Roumen Kassabov

𝗦𝗼𝗳𝘁 𝗟𝗮𝘂𝗻𝗰𝗵 𝗦𝗲𝗰𝗿𝗲𝘁𝘀 𝟮𝟬𝟮𝟱 📊🚀 How to test, tune, and kill games like a PRO. The Framework that decides if your game 𝗹𝗶𝘃𝗲𝘀 𝗼𝗿 𝗱𝗶𝗲𝘀 👇 Soft launch isn’t a marketing checkbox; it’s your game’s survival test. If your LTV isn’t higher than CPI by day 30, stop burning cash. Three months isn’t enough! You need six to see the real retention curve. Kill it with fire if it doesn’t move the needle; your studio will thank you. Philippines for crashes, Poland for retention, Germany for whales. Creative fatigue hits faster in small geos - refresh or die. Event-optimized UA is the bridge between installs and revenue. A bad D1/D7 ratio is a red flag you can’t ignore. If you can’t scale your UA mix in soft launch, you won’t survive global. No fake ads! Misleading retention data will sink your launch. Too many studios treat soft launch as a quick marketing step. In reality, it’s the most important survival phase your game will ever face. Here’s the 3-stage framework I’ve used to take dozens of games from idea to profitable global launch live on two & a half gamers: https://lnkd.in/dMU5VrDC Full Bible: https://lnkd.in/d36pENMT

103

13 Comments

The 7 questions I ask gaming founders that make 80% realize they're not ready for funding. 1. "What's your monthly burn rate if you don't raise?" Most can't answer. They've already mentally spent the money they don't have. Winners know their survival number down to the dollar. 2. "Tell me what you've learned from people playing your games?" "We don't have players yet." And you are raising $2M? In the age of AI coding etc., it's ridiculous if you haven't built something. Tested it. Killed it if needed. 3. "What games have you killed?" If the answer is zero, you haven't been building long enough. At Next Games, we killed 14 for every 1 we launched. Killing bad ideas fast is a skill. 4. "Who on your team has shipped a game that made money?" The uncomfortable silence tells me everything. "Well, our CTO worked at..." Working at Supercell or Riot isn't the same as shipping a profitable game. I need someone who's felt the pressure of their own game going through all the stages and finally starting to make money. 5. "What happens when a big games company copies your core mechanic?" Deer in headlights. If the incumbents can copy your idea in 3 months and crush you, you don't have a business. You have a feature. 6. "Walk me through your game's unit economics at scale." LTV: $5. CAC: $8. "But at scale..." No. The math doesn't magically improve. It gets worse as you go through bigger and bigger cohorts. If it doesn't work on paper, it won't work at scale. 7. "Why games?" "Games are a huge market." That's not a reason. That's a Google search. The right answer involves obsession, not opportunity. After these questions, most realize they need to go back and build. The 20% who have real answers? Those are the ones I might invest in.

197

42 Comments

Not every gaming job is on LinkedIn. My 7 favorite subreddits for job junting/networking: 𝐫/𝐠𝐚𝐦𝐞𝐝𝐞𝐯: With 1.9M+ members, this is one of the best communities for devs. Feedback threads, project showcases, AMA sessions, technical help, you’ll always find something useful here. https://lnkd.in/dnmZJAdr 𝐫/𝐠𝐚𝐦𝐞𝐝𝐞𝐯𝐜𝐥𝐚𝐬𝐬𝐢𝐟𝐢𝐞𝐝𝐬: One of the best places to find opportunities in games, often full remote. Quite big, it has 126K+ members and daily new gigs. This is only for paid collaborations, quite good for freelancers of short term contracts. https://lnkd.in/dmHUqHqw 𝐫/𝐠𝐚𝐦𝐞𝐃𝐞𝐬𝐢𝐠𝐧: A must for designers (265K+ members). Discussions on systems, mechanics, balance and design theory. Great if you want to get feedback on concepts, learn from others or simply network with other designers. https://lnkd.in/daAMw3c6 𝐫/𝐏𝐥𝐚𝐲𝐌𝐲𝐆𝐚𝐦𝐞: 126K+ devs and gamers testing each other’s games. Perfect to collect feedback, spot bugs early and improve your visibility before launch. Super recommended for indies/solo devs. https://lnkd.in/dtPGqfTE 𝐫/𝐠𝐚𝐦𝐞𝐝𝐞𝐯𝐜𝐫𝐢𝐭𝐢𝐪𝐮𝐞: A smaller but very engaged subreddit where you can post your work in progress and receive constructive feedback. Valuable for improving your craft and learning how others approach production/mrketing problems https://lnkd.in/d96u2k_A 𝐫/𝐠𝐚𝐦𝐞𝐝𝐞𝐯𝐣𝐨𝐛𝐬: About 17K members, only open positions and contract work. Smaller than big job boards, but quite good as it’s very active. https://lnkd.in/d93U4qFf 𝐫/𝐈𝐧𝐝𝐢𝐞𝐃𝐞𝐯: With 298K+ members, this community is the best for indies. You’ll find discussions on production, marketing, funding and collaboration. Great if you’re working solo or in a small team and want advice directly from peers. https://lnkd.in/dyVj_DbH Being active in communities is a career hack. Reddit offers some of the best in games. Any other subreddit worth joining?

400

17 Comments

$38M raised, 50% Steam mixed reviews received. Here's what went wrong with Stormgate's launch strategy. In 2020, two ex-Blizzard veterans set out to “bring RTS back.” $34.7M in VC. $2.38M on Kickstarter. $1M+ from equity crowdfunding. The pedigree, the war chest, and the hype were all there. Then Early Access hit: - Mixed reviews (~50%) - Art called “cartoony” and “low effort” - Combat that felt slow and weightless - The “Stormgates” mechanic missing until a year later - A “0.6” version number that screamed beta To their credit, Frost Giant fought back: - Rebuilt the campaign from scratch - Overhauled art, factions, core mechanics - Added Stormgates, shifting the meta to comeback plays By Aug 5, the campaign was a massive improvement. But the launch shipped half the vision (Campaign + 1v1), with Co-op, 3v3, and the editor stuck in public test. Reviews stayed ~50%. The tension now: Monetize a casual PvE audience and keep a hardcore PvP base alive long enough to make RTS relevant again. Lessons for niche genres: 1. Pedigree wins funding, but raises the bar. 2. Engagement ≠ transparency if the message backfires. 3. “Launch early” only works if your MVP is close to expectations. 4. Hybrid funding = freedom + shareholder-level scrutiny. If Stormgate pulls it off, it’s a blueprint. If not, it’s a cautionary tale. Either way it's worth watching. And I'm rooting for a turn around.

59

26 Comments

This is where to start when building your brand/marketing team in web3/crypto 1. Crypto culture is something else - to succeed you need to understand this. This is why companies like Coinbase are hiring Crypto Twitter native marketers to run their main X account. You also get another distribution channel, ie the person you are hiring also has a strong and often loyal following. Pitfalls here of course and you need to do DD, but there are many valuable creators and subject matter experts depending on your vertical focus. Many seasoned web2 marketers have failed to make the transition to web3 as they simply didnt understand crypto culture, the language, crypto branding. 2. Prices drives sentiment, not the "brand" The primary motivation for many in crypto is money. You need to recognise and understand this vs swimming upstream against this. Not saying dont build a brand, build a community - absolutely do this. Just dont kid yourself thinking best product or brand wins. People need to understand the "money flywheel" ie how could this brand, product, service bring value to me. 3. Dont throw marketing fundamentals out of the window Likes, comments = community or revenue. It doesnt. Having worked in crypto for over 6 years now seen and made a lot of mistakes - ie thinking you have to spend money on X,Y,Z to be relevant. You dont. Service providers will tell you that you have to do X content, work with Y Creator, be on X publication. Reality is same as web2 marketing - get the basic marketing and sales funnel in place first. So when you start marketing - people are clear on what you want them to do, where to go and how to support. Good luck - never lose your curiosity, just do your homework before diving in.

36

3 Comments

It’s easy to watch the AI talent wars and think they only matter to Big Tech. But this stuff doesn’t just affect OpenAI and Meta. As AI talent gets more scarce and expensive, deciding which parts of your AI systems to build in-house becomes critical. This isn’t a question of capability, because these teams are incredibly sharp. The bigger question is where you want to spend your best people’s time. I’ve talked to teams who want to build everything from scratch. Sometimes it makes sense, but more often than not, the opportunity costs are far too significant. At Decagon, our focus is giving you the controls to quickly build AI agents customized to your business, while abstracting away the complexity you shouldn’t need to worry about: managing inference clusters, orchestrating and parallelizing workflows, keeping responses grounded, etc. There’s a time to build, but there’s also a time to pick your battles. And with AI moving this fast, misjudging that can be very, very expensive.

245

28 Comments