Ep 180: Ervins Grinfelds
How a Bootstrapped Startup With 0 VC Money Grew to Service Zoom & Microsoft. The Importance of Software Testing, Skype Mafia & What Matters to Enterprise Clients
Ervins Grinfelds is the Co-founder of TestDevLab, a leading quality assurance and software testing service provider that works with global tech giants such as Microsoft, Zoom, and Twilio. TestDevLab has grown from a small startup to a company with more than 500 employees, all without external funding. Ervins is passionate about ensuring that software products meet the highest quality standards, and he brings invaluable insights into the importance of trust, scaling a service business, and the role of quality assurance in the tech ecosystem.
On this episode we talk about:
Why founders often only turn to QA when s%&t hits the fan
How TestDevLab expanded from a small team to a company with over 500 employees and global clients
The balance between people-driven services and product-based scalability
How to ensure trust and reliability
How TestDevLab has managed to remain profitable and grow without external investment.
We are on YouTube and Linkedin as well
Watch select full-length episodes on our YouTube channel > https://www.youtube.com/channel/UCP6ueaLnjS-CQfrMCm2EoTA
Connect with us on Linkedin > https://www.linkedin.com/company/pursuit-of-scrappiness/
Read the full episode transcript below
Uldis (00:02.458)
Hello, hello, hello, dear listeners. Welcome to another episode of the Pursuit of Scrappiness podcast. Whether you're building a business, running a team or just starting out in your career, we are here to bring you scrappy and actionable insights to help you become more productive. My name is Uldis Teraudkalns and my co-host is Janis Zeps. Hi, sir.
Janis Zeps (00:21.877)
Hey everyone.
Uldis (00:24.164)
Before we start, quick reminder, follow us on YouTube, Spotify and Apple. It helps more than you know. And in exchange to that, you will find almost 180 episodes there covering all topics to help you become a scrappier and better version of yourself in life and in business. Plenty to explore. Go subscribe and be the first one to know when we come out every Tuesday morning. And great news for the video lovers. As I said, you can now see us on YouTube.
So go on the pursuit of scrappiness on YouTube and get immersed into the world of scrappiness. About today's topic. We are kind of continuing our half intentional series of having different domain experts talking about different valuable topics to help you build. And today we want to talk to a founder who has built a formidable business.
by helping companies produce higher quality software. So please welcome Ervins Grinfelds. Hi, Ervins.
Ervins (01:29.609)
Thanks for inviting me. Happy to be here.
Uldis (01:33.218)
Yeah, so Ervins is the co-founder of TestDevLab. It's a quality assurance and testing service provider out of Riga, Latvia originally that has been on the market for more than 10 years and has accumulated an impressive client base, including the likes of Microsoft, Zoom, Discord, Pinterest, Twilio, and many more household names. So the company has not taken outside investment as far as we know.
and its revenue is north of 20 million euros. I don't know. That's like a few years ago. I don't even know where it's now. I couldn't find the data, but doing quite well with the Bootstrap model. So what's behind the business? So let's dive in. And the first question that I wanted to ask was, what should founders and maybe non-technical business leaders actually know about quality assurance in software development and why it's important?
Ervins (02:28.204)
Yeah, that's a good question.
Let me tell that many companies with whom we work, quite many founders or non-technical business people engage with us only when they see the consequences of lack of testing. No, that's true. And what it means, in many cases, it could be poor customer reviews for their product. It could be maybe frequent outages, what they see. It could be maybe security problems, they face. Sometimes it even could be legal
consequences if it's not designed properly like many situations and that situation when many companies engage with us. mean obviously it's maybe a bit too late but that's when you know people realize that it's kind of that they need somebody to help with software quality assurance. So that's one case. But the second example also is that they're saying that
there is a huge difference between good product and great product. And that's another aspect where we can help. Sometimes it's just not enough that product works. It needs to work perfectly on multiple different devices, on different network configurations. It has to be perfectly designed, let's say even from accessibility perspective or many other aspects. And those small details makes a difference, from good product to great product. So to give kind of some
summary, know, like one reason why testing is needed is to avoid huge consequences, be it security breaches, be it outages, be it very poor customer reviews. And second reason is to make differentiating factor from good to great product. So that's simply good.
Janis Zeps (04:17.549)
One thing you mentioned actually about this one, can totally believe that when people look for help after something has happened, not before, and even accessibility, like if you don't meet the standards of the industry, mean, like for example, you operate in US, it also can like open you up for even litigation in the worst case, if you have this. So like all these things you don't think about when you're just starting to build a business, but all these things that when you grow become more and more important, and especially like there's
also being in product is also there's this bias always that you know people who work in the team have certain set of phones and wherever very often iPhones or newest Samsung's and all works great but but her customers maybe have different phones and so yeah interesting business that that we don't think enough about anything
Ervins (05:06.986)
No, exactly. actually, kind of this kind of access to wide variety of phones is also something that we accumulated over the years. At the moment, we have more than 4,000 devices in our office. And it actually looks physically like that. know, like if we go to different storage spaces, we open up racks with devices and you can see hundreds and thousands of different phones, different kind of manufacturers, different operating systems, different screen sizes.
But imagine, know, like if product has millions or even billions of users, it's so important that we can test on exactly those devices, on exactly those operating systems, configurations, what potentially end users might have. And sometimes it may look, you know, like, hey, what's the reason to test on, know, iPhone 7 on all the operating system, but some specific product maybe have hundreds of thousands of users, you know, like, and if suddenly
some specific feature or product is not working at all on that phone, it could have huge consequences. Maybe it doesn't sound relevant for fresh company startup, know, like it's very small user base, but as the company grows, it's kind of user base, you know, becomes quite large. Those details kind of matter a lot. And then as a point, he also kind of spot on mentioned around accessibility, you know, like not only, mean, of course, on one hand, it's
right thing to do, know, like so many people can properly use those digital products. But on the other hand, there is digital accessibility act and there might be even legal consequences for many products companies if they are not designing properly their products. So that's another kind of angle to look at.
Uldis (06:59.532)
I can definitely attest from my personal experience you don't need a big number of users but if there is something that doesn't work on an obscure device in a corner case you will definitely have some user who has that case and they're gonna complain and then clog up your customer support and annoy your team. It's bound to happen especially if you have something that relates to using the camera on the phone. I think that is the
the place where you can really go into big trouble and really need some proper testing.
Ervins (07:35.264)
Yeah. And sometimes you like to say that our potential customers are companies who have something to lose. And by that we mean, you know, company could lose reputation, you know, like if they release bad quality product, it could be money, you know, like or legal consequences, you know, like if there's some, let's say, you know, like especially e-commerce sites, they do some active product sales and if there's outage of system,
You know, like of course there's kind of significant losses for the company. So once companies mature enough that they have something to lose, be it reputation, be it money, be it any kind of legal potential consequences, it's definitely a time to thoroughly think, you know, like what's their kind of testing processes, do they kind of test thoroughly their product, are they insured, you know, like against possible incidents.
Janis Zeps (08:09.057)
Mm-hmm.
Janis Zeps (08:30.069)
Good point, Black Friday coming up for lot of brands, Christmas or time after Christmas is the peak season. Any interruption there, it's just going to cost you so much that this, I don't know, extra money that you spend on extra QA will look minuscule. Makes sense.
Ervins (08:51.348)
Exactly. Sometimes, know, like there's an analogy, you know, like this testing is an insurance, you know, like you invest some amount of resources, money to make sure that, you know, like you properly test those products or kind of establish right kind of testing practices. And that significantly can mitigate the risks of potential bad outcomes.
And it takes some maturity for companies to understand it. And as I kind of told initially, sometimes companies realize it through pain, like once they kind of encountered some kind of problems. But some companies already know it and establish right kind of testing practices from beginning building products.
Uldis (09:35.746)
Okay but... Teams...
usually are small in the beginning in maybe in except for some some exceptional cases. So what kind of I don't know ways for for scrappy small teams to ensure a decent QA level and when when their engineers are few and the money is scarce and maybe maybe AI can help somehow here how to how to make sure that that you know you you get get a good result and not break the bank.
Ervins (09:40.94)
Yep.
Ervins (10:08.426)
Yeah, that makes sense. And I completely agree that for very small teams, it's probably not wise immediately to, let's say, get external testing team. It's not maybe most rational for very kind of early startup or product team. But there's a couple of things that I would recommend. Number one definitely is invest in training, just to get a bit of external help about QIM.
mindset and somebody you know externally could easily kind of explain for example some good practices like peer testing kind of come up with approach that even if there's no testers in a team maybe kind of one developer could properly review others code also do some black box testing also explain about risk-based testing kind of strategies meaning that to map what are most important use cases with highest
impact and at least to test those to make sure there's no kind of severe consequences. It could be also, you know, like making sure to establish testing as a discipline, even with developers, product owners or designers, sometimes just, you know, like engineers or other kind of members in a team don't like to do it, but it...
makes sense even to spend at least some time of testing the code or products that you're building. And about other part about AI, you to be honest, you know, that's a good question. And then we get asked it quite a lot.
I mean, there's no simple answer, but when it comes to testing, we don't see that AI can help more comparing to any other kind of discipline. AI is kind of good help also for engineering. Developers can get good help using AI tools. The same is testing. Of course, you can help to produce some test cases, scenarios, give some tips, but it's nowhere near to replace actual testing or actual testing on physical devices.
Ervins (12:19.58)
And one of the kind of key reasons why this like that because I mean it's in
In most cases, you know, products, they don't have any kind of good documentation, no good description. What is the exact behavior? Frequently product owners, founders, designers or engineers only in discussions with this testing team, we can get a sense what exactly is intended behavior. And it's, it's, it can't be done, at least not at this moment with any AI tools. They can be kind of great assistance.
for testers, but there's no tools that can replace good quality testing done by humans.
Janis Zeps (13:06.461)
You know what is like one thing you can never replace. It's like a CEO or highest level manager finding a bug and the new software that you can like nobody spotted but like it's always gonna happen to the main guys. That's like, yeah.
Ervins (13:22.796)
Hmm
Uldis (13:24.484)
That's something that I found as building products that once you step into this user mode, you suddenly become really stupid.
You know, might be building yourself some kind of really sophisticated platform and web service, but as soon as you become a user, suddenly you cannot find the right button, suddenly nothing works, suddenly... All of this experience building doesn't really save you from then being a clumsy user being stuck in all kinds of issues.
Ervins (13:40.908)
Mm-hmm.
Ervins (13:45.546)
Mm-hmm.
Ervins (13:56.235)
Hmm.
Uldis (13:59.98)
So yeah, let's talk a bit more about your business. So it's kind of selling a combination of software and human service. Maybe you can explain a bit more how it works, this combination. And I was just wondering how does this model scale? Is it easy to scale? And what challenges do you face or have you faced since you stepped on this journey and grew the business?
Ervins (14:27.498)
Yeah, so I'll start in a way how we typically pitch ourselves to existing and new customers. We like to say that we are one stop shop when it comes to providing different quality assurance services, products and different innovative laboratories. By services we mean be it manual test, automated testing, security testing, usability, accessibility, audio video, you name it essentially one stop shop for all different testing needs by lab
We mean we have multiple innovative laboratories in our offices where we can do, for example, network conditioning. We can emulate 2G, 3G, 5G networks. We can record any specific scenarios like airport network, network in different cafes, play it back in our laboratories to see how it behaves. have GPS emulation laboratory where we can actually kind of emulate real life GPS signals and test products.
that can apply some GPS signals. We have laboratories for battery and performance testing where we can see precisely what is the impact on battery life by some specific product. We even have chambers where we can change temperature, which sounds maybe a bit wide, but it's actually quite frequent request by many companies to see how application behaves. Let's say phone is in cold temperatures outside. What happens in
gives us kind of warmer temperatures so we can do that as well. that's a lot. And by laboratories, the last thing is audio video testing laboratories. Also specific rooms where we can do echo cancellation testing, noise suppression testing, and by video also dedicated rooms where we can test relatively video quality. In specific, we call it black rooms with now also lighting where we can relatively test what is video quality in different
network conditions, different devices, parameters. So that's the laboratory part. And as a sidebar, the products, we have multiple different products, like one is called load arrow, which is a source tool for load and performance testing. We also have invented many algorithms on our own for predicting mostly audio and video quality scores. One algorithm which are quite proud of is algorithm
Ervins (16:57.263)
that is trained to be precise.
score of video quality. Let's say in this video call we are at the moment where we are at the moment with our trained algorithm it could reliably say that video quality score is for example 4.5 and it will describe also with high detail why it rated it like that. For example it could describe because it had three freezes for one second and at some point video degraded
from HD to 360p and many other maybe can description so it gives very reliable and accurate information of the difficulties. That's...
Ervins (17:42.636)
algorithm solutions that we created. And question about scaling is a really good one. Of course, services part is purely dependent on people, on engineers, and that's quite hard. You know, like it goes down to resource management. That's the reason why we have opened offices not only in Latvia, in six cities, but also across all Baltics. So we have office in Tartu, Estonia, we have office in Vilnius.
Lithuania also we opened an office in Skopje, North Macedonia with around 80 people and we have office also in Malaga, Spain and that's the challenge of scaling team because the more demand we have on services part, the scaling happens by getting new talent, by getting new people and opening kind of new offices or extending existing one.
Of course, better situation is with product scaling, like products like Lodero, which is a SaaS-based load testing product, or our video quality detection algorithm. Those are more like SaaS products and tools which can scale way more easier from business point of view. And of course, from a business perspective, we are...
putting, let's say kind of quite high priority on product part to see how we can build them even better, how we can scale them even faster as it allows much better kind of scaling perspective comparing to services part. So that's a short answer.
Uldis (19:25.796)
How do you determine where to go, where to open a new office? You mentioned North Macedonia, maybe not a clear answer why, right? It immediately comes to mind.
Ervins (19:42.368)
Yeah, yeah.
It's a good question. mean, it's actually was an interesting story we had from Spain. had one girl, a really good sort of cult engineer. She was living and working in Spain, but originally she was from North Macedonia. And I think it was back in 2020, we were growing super fast and it was just not impossible to hire people in Latvia at that pace.
that we needed. And we started to explore different countries, looked where we can open office, where we could do some hiring. And then our employee from Spain, but originally, as I told, she was from North Macedonia, she told, why don't you try North Macedonia? She pitched to us that they are really smart engineers, good communication, good technical skills, culture-wise similar to us.
So saying that we thought, okay, why not to try? know, like initially we are not 100 % convinced, but we started to put advertisements, you know, like seeking for quality engineers and engineers in general and putting some ads, you know, like in a local market. And to our surprise, number one, there were a lot of applications comparing, let's say to Baltics.
you know, like typically, you know, like if you look for especially kind of experienced software engineers or quality engineers, you don't get very many kind of applications. So number one, a lot of applications and number two, quality of people who applied were really great. I mean, most interviews we had were really decent ones with well English, good technical skills, good motivation, good attitude. And we immediately kind of
Ervins (21:43.222)
got the sense that it's good market to hire talent. we went there, opened actual office, created...
also daughter company, legal entity, so we can properly hire people. And that also was relatively easy. That was back in COVID times and we did everything remotely. So opening company, signing office lease, opening bank account. It was done remotely, of course, using notary and so on. But it was an interesting kind of experience, but it was possible to do so. So we opened.
office and started active hiring and yeah as I told at the moment we have 80 people and really kind of good location for for talent about Baltics Estonia Lithuania
just made sense, know, like those neighboring countries and close by, easy travel, easy commute, visiting offices. In Lithuania, for example, we started to hire really strong, experienced country manager who helped us and still helps us to grow as a team, managers, know, like day-to-day kind of activities in Lithuania, and that turns out
also to be quite successful. think at the moment we have more than 20 people already in Vilnius office.
Uldis (23:17.7)
In terms of scale you mentioned that you have this product side and it would be a natural assumption that it scales easier so it must be bigger, must grow faster. But in your experience and in your business, has it been so? What's the kind of split between services and product and what's the growth? Can you still do a double digit growth every year or triple digit growth with services?
Ervins (23:32.736)
No. No, no, no, no.
Ervins (23:46.944)
Yeah, it's a good question. So number one, if you look at this year, we are targeting something like 26 million revenue. And if we look, let's say, from a financial perspective, only tiny part is product revenue. But
We look at it differently. We use those products more like as an accelerator or door openers, know, like we can call it like that because those innovative products, once we attract companies to use them, then we can start discussions about expanding cooperation also to use as that kind of testing services, be it test automation, be it security, testing accessibility, testing, and that drives
most income for us. And that's conscious decision. We are not trying to...
let's say price tag or earn kind of directly from products with a huge revenue streams with the user more to attract customers, secure them and then also kind of cross sell different services that we have. But saying that there is kind of trending line upwards even for direct product revenues that we have.
Uldis (25:12.674)
Okay. And you mentioned selling and I mentioned Zoom, Orange, Microsoft, Twilio, Discord, like great, great logos to have on your website. maybe you can talk about the sales channels that you have used that have worked and tactics that have worked best for you how to get to such clients. Product is one, yes, that you already mentioned. What else?
Ervins (25:36.16)
Yeah, yeah.
So number one, it's kind of one step back, but number one that helped a lot our company, myself and also second co-founder Andres, both of us used to work in Skype before funding TestDialog. Myself, I spent around eight or nine years working in Skype in Estonia in Sweden. Afterwards, I was managing Skype for iOS engineering team back then. Also that helped to build very
strong connections within company from engineering as well as kind of business kind of groups. And I'm sure everybody knows how important our connections and kind of personal relationships and probably you're also going to have a term like Skype mafia. And it means that even people who have the Skype who created their own companies or who work in different
operations worldwide, they somehow help each other, you know, like with recommendations, with referrals and see how, how, how, can, of, they can help each other. And that helped us definitely initially to secure initial deals, just using our personal network and, and, and close initial kind of contracts. And that works quite well, even though, you know, like personal network, personal
connections, that's something that works quite well.
Janis Zeps (27:11.45)
Well, you're your first non-Estonian from like the Skype mafia network. We're spoken to quite a few of them It's it's fun to see that there are not only Estonians Mmm nice
Ervins (27:19.396)
We had a couple of Latvians
Uldis (27:23.416)
Yeah, but yeah, but you're, you're not super loud about it. I mean, it's, crazy. You've been on the market for so long and obviously we're familiar with your company, but it's somehow, not, maybe a critique, but, just, just, just noting that, that, yeah, that, it's not that we have been, hearing that much about, you know, the Skype mafia in Latvia building this great company, et cetera. It's, I guess, maybe Estonians are a bit more active.
Janis Zeps (27:27.578)
Mom.
Ervins (27:53.322)
Maybe, yeah, that's actually a good point. I mean, we mentioned it, of course, but it's not that we are super vocal. But yes, myself, joined back in 2007. That was in time when just eBay acquired Skype and been there through kind of all journey. It was kind of later private equity company, Silver Lake, and I've bought and afterwards, of course, Microsoft.
Uldis (27:53.381)
in marketing is a...
Ervins (28:19.404)
So it was kind of an interesting journey, but a bit kind of continuing to your question about sales. So personal network connections, that definitely is one thing that works because, especially in software testing, because software testing is trust. Trust is one of the important aspects of what companies...
look for when they select a software testing partner, they need to make sure that they can trust with their confidential data, with their unreleased products, and get a sense that somebody will actually do good quality work for them. So that trust is important aspect also, why those personal connections works quite well. But saying that, there's source that works quite well for us.
And that is marketing, content marketing. We invest a lot in technical blog posts, technical articles. We go to conferences with technical speakers. Let's say discuss about audio video innovations, audio video testing. And those things bring us good high quality inbound leads. And there's nothing better than high quality
inbound lead who have found us, who have done research and write with specific need saying, hey, we would like to test audio quality, we see that you are great in what you do, can we have a chat, you know, and of course also the best possible leads and that's why we are investing a lot in content marketing, digital ads and getting good
good many high quality leads from that. Zoom actually was the one company who found us from different articles that we generate from blogs. And that works quite well. And the sad ones, which...
Ervins (30:27.324)
also works quite okay are conferences. But by conferences I mean only and only if we go with stand with speakers if we make a presence in conference. If somebody goes with a conference and not attendee it's waste in most cases you know because you know like it's we've never seen that to be working.
Janis Zeps (30:54.217)
You come back with free pens at least and USB drives.
Ervins (30:56.388)
Exactly. It's typically one salesperson trying to sell something to other salesperson and then exchanging in a business card. But if it's a good conference and if you have kind of booth, presence, technical kind of speech, then it makes sense and it works.
Uldis (30:57.078)
and a hangover.
Janis Zeps (31:03.785)
Yeah.
Janis Zeps (31:20.009)
I want to ask, a lot of companies who start in B2B space, of course the dream of clients that we just discussed that you guys have, they will probably start with, unless the network allows, start with smaller companies maybe. And what's the...
From your perspective, is there a difference and what is the difference of servicing like giants of the world versus servicing just normal, successful, good businesses that are like just local? Is there a difference and yeah, can you talk about that?
Ervins (31:52.542)
Of course, yes, of course, there is, know, like, and maybe, know, a couple of examples. So when it comes to those large businesses, big, big, companies, so one of the, I would say, you know, main reasons how they select and look,
testing partners is against this trust and confidentiality. They cannot assess if they can really trust their private confidential data to third party. And sometimes it's very long and painful process. They audit certifications. That's why we have all ISO certifications. That's why we are renewing our certifications every year. And what's interesting is that sometimes
Uldis (32:32.064)
Audits, certifications, things like that.
Ervins (32:43.982)
I can't mention exact kind of client names, but sometimes not only is they audit our company, but they also audit country where kind of testing services are actually coming from. Let's say if it's Latvia, they take a look, independent look, what is political stability, what is kind of legislation, how it will look like in the future, get some kind of different kind of insights to see if they can actually entrust
testing services for us. Many of those companies require quite a lot upfront investments that need to be done by us in order to build secure testing rooms, secure laboratories, physical security kind of requirements that need to be incorporated in our offices.
Uldis (33:39.874)
But does that come before or after signing a contract?
Ervins (33:43.88)
Well, it works like that. We know that if we will do it, there's quite a high likelihood that we will get the work. Short answer is after. But of course we assess when it makes sense and when it doesn't make sense to do it. And sometimes it makes sense to do it because all business is risky. We see that we do that investment.
and very likely we will get a really good long-term partnership deal with company X. And then we go for it, we do it.
But going back to your question, that the confidentiality trust is one of the primary reasons what bigger companies look for to make, to be absolutely sure that products what they will send to companies that they can trust fully. The price is secondary, you know, like especially for bigger companies. Of course they look, you know, like what is the most reasonable price, but if it would be only in market for cheapest price, we would not exist.
probably we would be out of the business on day one. And second reason also what the bigger companies like and value a lot, our ability to provide different testing services so that they don't need to look for multiple vendors. So if it's automation, we can do it. If it's accessibility testing, we can do it. If it's load performance testing, we can do it. If it's security, we can do it. And they value it a lot and it's one partner.
They did one contract, security assessment, and then they can use multiple testing services, products, use them in our laboratories. For smaller companies, of course, it's somewhat easier. There's no such high, typically, security measures. Of course, there still is, but it's typically not in that.
Ervins (35:52.841)
level.
Uldis (35:55.032)
But then I hope it's worth the money to get those clients.
Ervins (36:00.704)
Well, it is. mean, we are quite OK from a financial perspective. But it's also maybe not so related, what I can tell, and everybody probably can have seen it, that in the IT industry,
Excellent years we had 2019, 20, 21, 22. I mean, of course, partly due to COVID and huge digital transformations that happened in that time. I IT just skyrocketed, you know, like sometimes was the demand was so huge that the problem was hiring right people, right kind of experts to keep up with demand. And back in
2023 or at end of 2022, was a bit of a chill down, you know, like across kind of all IT industry, you know, like a lot of companies did as they over hired kind of people, lot of layoffs, lot of budget cuts, you know, like, it somehow slowed down industry. It's recovering now, but definitely, at least from our perspective, not in levels, you know, like how
it kind of used to be. Everybody's a bit more careful, you know, like these days, comparing to three or four years ago.
Uldis (37:23.162)
So you have built this business without outside investment or as we like to call it and others also customer funded. Is it a conscious choice? Is it your business model? it? Yeah. And have you felt like it has been holding you back at some points or, or you're completely happy with that?
Ervins (37:28.972)
Yes.
Ervins (37:47.734)
Yeah, it's it was conscious decision. And of course, kind of primary reason was freedom and flexibility. Of course, you know, if you are just two co founders, we can be very flexible in what we do, how we do what business decisions we take, where we invest profits that we have, what we do is that. So that gives freedom and flexibility. that was conscious decision. Second,
Of course, we are in.
mostly services business, which means that quite early after funding company, we already had healthy revenue stream, meaning that we were able to fund operational costs, salaries, office rent and equipment purchases, which not always is possible if you build a product. Probably a longer time will bring revenue.
us it was quite early enough in a business. So yes, conscious decision. mean, of course, never say never, know, like market is evolving, our company is kind of changing and everything is on the table. But at this time, we have been kind of profitable every single year, with good EBITDA numbers, good profitability. And there hasn't been, you know, like so far, let's say,
need for external kind of investments. So we'll see what future kind of will bring but let's this conscious decision from day one.
Uldis (39:31.321)
Are VCs knocking on your door or have they given up?
Ervins (39:34.93)
No, of course, know, like of course we and time by time we have discussions, you know, like we like to see what it means and but for us it's not so much that we need external money influx if there would be good strategic alignment which could maybe help at some point to
triple revenue in a few years or build some huge value that we would see outside of money. We would be open to discuss and see. mean, it's business in the end, but situations in all like ways is just, let's say, money for specific amount of shares in companies. That's something that is not very interesting for a company.
Uldis (40:30.352)
You mentioned that you're two founders and you need just two opinions to make a decision. However, they also say that two is especially 50-50 is the worst combination of what to have in terms of if you have disagreement and there's basically no way how to get out of a deadlock or resolution. How has that two founders set up worked for you? it ever been an issue?
or something that you had to overcome and figure out ways how to disputes or kind of stuck decision making.
Ervins (41:11.5)
Yeah, I mean, I believe we have been quite lucky in that sense. It's true that I have seen and heard, you know, like about many kind of different kind of stories that it kind of doesn't turn out well. But in our case, we've been kind of quite lucky. And I think there are a couple of reasons. One reason that I do believe both of us as a co-founders are quite rational and smart. If we have kind of different opinions, we like to rationalize it on a wide
board to see, you know, like what are different pros cons. And, and after that, you know, like, typically can sense, you know, like what to do in many cases, if it's not strategic decision, but just some operational decision, we typically don't mind, you know, like if, I have some kind of a good idea, I just go for it and do it the same, you know, like Andre's co founder, I don't mind, you know, like for operational kind of reasons we even don't need kind of to think.
you know, if it's strategic decision, of course we brainstorm and you know, try to rationalize, you know, like what is best. So far it's been kind of working out quite well, but I think also important aspect is that we are primarily services company.
It probably would be more harder if it would be primarily product company, you know, like we're building product with specific roadmap, business kind of plan. It probably could be more challenging for service company. is, at least in my mind, somewhat easier to share those on the responsibilities.
Uldis (42:57.24)
Well, since it is a service company and you have more than 500 people helping you provide that service and obviously those people are in some ways also the face of the company, something that your clients are facing and interacting.
Ervins (43:11.584)
Mm-hmm.
Uldis (43:14.588)
In terms of managing such kind of workforce, I just came up with this kind of thought experiment if you would have to go to a desert island with all 500 of them, what is the three management or organizational tools that you would definitely not survive without? What is kind of the main? It can be practices, can be tools, can be... What kind of is the major things for you?
Ervins (43:21.836)
Yeah.
Ervins (43:40.568)
Yeah, yeah, good question. I will probably, I mean, one, those are probably more like practices, you know, like because tools are probably not so interesting. Everybody can use collaboration tools, Google Cloud, Jira, Slack, Discord, you know, like whatever, but that is a given, of course. But I think what we...
A couple of things that we try to establish within our culture and processes is number one, when it comes to management, even myself, co-founder, as well as our technical directors, we like to keep policy where we are open to have discussions with anyone in the company. So anyone at any given point of time could come up to us, have a discussion, share some ideas, share feedback.
discuss anything, like what they have, like we like to create this kind of open policy, know, like where every single kind of employee, even intern, could come by, set up a meeting and chat about anything, know, like what's on their mind.
Uldis (44:52.857)
And do they?
Ervins (44:54.548)
Actually, yes, know, like many of them not, but many, many, many, years, you know, like, and, and, sometimes those are pretty interesting kind of discussions, you know, like it helps to understand what really happens in a company, what's the feedback, sometimes they have really good insights and feedback about what's happening within their projects. So, but yeah, so that's one thing. Second thing is,
We help, I would say, with more than 500 people. Many companies help, but I think for us it would be extremely hard.
To live without it's called like performance and development system There are people who love it and people who hate it but it obviously isn't any company, but but it would be impossible not how one because there's so many questions for Everyone like how can I advance in my career? How I can get higher salary when I can reach X amount of you know? Rays
what needs to be done, what certifications I need to do, what needs to happen in my project. And we invested a lot in building quite robust and clear performance and development system with clear expectations, salary levels, what needs to be done and how to advance in Korea. And again, like I'm saying, there are people who love it, people who hate it, but without such system, would be just impossible to
manage 500 employees and give some basis for everybody how they can operate.
Uldis (46:41.744)
No, it's extremely important, especially in a competitive labor market situation. When people come and ask, what do I need to do to get X amount of salary and you don't have an answer, it kind of takes all the power away from you.
Ervins (46:55.488)
Mm-hmm.
Uldis (46:59.788)
and basically then you end up all the time bargaining trying to kind of fight off some kind of poachers and all these things so it's really great that you have it kind of under your control.
Ervins (47:14.41)
Yeah, and another probably kind of worse mentioning, like, I wouldn't say.
We try to work from office. Not everybody, know, like, but for example, sales, management, marketing, design, HR, we try to be almost every day. Of course, it's flexibility, but we try to be almost every day in office because especially when it comes to brainstorming different initiatives about marketing, about sales.
approaches, what works, what doesn't work, what we can change, how we can come up with some new products or new testing initiatives. At least from my perspective, it is possible remotely, but it is nowhere near as efficient. if like-minded people come in the room, brainstorm, speak each other and then come up with some great new initiatives, it just kind of works better.
Again, there are people who love it, people who hate it.
Uldis (48:27.056)
It seems like this is the new, this might be the new polarizing thing, know, vaccines or Democrat or Republican, and now it's like back to the office or fully remote or I think it's like so strong opinions on both sides. It would be interesting to have some strong founders debate this in a proper debate format.
Ervins (48:33.718)
You
Ervins (48:40.149)
Yeah.
Ervins (48:49.16)
Sure, but...
But still we try to keep it kind of flexible with some exceptions. Of course, there are a couple of projects, know, like we are testing happens from secure facilities, what I mentioned before, and there's just impossible, you know, due to high confidentiality of products being tested, it's just impossible to do it from home. Imagine like this so sensitive, let's say builds or products that cannot be leaked anywhere and it
it can only happen from office and in those situations of course there is no other possibility than to carry out testing from office from specific secure facilities.
Uldis (49:36.29)
All right, all right, that means thanks. Very interesting stuff and obviously very critical. And sometimes we tend to over overlook this, you know, going to go production and then let the users test it and move fast, break things and it's going to be fine. And, who, and who, you know, has this obscure Nokia phone anymore or whatever. Let's just focus on, on Samsung and iPhone.
Ervins (49:52.492)
you
Uldis (50:07.094)
So yeah, very important to keep things clean and neat and well tested. So thanks for sharing your experiences. Really happy for the story. Do not hesitate to be more vocal about it. Also in the local tech scene and circles, I think you've done a great job and people should know more about it and hopefully be inspired. So thanks a lot, Ervins
Janis Zeps (50:17.033)
Thank you,
Ervins (50:34.316)
Thank you and thank you for hosting the podcast. It's really good having conversations.
Janis Zeps (50:41.117)
That's awesome. Thank you.
Uldis (50:43.278)
Alright, and to the listeners, we see you next week. Click that subscribe button and don't forget to see us on YouTube as well.
Janis Zeps (50:50.058)
test that subscribe button on Spotify please listeners for us go and try to test whether it works thanks
Uldis (50:53.1)
Yeah, let's see if it works. Alright guys, thank you.
Ervins (50:59.34)
Thank you.
Please note that the transcript text is AI-generated. We apologize for any potential errors or inaccuracies. Thank you for your understanding.
