Due to their ever-changing nature, game projects are incredibly difficult to plan. By now, most gamers are very familiar with the frustration of having their most anticipated game delayed and pushed back multiple times. In such situations, no one suffers more than the developer; publishers rarely pay for extended development.
Games vary in size, and teams vary in velocity. Still, the main production phase can span anywhere from several months for some of the smaller mobile games, to three or four years for big PC and console titles.
Game projects are divided into specific phases and milestones, and each milestone has a set of criteria that has to be fulfilled. If the project is being funded by a publisher, the developer will only be paid once the milestones are delivered, reviewed, and approved by the publisher.
The production process allows for better structure development, estimating its costs and increasing the chances of finishing the product on spec and on time. Unfortunately, big design pivots, unforeseen technical issues, licensing problems, and financial pressures are commonplace in our industry. Experienced producers always push for a buffer of an additional 15-25% development time on each milestone, and quite often that still isn't enough.
Note
These production processes are essential even if you're working on your own or as a part of a tiny independent operation. A set of deadlines and even loosely defined short and mid-term goals to work towards will help you focus, stay motivated, and increase the chances of finishing the project.
As we already know, game development is risky and expensive. To minimize the risk, during the life cycle of a project it will likely have to go through at least one greenlight gate—a point at which the fate of the game is being decided. A failure to greenlight will force the team to go back and iterate on the idea or result in the game being canceled altogether.
Before real production starts, game designers work with their producers to create and present the initial greenlight documentation to the key stakeholders in the company, hoping to convince them that the concept being proposed is a wise investment of time and money.
Once development starts, a version of the game itself is what's being shown. It's a common practice to start the project with the aim of spending the first several months on creating a so-called Vertical Slice. Vertical Slices are essentially demo versions of the main game hoping to showcase its potential, prove the artistic vision, and validate basic gameplay mechanics. Think of them as vertical slices of a cake: it may cover a very small area of the final product but contains all major ingredients (game systems are essentially horizontal layers). It serves as a good indication of whether it's worth the commitment to make the whole cake/game.
It's not uncommon to use Vertical Slices to present the game to the press, create teaser trailers, as well as to try and get external investors or publishers on board with funding the project. However, at diligent studios, many games will get canceled before they get that far.
While each studio and game producer can have their own way of running a project, a variation of the traditional set of milestones and production phases (borrowed from the movie industry) is employed by most:
- Concept phase: An idea is born! A concept document (usually in a form of a 5-20 slide presentation) is created and pitched to key stakeholders. If the project is given the go-ahead, the initial preproduction team is assembled and tasked with expanding the concept.
- Preproduction: A crucial period where the most important design decisions are likely to be made. Core gameplay mechanics are being validated by prototyping. The game's scope, art direction, technical requirements, production schedule, and team size are all being established. At the end of preproduction, the GDD should be finalized by the designer and approved by key stakeholders.
- Production: The team begins to execute on the agreed design, writing production-quality code, art assets, and content. As a rule of thumb, the further the game is into production, the harder it becomes to make large changes in product design.
- Pre-alpha: Depending on the length of the project, several interim production milestones are usually set, with the aforementioned Vertical Slice often being one of them. These give the team a defined mid-term goal to work towards and are useful even if there is no publisher (and therefore payment) associated with them.
- Alpha: At the end of the Alpha stage, the product should be feature-complete, meaning the game is playable from start to finish (should it have one) with all functionalities and content roughly in place. That said, the quality will be far from final, with many bugs left to be fixed and various improvements and changes to be made, often based on the results of playtesting.
- Beta: Beta represents a much more complete version of the game. In theory, all of the content is locked in place and the only changes being made from there on are bug fixes, balancing changes, tweaks, and polishing. Some companies will conduct public beta tests that are either closed (invite only) or open to everyone.
- Gold candidate: Once all important issues have been addressed, a release candidate can be approved by the publisher to put the game on the path to distribution! The gold status itself goes back to the old practice of creating a GM (Gold Master)—a version of the game that would be signed off and used for mass duplication of the final product.
- Release: It's time to celebrate! Your game has beaten the odds! Making games is hard and expensive, and the vast majority never see the light of day.
- Post-release: Depending on the post-release support plan, the team will either drop the milestone structure and handle improvements and additions on a sprint-by-sprint basis or create additional milestones around the creation of DLC (downloadable content) and larger expansions.
Let us delve into a slightly less traditional and more experimental approach to game production. While most game studios avoid talking publicly about unreleased and failed game projects, companies such as Supercell and Wooga often talk about the huge amount of games they have killed (canceled) at various stages of their life cycle. Some projects get canceled as late as during a soft launch that is, a test release in one of the smaller market such as Canada, Australia, Philippines, or the Netherlands.
Isn't canceling 9 out of 10 games hugely wasteful? And why would a nearly finished game ever get shelved?
Companies that adopt this model often operate in the mobile market, where the costs of marketing and user acquisition far outweigh the entire development cost. As the mobile market is a hit-driven one and studio resources are limited, to release a poorly performing product would mean to not work on something with a potentially much bigger upside—an opportunity cost, you may say.
The process of gradually culling less promising projects is what validation funnel is all about, and there's much to learn from it, no matter the type of game you work on and the markets you operate in. For one, games are never sure-fire hits. To allow for a high cancellation rate (especially in the early stages of the development) is to enable your team to take more risks and be creative. It's also important to give the teams the power to kill their product, rather than to have that decision flow from senior executives—the former is empowering and inspiring, the latter antagonizes and demotivates. If the decision has been made well into production, a postmortem presentation should follow. The aim of the postmortem is to analyze the production process, explain the reasoning behind key decisions, and share any learning from the project across the entire company.
The following is the validation funnel and game development process employed at Space Ape games at the time of writing. At the core of the process are small teams supported by shared company resources and outsourcing. While this funnel is focused on free to play mobile games, a similar approach could be taken with any digital product:
- Ideation: At Space Ape games, roughly a day per month is dedicated to an initiative during which the company can come together to form self-directed mini-teams that collaborate on something outside their day-to-day duties. This initiative often has a set objective and creative constraint; it may be around T-shaping, branding, improving existing games, or coming up with new ideas and prototypes. If new games are the focus, the lineup of game concepts and prototypes is voted on by the wider company. Ideas that are popular and deemed viable are then expanded upon and taken into preproduction, or put in the backlog awaiting a more suitable time. The key differentiation here is that new game ideas are not dictated by the executives or creative directors; they form and gain traction organically from within the whole company.
- Preproduction: A team is formed around the idea and works on its design. Core game loop, game pillars, target audience, brand, art, and technical direction are all being defined. The preproduction phase of Space Ape games is kept short and often ends with market sizing.
- Market sizing: This validation phase usually includes market research (looking into industry trends and competition) as well as testing our idea for a brand and potential user acquisition costs. To do the latter, we will often create a set of test advertising campaigns using a few potential art styles and brands for our game. The game will only pass if there's space for the product in the market and potential user acquisition costs are acceptable. It's possible to delay market sizing and begin the work on the internal prototype, but ultimately, all games need to make sure they can acquire an audience if needed.
- Internal prototype: The game concept is now ready for execution and a more polished and feature-rich prototype is being created. The team will now extensively iterate on gameplay and gather feedback from internal and external playtesting. At the end of this phase, a company-wide test is held. If the team is satisfied with the feedback and believes in the product, and if there are no red lights on the horizon, the game idea will live on!
- Alpha: The team switches over to writing production-quality code and continues to develop and playtest the game. The difference here is that Alpha builds can be used to gather external validation. This only happens if the team itself decides to seek real-world insight into certain aspects (such as experimental controls or multiplayer code). In such cases, a test version of the game will be released in a small territory, often with a placeholder name and under the umbrella of a brand new publishing name. The only way to test unknown quantities is to release the game to a real market and see how it fares. As everything is subject to extensive changes and most of the game will still be missing, these early versions of the game do not allow for any in-app purchases and are not ready to provide real insight into important stats such as user retention.
- Beta: A natural expansion of the Alpha stage with production going full Steam ahead. As always, company playthroughs, playtesting, and external validation help to push the game in a more refined direction. At the beta stage, the metagame should be validated (at least internally); this means the inclusion of long-term progression and features focused on improving player retention. The game is still likely to undertake major changes and will abstain from including any in-app purchases.
- Soft launch: At this stage, the game is released in a few territories, with the official title and branding. The game will remain in soft launch for several months as the team works on additional features, balancing, and polishing the product and improving the KPIs (Key Performance Indicators). Such KPIs include user acquisition costs, player retention, purchase conversion rate, estimated LTV (lifetime monetary value of a retained user), game session length, and frequency.
- Release: As the game is nearing worldwide release, the company will work on the game's marketing and user acquisition strategies. By then, the platform representatives (in this case Google and Apple) might already be aware of the product and its upcoming release, increasing the chances of receiving the ever-elusive featuring (for example, a prominent promotional banner on the virtual storefront).
- Live ops: The worldwide release of a free to play game marks the end of the official development phase and the beginning of the live operations era. The game team will continue to work on designing and implementing new features and content, and the game's live operations managers will ensure a steady stream of engaging events. In this world, the release is just the beginning.
Thanks to market sizing, frequent playtesting, and multiple stages of internal and external validation, teams at Supercell and Space Ape games can take risks with their ideas, yet minimize the unknowns when launching a finished product. It may be difficult for a game to make it to the end of the funnel, but once it does, it's much more likely to become critically and commercially successful, warranting ongoing support and investment.
As always, it's important to note that no process is set in stone. While every phase described here is important, the funnel approach is likely to shift and evolve along with industry practices, the company, and its culture.
It's possible that a game project you'll be working on will have a publisher other than your studio. The relation between the publisher and the studio can either be internal (in which case they own the developer) or external (the publisher is connected to the studio).
Whatever the relation to the developer, the publisher will not only release the product under its own name and market it, they are also very likely to cover all of the costs of development. Deals and royalty structures vary greatly, but in most cases, publishers are the ones who keep the cash flowing between all involved parties and are the ones who are set to gain (or lose) the most. Publishing can get very complex; sometimes more than one studio will work on a single game, other times an external owner of an intellectual property (IP) will be involved (the IP itself can be a brand, a book, video game, or movie universe).
When a publisher is engaged in the project, the detailed milestone structure mentioned previously will be of paramount importance as it will carry financial outcomes for the developer. Publishers will have to approve the GDD and provide feedback on each and every milestone; this can, of course, limit the flexibility of the developer, but it will also help ensure the project stays on track and has a chance of being released. Sometimes, publishers also take care of initial ideation and seed the game pitch to the developer.