Getting a game finished doesn't mean the work is done. However closely you followed everything covered so far — planning, building, testing — you can still watch a genuinely good game get ignored, and a mediocre game with careless publishing get remembered mostly for its problems anyway. Publishing is the stage most first-time developers underestimate the hardest, because it looks like paperwork sitting next to the "real" work of making the game. It isn't. For a lot of games, especially the ones without a publisher steering that side, it decides whether the previous three stages of work ever reach anyone at all.

Everything here splits into three overlapping problems: how you plan to make money, how you get people to notice the game exists, and what you do once it's actually out.

Money first, because it should be decided early, not bolted on after the fact. A revenue model is just the mechanism by which your game turns into income — a straight upfront purchase, free-to-play with in-app purchases, ad-supported, or some mix of those. Whichever you pick reshapes decisions you'll make throughout production, not just at launch. A straight-purchase game gets judged almost entirely on how much content it delivers for the price, and that judgment is unforgiving: The Order: 1886 had genuinely striking visuals and set pieces, but a five-to-six-hour campaign at full price left most reviewers focused on length over anything else, and the backlash defined how the game is remembered more than its production values ever did. A free-to-play game with in-app purchases avoids that specific trap but walks into a harder one — you have to design an economy where spending money feels optional and fair, not extractive, which is a genuinely difficult balance most attempts get wrong. Supercell has built an entire company on getting that balance right across Clash of Clans and its other titles, and it's worth studying exactly why their systems don't feel hostile the way plenty of copycats do.

Once the model is settled, the game still has to reach anyone. This is the part where budgets get uncomfortable: on anything above the smallest scale, marketing spend now regularly matches or even exceeds what it cost to build the game in the first place. GTA V's combined development and marketing budget is widely put at around $265 million, and marketing made up close to half of that on its own — for a franchise that could, in theory, coast almost entirely on name recognition. If Rockstar still spends that much on getting attention, treating marketing as an afterthought for anything smaller is a mistake.

You don't need Rockstar's budget to take the right lessons from it, though. Building a presence on social platforms before launch, sending build codes to outlets and content creators who cover your genre, and pushing for decent placement on the storefronts you're launching on all matter more than any single big splashy moment. Emotional honesty travels further than most people expect, too — Moon Studios leaned hard into genuine streamer and YouTuber reactions in their marketing for Ori and the Will of the Wisps, letting real, unscripted emotional responses do work that a scripted trailer voiceover never could. None of this needs a huge budget to execute, just consistency and enough lead time to actually build an audience before the release date, instead of starting from zero on launch day.

Showing up at a game dev event or expo, even a small local one, does something none of that can: it puts your game in front of strangers who'll play it and tell you exactly what they think, face to face, without the filter that online feedback usually comes with. That's worth more than most developers expect going in.

Then the game ships, and a different kind of work starts. Problems that stayed hidden through your own testing and even your beta always surface once a large, varied audience gets its hands on the game — different hardware, different playstyles, edge cases nobody thought to check. How fast you respond to that matters more than most developers expect: plenty of players will drop a game at the first serious bug and never come back, and word about a slow or absent response spreads faster than word about a fixed one. This is also where the documentation habit from earlier stages pays for itself directly — hunting down the cause of a live issue is a very different job when you have records to work from instead of just people's memories of decisions made months earlier.

Maintenance isn't only about fixing what's broken, either. Plenty of it is about improving what already works: rebalancing systems that turn out unfair once real players stress-test them, and shipping new content to keep people around. No Man's Sky is still the reference case for what sustained maintenance can actually do to a game's reputation. It launched in 2016 to a genuinely brutal reception, criticized for promised features that simply weren't there. Hello Games kept shipping free major updates for years afterward — new mechanics, multiplayer that hadn't existed at launch, systems overhauled from the ground up — and the game's standing today looks almost nothing like it did in its first week. Very few launches recover from a reception that bad. The ones that do, do it through exactly this kind of sustained, unglamorous post-launch work.

A few things worth carrying into your own plan:

  • Decide your revenue model early — it shapes design decisions well before launch, not just the price on the storefront page
  • Start building an audience before release, not on release day
  • Budget real time and, where you can, real money for marketing — it competes with attention, not just with other games
  • Track live bugs and player feedback the same way you tracked them in testing, and respond fast
  • Plan for post-launch content and balancing from the start, not as an afterthought once reviews come in

Publishing and maintenance don't get talked about as much as the fun parts of making a game, but they decide whether anyone ever finds out the fun parts exist. That's the last stage in this series. Next time, it moves from theory to practice: opening an engine and actually building your first game.