How Two Friends Built a Roblox Game From Scratch

In early 2025, neither of us had ever opened Roblox Studio. Chris was a builder and creative thinker. Gio had a background in business and product. Neither of us was a Luau programmer. We had one thing in common: we believed two people could build something that thousands of players would love.

This is the story of how Goonblox Importers went from a shared Discord call to a live multiplayer game with 22 pet families, 4 explorable zones, and a community growing every day.

Starting From Zero

The Roblox developer ecosystem can feel overwhelming. There are studios with 20+ engineers, millions in funding, and years of experience. We had none of that. What we had was a simple idea: what if a merge game could also be an idle game?

Merge games are satisfying because of the dopamine loop — combine two things, get a better thing, repeat. Idle games are compelling because they reward you even when you're not playing. Nobody on Roblox had combined the two well. That was our gap.

We started with a 5x5 grid, a puppy model, and the merge mechanic. Within a week we had pets combining through 4 tiers. It was ugly. It was buggy. But when we watched the first pet evolve from a tiny puppy to a larger, glowing version — we knew we had something.

The First 100 Hours

Building a Roblox game as a two-person team means doing everything. There is no "I'll handle the frontend, you handle the backend" — both of you do all of it.

Our first 100 hours of development taught us:

• Luau is forgiving. Coming from other languages, Roblox's scripting language was surprisingly approachable. The documentation is excellent.
• Client-server architecture matters immediately. Unlike single-player games, every action in Roblox needs to be validated on the server. We learned this the hard way when our first playtest had someone giving themselves infinite coins.
• Roblox Studio is powerful but quirky. The 3D building tools are incredible. The scripting IDE is... functional. We learned to love external tools early.
• Playtesting reveals everything. Features that felt brilliant in isolation fell apart when a real player touched them. We started playtesting every day.

From One Zone to Four

MergePets launched internally with just the Meadow — 7 pet families (Puppy, Kitten, Bunny, Hamster, Bird, Turtle, and the beloved Frog Wizard) on a sunny green field. But we always planned for more.

The zone system became the backbone of progression:

• Meadow (free, 1x CPS) — Where everyone starts. Friendly, bright, forgiving.
• Jungle (T5 sacrifice + 50K coins, 2x CPS) — Dense canopy, exotic pets. Your first real milestone.
• Arctic (T6 + 500K coins, 4x CPS) — Ice caves, snow, and Huskies. The difficulty ramp hits here.
• Desert Pyramid (T8 + 2M coins, 8x CPS) — A full Egyptian pyramid you explore from the inside. 980 BaseParts, zero imported meshes. Plus a secret chamber with a 1M coin reward.

Each zone took weeks to build. The Desert Pyramid alone required designing columns, braziers with animated fire, hieroglyph panels, and a lighting system with spotlights casting through windows. All from primitive Roblox parts.

Systems on Top of Systems

What separates a prototype from a game is depth. Over months, we layered system after system:

• Rarity system — 6 tiers from Common (70%) to Mythical (special). Plus Golden and Rainbow variants.
• Combo system — Chain same-family merges for escalating multipliers and gem drops.
• Smart quests — AI-adapted daily quests that detect whether you're a Collector, Grinder, Explorer, or Competitor, then serve quests you'll actually enjoy.
• Home building — 5 architectural styles, 3 upgrade tiers each, furniture with CPS bonuses, daily bonus objects.
• Pasture system — Place pets in pastures for passive income. Feed them, groom them, watch them generate coins while you sleep.
• Leaderboards — Per-zone merge count leaderboards with physical statue pedestals for the top 3 players.

Each system feeds into the others. Merging creates pets. Pets go into pastures. Pastures earn coins. Coins unlock zones. Zones have new pets. The loop is self-reinforcing.

What We'd Tell Other Small Studios

If you're two people thinking about building a Roblox game, here's our honest advice:

1. Start with the core loop. Don't build 20 features. Build the one thing that makes your game feel good. For us, that was the merge animation.
2. Playtest daily. Not weekly. Daily. Watch someone play your game without helping them. The things that confuse them will surprise you.
3. Build security from day one. Every RemoteEvent needs server validation. Every currency transaction needs a mutex. Every action needs rate limiting. Exploiters will find anything you miss.
4. Ship fast, iterate faster. Our first version was embarrassing. But it was live, and real players were giving us real feedback. That's worth more than months of polishing in private.
5. Stay fair. We made a deliberate choice: no Robux gambling, no pay-to-win, transparent odds. Players notice. Parents notice. And it's just the right thing to do.

What's Next

Goonblox Importers is still just two people. We're still shipping updates every week. We have a weather events system ready to launch, a merge race PvP mode in testing, and plans for seasonal events throughout 2026.

The Roblox platform has given two people with no prior game dev experience the ability to build something real — something players come back to every day. If you've been thinking about building your own game, the barrier to entry has never been lower.

The hard part isn't starting. It's shipping.