intense is around a year and maybe every week for an hour there is a regular meeting or online call.


What is the myth about development among companies?


Martin Hurych

What's the myth among the uninitiated in terms of getting started, finding companies, or even just taking them on, they'll figure it out and I don't have to worry about anything? Is there such a thing as a chimera?


Pavel Schiller

It still occurs to me to say that there is definitely not one single right way. I think the myth is the underestimation of how much energy and time they themselves will have to put into it, the underestimation of how much work it all is. Because if it's someone who's never been around electronics, development, manufacturing and software development, and is from an engineering background that's generally less abstract, they can't really imagine it. Even a layman can imagine how much work it must be to invent a shaft or a gear, because you can see it. So it's a total underestimation of the overall labor, time, duration, and how hard it is to correctly estimate or calculate how much labor it will be.

At the same time, there are many variables that will change over time, and you as a customer often want them to change. Because if you get it right, some section, some part, we'll do it, we'll deliver it, you touch it or your customer touches it and it changes. If we have it too fixed, ideally contractually with all sorts of crazy stuff, it's going to get terribly expensive because we're going to be negotiating all the time about what the change was, how much it was going to cost, why it was, and who caused it. That's going to make the project terribly expensive and slow it down. Of course, there are cases where there are giant projects, tens of millions and upwards, where it makes sense to really calculate it and make some sense and it makes sense to spend 100 hours negotiating for 10 million. But it doesn't make sense for a million, and it's all about time. Our customers ideally need to present the new thing at a trade show somewhere in January or February and if they do and they are first, being first in that market is often a huge advantage. Even if at that point it happens to cost you 200k more to develop than it needs to, it's small change compared to being first to market.


What should the initial project brief look like?


Martin Hurych

You said I'd define at least a little bit. How well do you define what I, as a customer, have to give you so that we don't get into dead ends?


Pavel Schiller

I think every customer has it a little different. Some people have the Word for it, some people want to talk, some people want to record a video. As a customer, I would pick what's closest to my heart and really put it into some tangible form as much as possible, even if it's a video, Word, document, diagram, depending on the technical skills that person has. For that reason, everybody has it differently. But that doesn't mean anything, that's input to us and then you and I together respond to that and we're already putting it into more technical forms like some diagrams and charts. We usually ask you a bunch of questions and you have to answer them or find out the answer and then only then is there some kind of brief that can be developed effectively. But it's still not the point by which you can tell exactly how many hours it's going to be. If someone wants it exact, they'll usually pay for it because we have to put a margin in there.


Martin Hurych

Does this mean that I just need some rough idea, or do I need to know some technical specifications and what it should do? What do I need to have lined up in my head before I send you the Word, what does the Word need to contain?


Pavel Schiller

I would probably give the advice not to worry about it, rather start talking about it as soon as possible. A lot of companies just want to get it well specified, spend 2 years on it, work on some super exact brief that they come to us with. But that doesn't add that much value compared to if they do a workshop or an online call with us and we only start to build something from that that is usable for those developers.

Whether you want to develop throttle control in a car or a shuttle is also a big influence. With a car, I think everyone can imagine and know that it's not going to be 1 and 0, but that it works in some sort of linear fashion. If it's a shuttle throttle pedal, which nobody knows how it works, you might write in the input that pressing the pedal means you're going to add throttle, and to the programmer that means it's a switch. That means 1 and 0, so on and off. We all know the car, so there's an assumption that we'll figure it out, but we often do things where we don't understand that use casu so perfectly. On the other hand, we have just the experience to ask them if it's supposed to be linear, exponential or if it's supposed to be off and on.


What is the next step of cooperation?


Martin Hurych

So together we somehow clarified what it was supposed to be, what it was supposed to do. What is the next step in our collaboration?


Pavel Schiller

For us specifically, the next step is that we agree when we start and then there is some more detailed analysis, which is paid for, if that makes sense. If it's not needed, it's not done. Sometimes a proof of concept is also done to check if it makes sense to do the project or not, if there is a blind spot. Then we start working together and we usually have a regular calendar with the customer cally every week for an hour. Sometimes we cancel when it's not needed, sometimes we call each other on an operational basis, but it's regular so it's not operational every week. Then every 14 days or every 4 weeks we do a report with what was done and invoice it at the same time. We do it in such a way that the customer is so closely part of that development team that it's as if they own that team.


Martin Hurych

Does that mean you're diving into his development team?


Pavel Schiller

If he has it or he has the mechanical one, it's possible and if he has no development, he dives to us, but we operate as transparently as possible. The best tool for transparency is also the invoice, as I like to say, it heals the relationship because it's terribly important for the customer to know how much what costs on a regular basis, to make good decisions. The minimum product and a dream vision that no one would pay for often differ. So it's good, even within that money, to make it as continuous as possible so that he can make the most informed decision with us. We have already done so many successful projects worldwide that we are able to advise them on what is good to develop when and to wait until the market pays for it.


How to manage the project budget within the given limits?


Martin Hurych

When you said it was worth 100 hours to negotiate a ten million dollar project, but you wouldn't do it on the small one, I could totally see the audience's pimples popping and hair standing up in horror. I get what you're saying, that often it's not clear where we're completely going, however I do have a budget for it. How do I, on your client's side, coach it so that I can get that whole project, even if it's not exactly described, in quality, in time, and in some kind of an outline budget so that I can show it at that show in February?


Pavel Schiller

I'm not saying they shouldn't negotiate, but let them put those other factors into that negotiating trinity, their time or their staff time, and also the value of time when they're going to be looking for a supplier for 2 years.Some companies have been looking for 5 years and still haven't found one and have been known amongst developers that we don't want to deal with them because we only deal with them and we want to develop. Act in proportion to the size of the job so it makes sense.

If I were in the role of a customer, and sometimes I am also in other services that are similarly harder to price, such as marketing, I would mainly try to be constructive. The moment they stop being constructive, they think of something but don't want to understand that it's going to delay or make the project or product more expensive, then trust is already falling and it's a problem. I would try to be constructive and see the real things that are going on so that that trust doesn't get broken unnecessarily because that usually makes it more expensive and delays it for a long time. Otherwise, with us, they have these estimates of deadlines and budgets, they usually correspond to x, y functionality and some things and they see that by being in communication with us on a weekly basis. Of course, sometimes unforeseen things happen, but at least they know about those in time. So it's wise to give yourself a year's headroom for that. Often someone will come in and want to develop something with certification to sell it in January, it's just that some things like certification processes and testing can't be accelerated even if you pay a billion crowns for it. I like to say that if I plant potatoes, they won't grow for a billion a month.


Martin Hurych

So that means that I should prepare myself for the fact that I'm buying the head and the resources rather than the final product, because if I'm buying the final product, we can price it in time and for some money.


Pavel Schiller

That's the big difference, I'm not buying a final product and I'm not buying a type of service like building a house or something like that. It's better to approach it as if I'm hiring my own development. That's also where realistically it would probably be different than what they envisioned in the beginning, and it's a misconception that development is too much work, so I'm going to throw all the development risk on some contractor. Realistically, it probably can, but you're going to pay for it anyway, or someone is going to under-shoot it and then you're going to haggle and get mad at each other. I'm fond of saying that it usually makes sense to develop things that have the potential to make hundreds of times that. So whether it costs a million or two shouldn't matter, and if it doesn't, it's not worth developing.


What can be requested in the project?


Martin Hurych

I said at the beginning that you develop electronics, but that's not quite it. So what do I get from you or what can I theoretically come to you with?


Pavel Schiller

Someone comes along who wants the whole ecosystem, so electronics, firmware, mobile app or some other app and the whole ecosystem. But it can also come from someone who has their own development. Typically in the Czech Republic there are older companies that have maybe 5 developers, and once upon a time electronics didn't have any software. Then there was a little bit of software and as time goes on there is more and more software and it communicates with each other and so on. So we often do work for companies that have in-house development and we only do some part of it and most often it's something software because they historically have more hardware developers. But sometimes a company will come in that's software again and they just want to do the hardware.


F**k -up or what is it not supposed to look like?


Martin Hurych

So basically, I get the intelligence of the thing from you right down to the final control. You also said at the beginning that one of the parameters of choosing the right partner is communication, you describe here how close you are. Come up with a fuck-up that went wrong because of communication?


Pavel Schiller

I don't know if it was a communication failure, but it was a Dutch customer, an interesting project. We often take customer references as well, typically when they leave someone and that supplier or their past partner was just bad. It's been a while since I've gone there and called and asked how they see it. We didn't do that here because I guess they've had some experience with it, we did it afterwards. In short it was a company that was mismanaged from our perspective. They had these huge visions and they ran out of money and then they were just pulling the project like a cat of kittens probably with the goal that within those bids and within the first billing and its due date something would develop and then they would move on. For us, we ended up with an unpaid invoice because we saw too much potential for the project and got talked out of a lot of things like the deposit, the length of the payment, the invoicing interval and so on.


How does Pavel choose his customers?


Martin Hurych

This is a good moment, let's turn this around for a second. You said you had an overhang of inquiries. When we were advising your customers how they should choose a partner, how do you choose a customer?


Pavel Schiller

The first is the humanity and communication. When we start talking about the project, it's pretty obvious right away if we're going to get along. So basically it doesn't matter what we develop as long as it's what we know and what we're good at. Most of the time they have a parameter that they've been burned somewhere, they've found out that it's bad somewhere, and they've had some experience with it and they're not naive. Those are probably our best customers. Then, of course, we pick it based on what they want to do and ideally if we're good at it and ideally if it's the whole ecosystem. Lately, we've been enjoying bringing things to life that never had electronics in them and helping those customers be market leaders or make such a competitive ploy that we feel good about it afterwards.


Martin Hurych

I have a great idea, I haven't had a chance to get burned anywhere yet, I like you a lot, we sat down humanly. Are you gonna marry me, or is the fact that I've been burned somewhere and the assumption that I'll know how the collaboration works important?


Pavel Schiller

If there was someone next door who got burned and was just as likable and with all the other identical parameters and we were forced to choose, he'd probably get the edge. But otherwise we'll take, and we're good at that too, because we have the experience and we know what we have to do

up front and repeatedly warn. By the time everything is worked out, which is a difficult phase, it's just an investment, it hasn't made that customer a penny. It's been maybe 2 years, so he's looking forward to it turning around to the point where it's cash flow positive for him. He's used to that manufacturing and he's used to maybe from that engineering where he sends a 3D model to some CNC, he'll probably send a drawing to go with it and the same shaft will be coming off that CNC for maybe 10 years or more. It's more complicated with electronics. There is some engineering around keeping that production going versus maybe somewhere there is none and that has to be accounted for as well. We're already telling them up front to account for that in their business model and when we get to that stage, not to say we designed it wrong when there's always something that has to be addressed. It's being dealt with because somebody stops making parts, for example. After all, it's a thousand parts on some board and there could be a thousand variations of why there's a failure and that manufacturing is not able to figure out why they're making the crap that's happened. We have to deal with all of that. The other thing is that until you produce thousands or tens of thousands of things, it's not a run-of-the-mill production. With 100 pieces of shaft, I guess you can say we can make a million of them already, but with electronics, it takes a little bit more gradual.


When would you rather build an internal development team?


Martin Hurych

Now you've brought me to a devil's advocate question. I once heard a famous interview with the head of Ford about why there was such a big difference between American and European classic car companies and Tesla a few years ago. He was describing it as Ford actually outsourcing everything and

he has no rights to anything. If they need to change something, he mentioned the car seat, they have to beg the supplier to change it. Then he converted it to this software, which is actually impossible to write because you have a bunch of components that are incompatible and you don't have rights over them, whereas Tesla is a computer on wheels. So shouldn't I rather ditch you and build it in- house since it's a core competency?


Pavel Schiller

That's a great one, because most of the time it's the same cases where the customers are unhappy and they come to us. They probably had the business model set up in such a way that maybe they don't have the right to do that, they probably didn't pay for the development and then they bought the finished solution for a fixed price. They probably had a budget for how much it was going to cost, so they were buying a ready-made thing that somebody had developed. Dissolve the development into the price of the products as well and then you're really, like, buying a finished product, but at the same time you're folding your own out of it, which you don't really want to do. Making a living just developing for an hourly rate is tough, it's not some great business compared to having a scalable product. So a lot of companies try to do it in a way where we'll develop it for free and sell you pieces for a clear price and you pay us for them. But then what happens to Ford, they go off and look for somebody else or they start their own development.

In our country, it works absolutely transparently, even in the fact that they have the data and the intellectual property. If they stop liking us, they can go anywhere else, they don't have to start from scratch, they have the source data. That's what's worked best for us and it's actually saved those

relationships several times. If someone was disappointed about something, that things happened that maybe no one could have known, I said I totally understand, we're not going to hold them back and we're not going to hurt them because they paid us to do what we did.

That's exactly how it breaks down here in small companies, and it usually comes down to the customers' need to know how much I'm going to pay for each piece, I don't want to pay for development and then production. At the same time, it's the need of those development companies to have a scalable business. We've found that scalability in that we develop it for customers, it's all theirs, they can pack it up and walk away at any time. Most of them who don't have their own development or manufacturing or don't have it sort of established and don't know the electronics, they leave the manufacturing to us. We provide them with the manufacturing, we supply them with the finished pieces and we get some small percentage of that because we don't dilute the development. That's the transparency for that efficient decision making and also the freedom of not being held up by one of the suppliers.


Martin Hurych

So is this something that Radek Alenka here once called Team as a Service, an internal team that sits somewhere else?


Pavel Schiller

That's right, that's the model that works best for us. It's called maybe point shopping, but that's for the individual, we don't do that, we almost do team point shopping. For a lot of companies it's also advantageous because even for the busiest part, when they're developing the most, those people who need 3 specialists or 4, they wouldn't hire them and they wouldn't have an experienced architect and designer there yet. They wouldn't even hire all of them, so if they did hire them, it would cost them maybe similar to what it costs us. Because we charge them for the actual work they do, and then when the project is at that later stage, when it's just lightly upgraded or maintained, it's awfully handy to have that development team. But you don't want to pay them, so with us they pay less, just for the hours, and if they buy the products from us as well, we get some fees from that, which gives us some stable footing and scalability. Of course we treat it kind of like an SLA, like a software service contract, except we tie it to those manufactured pieces because it's more customer friendly. It's proportional to how much they sell, that's how much they pay. It's a no-go for some people, which we don't mind either, but of course we're most comfortable with those who want to not worry, want to give it to us and trust us.


What to look out for when moving into production?


Martin Hurych

What have I forgotten in this cycle of collaboration?


Pavel Schiller

Then there is the production, or the introduction into production, and in between there is often some certification according to standards, where the equipment goes. We cooperate on this at least, and for those customers who are not from the industry, we handle it. Once it's certified, tested, it goes

into mass production.

You have to take into account that the mass production takes some time, so you can only figure out some things when you make 100 or a thousand of them at a time and then maintain it somehow.


How do I insure against supplier defection?


Martin Hurych

You mentioned the SLA. If I'm your customer, what should I demand at this point that you don't run off somewhere else and tell me I'm already a small uninteresting customer to you and that you're baking on me?


Pavel Schiller

I think that's one of the factors of what's going on, that's the risk of hiring companies like us. I think the main reason why those firms then ditch the smaller clients is not because they're small, but because they're unfortunately annoying. They feel they've paid enough and then it should all be free. I think we need to acknowledge that it's quite a live thing, that some of those engineers and developers' hours of work around it will be there and if I accept that, it will stop annoying me. The contractor or partner then gets a sense from me that I'm okay with it, and it's not like I'm bugging them with three-hour jobs that are crushing three-thousand-dollar jobs. On top of that, I'm pissed at him and I'm not paying his invoices.

If even that doesn't work, it may very well be that the company has grown astronomically and is no longer delivering. I don't think that risk is so bad to take because it's very unlikely and if I have the source data, I can go anywhere else. Plus, if I behave nicely, they'll be happy to refer me to a smaller friendly firm for whom I'll be a good customer. If I'm not a good customer, they won't even recommend me to anyone and just want to get rid of me a little bit. I try to be a good customer for those key suppliers too.


Summary


Martin Hurych

If there were a few sentences left in the information noise from this podcast, 3, 5, what would they be?


Pavel Schiller

I really like the thing that has moved me the most in my life. It's that information + thinking = actions and actions create those results. For example, if something in your life isn't going well or isn't going effectively enough, the information is the problem because you can't really control the thinking. So look for

quality information and that will help you in that development as well, because I think a lot of times people don't have that information or don't even want to know. The other thing is that I think relationships and trust are important for the business and you need to nurture those relationships. It's not that only one side has to try and if that trust ends, don't do it together because it's just going to be misery. It's maybe even a little bit related, because when people have the same information, then they get along.


Martin Hurych

It's true, it's all about communication. Thank you, Paul, for coming in and I wish you can ditch those little annoying clients because you have a hockey stick in front of you and you're growing rapidly. Fingers crossed.


Pavel Schiller

Thank you, too.


Martin Hurych

You see, we've peeked under the hood of external development. If we've uncovered a part of a slightly different bubble, if we've brought you a thought, an idea, then Pavel and I have done our job well. In that case, don't forget ideally to subscribe, because it is the most valuable for me. Anyway, it will also help if you like or if you forward this episode to a friend, a friend, a colleague, a colleague who is just thinking about some development or whatever else we are talking about here at Zážeh.

Be sure to check out www.martinhurych.com/zazeh, where Paul has already prepared and has a bonus hanging somewhere that summarizes the course of an ideal collaboration between the two parties. I can't help but keep my fingers crossed and wish you success, thanks.


(automatically transcribed by Beey.io, translated by DeepL.com, edited and shortened)


bottom of page