Sergey Konstantinov

I heard a story yesterday from a new friend about an unsuccessful software development project that lasted for 1.5 years. It is over 1.2M USD in total budget spent to date and still needs to be finished. How do I know it's not successful? The phrase that he said: "They will probably deliver it this Autumn, but it will be a very raw product that no one needs" So, what went wrong? First of all, the contractor's choice. They selected a large software development agency with offices in 10 countries, and that seemed trustworthy. What you can expect in this case? - You will get your product done with a good quality - You will pay 30-50% more than for small consultancies - You will have a chance to meet someone in person somewhere in your country - Your product will be finished (but you obviously need to pay for every hour worked). What's the downside except the price? One main thing: Your product will be your responsibility only. Was your idea bad? Well, you paid 1.5M to validate it; no one really cares if something built will never be used. In 2V Modules, we have a different approach, which I'm a big fan of. If we start working on something, we dive deep to understand what this product is for, what pains we are addressing, who the ICP is, and whether it was validated or not. Even if it's an "internal" product, there are still multiple ways of building it. It's not only about the business analysis stage, it's about finding the right solutions. So I think small teams are often a better fit because they actually CARE what they are building, not only HOW it is built.

11

2 Comments

In the previous post, i gave a rundown of cozytown.eu. A project i am working on while looking for my next career opportunity. This time let's talk about the tech stack. Even if this stack may not be super relevant for all companies. The Cozytown stack is not necessarily geared towards millions of users. It will currently break under heavy load. And that is a conscious decision. If this takes off, cool! awesome! and then i can deal with scale at that time. I do realize that his might be an unpopular opinion in developer circles not to have a service that is not always up and doesn't auto-scale and redundant across the globe. Cozytown is currently hosted on 1 virtual machine running inside a docker container. It uses SQLite for storing trip information and block storage for image assets. The backend is written in Go with vanilla Javascript, Css and Html for the visible parts. Everything runs as a monolith. All in all, a simple stack for a simple product. The release cycle consists of tagging a git commit, build a docker image and deploy it to my 1 node docker swarm. That one node also hosts www.brandmaps.io and other random ideas i get from time to time. Remember that in the end it is all about the product. No product means no customers. A product means potential customers. It means it can be validated. Is the idea good? Should i pivot? Should i double down? Should i drop it for another idea? Questions that cannot be answered if my time is spent on things other than shipping. My expenses for this is currently 5 euro a month + domain costs.

7

Hi No Code Gang: Gamification and Leadership, First Attempts About two months ago, I decided to build a close-knit in-house team. Previously I: - Hired contractors who came and went. - Delegated team communication to the project manager. - Was detached from project details and company community. Team members only knew each other within their projects. Casual calls had low attendance. I wanted to bond with everyone I worked with, grow together, vibe, exchange memes, and more. Plus, I wanted a team to eventually switch to our own products after custom development. Honestly, I used to see people as functions. Looking back, that approach makes me sick. In the last two months, everything has changed: 1) HR It started with HR. Besides asking about experience and values, I asked: who would I be friends with? Who would I want to spend every day with? 2) Hi No Code Game With my friend and mentor, Volya, we tested the idea of turning work processes into game mechanics: - Work would be more fun. - Challenges would seem easier, like a "Level 10 boss" is easier than an angry client. - Achievements in the game are more exciting than hitting KPIs. - The game environment is immersive, reducing distractions. But we thought it was too early and needed basic processes first. However, I couldn’t wait, bursting with ideas. Here’s what we’ve implemented so far: 1. Professional Names:   - Female Developer: DEV Girl   - Male Developer: DEV Boy   - Female Project Manager: PROD girl   - Business Development Male: BIZboy   - Designer Male: UX boy   Grades like in rap lol:   - LIL - Junior, beginner   - OG - Mid-level + 2. Selected Universe for Reference: GTA. We started building around this theme. 3. Custom Avatars in Midjourney, themed around GTA and the Bloods gang with red symbolism. 4. Mission = Milestone in a project for the project team or a sale for bizdev. Problems in projects = missions too. 5. NoCoins - in-game currency earned for daily reports, completing missions, or values. They can be exchanged for merchandise or money. 6. Gang Gatherings = weekly calls where everyone shares how their week went and posts weekly photos in a thread. 7. City Map Exploration = learning new tools or market segments, or starting internal projects. 8. My Favorite Mechanic - Not About GTA: Everyone wishes each other good morning every day for over a month! Love it! It’s fun to see which mechanics take root in our culture. For example, a week ago, three of us shot a reel from WeWork, and deleveloper Stasya edited it in a GTA style. Support the reel: https://lnkd.in/eKCB_ggx Make your companies a game, it’s awesome! Next week, I’ll share our sales/projects successes and failures and our transformation into a SaaS factory. This post was written for my Russian blog on Telegram (t.me/failblog). Keep in mind that I also have an English blog.

2

2 Comments

How Olearis Helped Parlor App Increase Revenue by 3X in Six Months Parlor is a social app that enables people to make random calls, finding the right person to talk to instantly. It might seem like a phenomenon nowadays, as it’s commonly believed that people are not willing to talk anymore. However, it appears we all underestimate social hunger, especially in the Arabic world! Parlor has been on the market for years, amassing 7 million downloads and achieving 2.5 million sessions per month. Originally, Parlor’s monetization model was based on advertising. It was assumed that people would prefer to watch ads rather than pay. But as time passed, the app became outdated, and together with Parlor founder Joel Schwartz, we bet that better UX and an ad-free experience would increase user engagement. Following YouTube's example, we offered an ad-free subscription that also included additional features, such as advanced search filters. Two months after its launch, revenue doubled! Then we undertook a full app redesign and significantly improved the UX. This launch was another huge success, leading to a x2 decrease in the number of user actions and a x2 increase in time spent in the app. In other words, when the UX became clearer, people stopped struggling with the interface and started spending more time actually using the app! And voilà! Revenue tripled with no investment in marketing! This is a great example of how a product with good potential can benefit from the right product solutions and better UX.

10

1 Comment

Is it finally time to say goodbye to ugly Objective-C wrappers for Kotlin shared code in our Kotlin Multiplatform apps? 🚀 Even though we started working with KMP less than a year ago, I still believe we are in the group of early adopters of this technology. Until now, we have used KMP to build two different apps which have roughly 80 downloads per day 📈 With that being said, during this year, one of the main issues of our iOS team had with debugging the compiled shared code was that KMP is currently generating Objective-C headers from Kotlin code 🤔 We are aware that Google will release Swift interoperability during this year, but I always thought to myself, there has to be a better way to do it. And, totally by accident, I found it today 💡 So apparently, there is a framework: https://skie.touchlab.co/, which can be used in KMP projects to generate Swift code based on generated Objective-C headers 🤯 I still didn’t try it out, but I plan on doing it in the upcoming week. Community developed frameworks rock 🤘🏻

27

3 Comments

https://lnkd.in/djDyhZNz Finnish payment API coders are trying to pre-empt even further insanity from the Finnish government: "My only concern with this scheme is that it only allows for tax rates that are easily represented by rational numbers. There's a risk that future legislation introduces tax percentages (say π%) that require an unfeasibly large numerator and/or denominator to achieve the desired accuracy." Relates to: https://lnkd.in/dgYgwyQN

2

What I read last weekend ** 𝗧𝗵𝗲 𝗥𝗶𝗴𝗵𝘁 𝗞𝗶𝗻𝗱 𝗼𝗳 𝗔**𝗵𝗼𝗹𝗲 ** Turns out, good and horrible founders have the same characteristics, but very different characters. How do you tell them apart up front? (It’s super easy after a few days of working together though) --> https://lnkd.in/dsU9wD6J ** 𝗛𝗼𝘄 𝘁𝗼 𝗯𝗲 𝗮 𝗴𝗼𝗼𝗱 𝗹𝗼𝘀𝗲𝗿 ** You are Volkswagen, it’s 1949. You want to sell the car Hitler commissioned. To Americans. The car is about half the size of an American car, but gas is still cheap. Ah, the car also looks a bit odd.  Your ad agency has a secret weapon though: humor. These ads have everything I miss from this region: a sense of humor, the clever, self-deprecating kind. Copywriting that sounds like a human had written it. --> https://lnkd.in/dUwv4T3T ** 𝗥𝗶𝗰𝗵 𝗣𝗲𝗼𝗽𝗹𝗲 𝗗𝗼𝗻'𝘁 𝗧𝗮𝗹𝗸 𝘁𝗼 𝗥𝗼𝗯𝗼𝘁𝘀 ** Robo-advisors for investments were statistically better than humans. Nevertheless, they did not succeed. Why? We will soon all talk to AI agents (and those are probably way better than the lets-not-even-mention-them-anymore chatbots of yesteryear) but here’s why they still won’t replace many human ones.  --> https://lnkd.in/dHYV7gqx

13

2 Comments

"Becoming a 10x developer by harnessing the power of generative AI." For those who follow me long enough know that I was a bit set back by this title of a talk given by EUR ING Ioannis Kolaxis MSc at jPrime today. I nevertheless went in and heard an exceptionally well balanced talk, backed by actual numbers in what and which cases generated code is worse / non functional, how the code feels like (a bit like the pieces added by short term contractors on large code bases), but also where GenAI is actually helpful, like for discovering and experimenting. In all fairness, if the machine reads the docs, I am happy to some extend, and if than communicates to users why stuff is the way it is, so be it. The other learnings assure me that the qualities I try to achieve now for quite some time will be value in another decade, still (see attached image). I think there will be a recording, and I really recommend that.

44

3 Comments

𝗣𝗿𝗶𝗰𝗶𝗻𝗴 𝗶𝘀 𝗺𝘂𝗰𝗵 𝗺𝗼𝗿𝗲 𝘁𝗵𝗮𝗻 𝗷𝘂𝘀𝘁 𝘁𝗵𝗲 𝗽𝗿𝗶𝗰𝗲 - Meetup.com just increased their prices by over 100%! Bold or desperate move? Here's why vote for the latter and are breaking up with Meetup (and no, it's not because of the new price, but because of the PRICING 👇 ). We just received a poorly communicated email from Meetup.com informing us that our standard plan is increasing to $99 per 6 months (that's $198 per year). This is a significant increase, nearly double what we were paying before. How much of an increase exactly? hard to tell (more about this later). The new price isn't necessarily off, but the pricing strategy is, which is why I believe it's a desperate rather than a bold move. 😢 No increase since 2019 - Regular, smaller increases are better than a +100% hike every 5 years! 😢Lack of price transparency in communication: The email didn't mention our previous rate, likely to avoid highlighting the steep percentage increase. Trust and transparency took a hit here. 😢Lack of price transparency on the homepage: Hiding prices behind a registration wall creates a perception of secrecy, tricking and perhaps ... discriminatory pricing. 😢Light rationales: The communication about the rationale behind the price increase was light ... to say the least! one of the points explains that the increase is partially justified in the fact that Meetup is now "identifying and fixing bugs as quickly as possible." 😮😮 (btw - that's why when we work on our pricing projects we start from the communication!) Goodbye Meetup: ☑ Is $99 per 6 months too much? Not necessarily. It's still not our biggest expense for events. ☑Will we break up with Meetup? Probably, but not because of the price—because of their way to approach pricing. ☑ What's the difference? The new price point might be acceptable, but pricing is much more than just the price. It's about how you manage your prices and how you communicate value to customers. This relationship is about transparency above all. Take-aways: ✅𝗣𝗿𝗶𝗰𝗶𝗻𝗴 𝗶𝘀 𝗺𝘂𝗰𝗵 𝗺𝗼𝗿𝗲 𝘁𝗵𝗮𝗻 𝗷𝘂𝘀𝘁 𝘁𝗵𝗲 𝗽𝗿𝗶𝗰𝗲! The price tag comes last and usually is not the (main) reason why people leave your services ✅𝗧𝗿𝗮𝗻𝘀𝗽𝗮𝗿𝗲𝗻𝗰𝘆 𝗮𝗻𝗱 𝗙𝗮𝗶𝗿𝗻𝗲𝘀𝘀 are increasingly key topics in pricing. Some companies made them their strengths. Some others, not. ✅𝗖𝗼𝗻𝘁𝗶𝗻𝘂𝗼𝘂𝘀 𝗲𝘅𝗲𝗿𝗰𝗶𝘀𝗲: working continuously on pricing and adapting constantly has 2 advantages: avoids sudden spikes to customers, and "trains" the organization to deal with price increases. ------- 👉 I write about product pricing, product monetization, and creating top product organizations. Follow me to receive my articles and to be informed of the next events! #pricingstrategy #pricingtransparency #priceincrease

9

6 Comments

While we at Mirego are known mostly for building digital products from scratch, we embrace the fact that companies more often than not have built their digital ecosystem over the last few years, and it does not make sense to start over. We have built our recipe to take over existing projects and code bases, and my colleague Samuel Garneau has shared some key elements of our philosophy. It's well worth a read! https://lnkd.in/dPG_76J2

14

In the advent of AI, many startup enthusiasts want to wrap AI in a paper and sell it as-is. However, AI is not often saleable as a standalone solution. AI should be a polish on your idea, not the other way around. In other words, if you cannot detach AI from your solution, in most cases, you have no real problem to solve.

5

This is fascinating. Look how easy it is to identify the troll / bot accounts pushing disinformation at us? In this case, Russian bots about Ukraine. #data #politics #politicaldata #sentimentdata #news #bots #ai #automatedaccounts #fakenews #disinformation #cybersecurity #datascience #misinformation

2

Bubble for Enterprise? Yes, it’s possible. Some time ago, Mike Guzowski asked me about the Enterprise features of Bubble.io, and I promised to write a post about it. Finally, I have the time, and here it is - a short summary of features that could be interesting for big companies looking to enter the #nocode space. Bubble's Enterprise plan is specifically designed to meet the demands and compliance requirements of large organizations. The biggest difference between the other plans is that on Enterprise you can migrate your app to an isolated instance of Bubble-managed AWS server, giving you choice of hosting region, a static IP address and greater control of server resources and updates. While you cannot use your own cloud, this gives you separate instance from all the other apps, with your own dedicated resources. You can host one or more apps on this server, adding capacity if needed. 🔒Security and Compliance Bubble's Enterprise Plan includes SOC 2 Type II compliance and GDPR, regular penetration testing and advanced user authentication mechanisms, like single sign-on. 🤝Collaboration For seamless team collaboration, Bubble's Enterprise Plan offers centralized user management. This feature allows administrators to control access levels across various apps. You can also use SSO integration to simplify user authentication. 🆘Support Enterprise users get priority access to Bubble’s dedicated support team, which includes a Technical Success Management team. This team assists with app optimization, troubleshooting, and security queries. Additionally, each enterprise client has a dedicated account manager to ensure personalized support and quicker resolution of any issues that arise. If you have more questions, let me know and I’ll try to answer all of them below 👇

9

3 Comments

I never imagined how disastrous a missing approval process is 🥵 𝗛𝗶𝗿𝗶𝗻𝗴 𝗠𝗮𝗻𝗮𝗴𝗲𝗿 (𝗛𝗠): Hey, did we start the interviews for the Senior Engineer role? 𝗥𝗲𝗰𝗿𝘂𝗶𝘁𝗲𝗿 (𝗥): Yes, we already shortlisted three candidates. The first one seems like a great fit! 𝗛𝗠: Wait, did we get the budget approval for this position? I haven’t seen any confirmation... 𝗥: Oh no, I thought you got it approved last week in our Slack conversation. I’ve been moving forward based on that. 𝗛𝗠: We discussed it in the hallway, but it never got formalized. We might be in trouble if we don’t have the budget cleared. 𝗥: This is bad. We've already invested time in the candidates. 𝗛𝗠: And what about the candidate? We can’t afford to backtrack now 😧 Most companies still use fragmented places online and offline for approvals or even reuse tools like Asana or ClickUp. This leads to a non-scalable and easy-to-break approval infrastructure. Bringing all workforce data & decisions into one platform ensures your team's and your firm's reputation remain intact 😇

13

What is your perfect developer look like? Everyone who has been hiring even once in life wants to find that one perfect candidate who will take your company to unbelievable wins. I don`t know how you are, but I saw a lot of job ads that just screamed that this job is for the Chosen one only and if you are not - you have no chance to get the offer. So, I decided to talk to you about what a perfect developer looks like. Let me be the first to describe this specialist. I will consider you as a perfect developer if you : ✅ don`t try to be “perfect” ✅ don`t afraid to say “I don`t know” ✅ don`t stop learning ✅ are curious, responsible and goal-oriented ✅ don`t lie and cheat ✅ always give honest feedback ✅ respect your team, clients and their time ✅ know how to do the job you were hired for ✅ flexible and adaptive ✅ don`t try to be the loudest person in the room Yes, in my list there are no diploma requirements or experience with the top world IT companies or ability to work overtime or on weekends. I don`t consider employees as slaves or robots or instruments to achieve my goals. First of all, I value the people I hire, their identity, desire to work, contribution to the company`s success. So, I beg you don't be too strict and picky. We are all humans after all. P.S. What does YOUR perfect developer look like? _-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_ Hi, I`m George Fironov - Talmatic`s CEO! My team and I can pick the right developer for the clients' project. DM 📩 me if you need my help and don`t forget to follow me and Talmatic to learn more about our services 🤗 #developer #hiringdevs #staffing

13

10 Comments

Very interesting podcast about the current state of big tech. Spoiler - it's in deep s**t :) The central thesis of the discussion is that big tech has been faking technological progress for quite some time just to keep Wall Street happy and the chickens are finally coming to roost. https://lnkd.in/dZSmXxtn

89

31 Comments

I think private kindergartens in Croatia currently behave like developers in their golden era. 1. Pricing Autonomy: Similar to developers setting their own prices, private kindergartens seem to have the freedom to charge whatever they see fit. 2. Unlimited Demand: Just as developers could demand anything or anyone they wanted, these kindergartens can seemingly do the same. If they're not satisfied with a parent or child, they can easily replace them from a pool of applicants in waiting lists.

14

Product-market fit is not enough anymore. You need position-market fit: There was a time, not too long ago, where in startupland one could build a thing that solved a real problem, put a price tag on it, and see if the market wanted it: • If they did, then we would have said he had found product-market fit • If they didn’t, we would say he didn’t and it was “back to the lab” That time, for better or worse, is long gone. Here’s how Enzo Avigo and I break it down: In 2024, a utility provided through software can’t make a dent effectively anymore. People’s heads are overstuffed with competing products, messaging, and narratives. It’s hard for a product alone to get a market edge. The main exceptions are new tech or hyper-niche markets. ChatGPT, for example. But that is a rare instances of breakthrough technology where the “product” itself carries the bulk of the impact. Most companies don't have that luxury and are not in such a position. So, if a product alone isn’t enough, then what is enough? Enter position-market-fit. • If “product-market-fit” means that you’ve found the right kind of product that the market wants… • “Position-market-fit” means that you’ve found the right combination of product/brand/marketing/pricing/go-to-market/sales/etc in a given domain. The Importance of Brain Estate The fundamental reason why “position-market-fit” is so important is that it operates more at a personal and subconscious level. Our brains can only conceptualize a finite set of “characters'' per domain. Similar to the "Dunbar number" rule, which suggests we can maintain stable social relationships with up to 150 people, our brains are wired to understand only a finite number of company-market associations. Gaining a strong positional edge, or nailing “position-market-fit” is the exercise through which a company, with the right combination of product, brand, pricing, marketing and go-market is able to conquer a certain portion of consumer “brain-estate.” The Story of Startup Success If you step back and analyze some of the best startups from the last decade, you'll see they excelled at this. → Are you building in an established market dominated by large incumbents with feature-bloated, clunky software? In that case, you might want to position your product as a high-craft, premium option. Does Linear ring a bell? → Alternatively, if you're entering a highly commoditized market dominated by a few corporate-looking brands, consider positioning yourself as the quirky, fun company that doesn’t take itself too seriously. Does Arc Browser come to mind? Their pricing strategies are almost a second-order effect of their employed market positioning. And that’s the power of position-market fit. — So how do you build your own position-market fit? We cover that in the deep dive: https://lnkd.in/e6J_zPzW

855

113 Comments

I know I’m not the only one who loves digging deep into big iOS project optimizations, but how on earth did I manage to miss differences between static and dynamic library linking and how they can affect project performance? 🤯 So in my free time, I quite often read how big companies solve their big iOS project issues. Recently, I noticed that quite a lot of them talk about how properly linking both their third party libraries as well as their local module dependencies positively impacts their iOS apps build times and app size 🚀 Then, I dived in into static and dynamic library linking, went through a couple of articles, and these are my key takeaways about differences between these two, and when you should use which: 1. Static libraries are libraries that get linked to your iOS application directly at compile time. You should use them when you want to increase performances of you iOS apps and lower build times, but they might affect you iOS app grow in size 📈 2. Dynamic libraries are libraries that get linked with the application at the run time, and aren’t a part of app executable. They should be used mostly when you are developing reusable frameworks shared between multiple applications, because in this scenario, the the library file is shared in memory and is reducing the overall memory footprint 📉 3. SPM packages are by default static, unless you explicitly mark them as dynamic. I was curious about this one because I am a big SPM fan boy 📦 Investigating of how big iOS systems work has always helped me get a much clearer perspective about how iOS apps actually work, and I always find out something new about it, it never gets boring 🤘

85

6 Comments