How to Set Your Game Up for Success (Pre-Production & Forecasting)

Most game development pain is self-inflicted. Not from a lack of talent or ambition, but from entering production without having answered the questions that pre-production exists to answer. Levels get thrown away because player abilities were never finalised. Features get cut because scope was never properly calculated. Games ship at a fraction of their potential, or don’t ship at all.

It traces back to one phase being underestimated, rushed, or skipped entirely.

I sat down with Riad, CEO of Codecks and veteran game developer, to break down what pre-production actually is, why studios keep getting it wrong, and how to turn it into a forecast that holds up when production gets real.

What Makes Pre-Production a Distinct Phase?

Pre-production isn’t just “the phase before you make the game.” It has a specific job: removing uncertainty.

Not just about what the game is or where the fun lives, but about how you’re going to build it. How long does one level take at near-final quality? What does your character pipeline actually look like, end to end? How many animation resources does one enemy require?

The tool for answering these questions is a vertical slice, a thin but complete cross-section of your game built at close-to-final quality. Think of it like cutting a cake from top to bottom, you get a taste of every layer, not just one.

The vertical slice isn’t just a demo or proof of concept. It’s your pipeline data. Once you know it takes eight weeks to produce one level with your current team, the math becomes straightforward. If you have 80 months of production runway, you can calculate how many levels, characters, and items are actually within reach. That’s your real scope, not the wishlist.

So Why Do Studios Keep Skipping It?

Because the conditions that make pre-production hard are structural, not just cultural.

Publishing contracts take around six months to close. That means you need to start the process half a year before your current project ends, and half a year before that for pre-production. The people you need for a vertical slice are often still finishing something else. The timing rarely lines up cleanly.

Then there’s the other pressure: feeding the beast. When you’ve got a team staffed up and burning budget, there’s enormous pressure to start producing something, even if you haven’t answered the core production questions yet. So you start anyway. And the cost of that shortcut comes later: rework, wasted assets, scope collapse.

It’s not that studios don’t understand what pre-production is for. It’s that the business conditions make doing it properly genuinely hard.

How Do You Translate Pre-Production Into a Forecast?

The vertical slice gives you pipeline data. The next step is organising your project in a way that lets you actually use it.

Riad’s recommendation: structure your backlog to mirror your game. Instead of a sea of disconnected tickets, organize by entity type, characters, levels, items, systems. The way players experience your game is probably the clearest mental model for how to think about your scope.

Once your backlog reflects your game’s structure and your pipeline data tells you what each entity type costs to produce, you can start calculating resource needs by department. Where are the bottlenecks? Where do you need to bring in outsourcing? What has to get cut if you’re short on animation resources but long on engineering capacity?

That’s forecasting. And it only works if the underlying structure is in place.

What Should Teams Track Once Production is Running?

The forecast you build in pre-production is not a document you file away. It’s a living tool.

You don’t need to track time to the hour to make this work. Even something as simple as counting how many assets get completed per sprint is enough to validate or challenge your pipeline estimates. The point is to continuously update your scope understanding as real data comes in.

If you planned 30 levels in a year and one month in you’ve completed one, you have a decision to make, improve the pipeline, or cut scope. But you can only make that call if you’re tracking.

There’s a useful balance to strike here. More planning isn’t always better. At some point it becomes busywork. Apply estimates and intuition where it makes sense, but validate them as you go. The goal isn’t a perfect plan; it’s staying close enough to reality that course corrections are manageable.

Why Doesn’t the Current Tooling Support This?

Because most tools weren’t built for games.

General purpose project management tools are customisable enough that studios can bend them toward game dev workflows, but there’s a meaningful difference between “customised for games” and “built for games.” The former leaves every studio to figure it out from scratch. The latter understands the problem from the start.

In practice, a lot of producers end up doing forecasting work in Excel: exporting data, building formulas manually, reconciling numbers that go stale as soon as something changes in the project. The result is that something that should be a weekly habit becomes a monthly ordeal. And when updating your forecast is painful, you do it less often, which means you course correct more slowly, which means more compounding damage before you catch a problem.

This is part of what Codecks is working to solve with its forecasting feature. The idea is to bring together team velocity, game entity structure, and pipeline definitions into one place, so you can see exactly where your bottlenecks are and what adjustments to make, without needing to wrestle data into a spreadsheet first.

The Bottom Line

Games are one of the most complex development processes out there. You’re projecting 18 months ahead, managing cross-discipline pipelines, building assets at scale, all under the constraints of a real budget and a real deadline.

Pre-production exists to reduce as much of that uncertainty as possible before production begins. Studios that invest in it properly, building vertical slices, identifying pipelines, structuring their backlogs by game entity, and building a forecast from real data, enter production with a real roadmap. Studios that skip it enter production hoping for the best.

As Riad puts it: a lot of games don’t fail from a lack of talent. They fail from a lack of planning and scope control.

If you want to dig deeper into Scope-Driven Production and how Codecks approaches forecasting, the video above covers it in full. And if you’ve been wrestling with forecast hell in spreadsheets, we’d genuinely love to show you what we’ve been building.

Have questions about how pre-production works in practice? Come find us in the Codecks Discord, the conversation is always open. 🃏