Kickstarter.com uses this model. Creatives post their idea and a total project budget that they feel will cover the cost of the endeavor. What's even cooler is that its fan-funded so when your product releases you have a built-in audience. Creators don't have to compromise their vision but the marketplace decides on what eventually gets funded. There's full transparency in the projects on Kickstarter but the creators go beyond that and actively engage the community that funded the project on a regular basis.

"But in the current climate where basic features (such as normal maps, physics simulation and so on) are so well-understood, do we really need to iterate so much? Do we want to play it so totally safe we never let our baby out the front door, or do we want to go forth and meet our fate?."

I think you are forgetting one key component that is addressed via iteration and that's usability. Throwing iteration out the window or saving it for the sequel is not something games can't afford as poor usability and poor user experience is what makes players drop the controller within the first 5-10 mins of the experience.

Well managed iteration is what I believe you are advocating.

This would work particularly well in grant-funded productions that do the planning/money-collecting up front. I like the transparency angle a lot and completely agree that iteration should not equal unbounded iteration, that a hybrid lean practice should still be in play, that we can use the pressure of a time limit to our creative advantage. But the biggest plus is simply in preparing to deliver work for pay equally in a way that promotes consistency and stability.

Nice, Tim. I'm officially stealing this idea.

I expect royalties, Glenn.

Excellent article, Tim.

As a project manager in the video game industry, I think that what ails us is not a lack of process, but rather a lack of planning. What you have described is similar to "traditional" project management in that the planning is done up-front: "here's what we're going to do, here's how much it will cost, now pay us and we'll go do it."

I do feel that the idea of "just making a to-do list" is a bit simplistic - I'm assuming you were being a bit sarcastic with that comment - but I agree with the core idea. Project management techniques and tools are simply scaffolding that should hold up the tent in which the tasks are done. Don't apply techniques that don't work simply because they are popular. Instead, do what makes sense to the team, fits the size and budget of the project, and Gets Things Done!

Yes Karen. The idea of a "to-do list" basically goes down to that point: plan it, then do it. Rather than: 1.) doing without much planning (aka crunching); or, 2.) pontificate on process and endlessly iterate.

I agree with the last statement. Doing up a good budget works.

On the other hand, milestones are really helping to keep the quality of the game that the publishers are paying for high. Seems to me that the project-based budgeting you're suggesting is just laying out your resources, how much they're worth and how long it'll take. Which seems like plain-old budgeting!

Maybe this whole ideal you describe, with the iteration and multiple games, would work better for a big company. Like in Hollywood where a studio can shovel out many movies and as long as one hits, the others don't sink them.

Personally, I find milestones to be a safety-net in an uncertain industry. Say my company has budgeted for 3 programmers, 2 artists, a designer and a producer for a milestone. They're paid and happy. If they do what was set out and complete the milestone, which normally has some leeway based on problems that may occur, everyone's happy and things continue. If there's a big issue, for developer or publisher, you can re-evaluate and decide what you want to do next.

Derek, the article speaks of "milestones" not from the perspective of project management, but from the perspective of budgeting and greenlighting. Even in a film, where there are no "milestones" in the sense of staged deliverables with installment payments, there are still detailed shooting schedules.

Ah, that makes more sense. And highlights why I'm a designer and not a producer. ;)

I think a lot of companies are budgeting along these lines at the moment. When looking at recent career opportunities around the industry, I regularly see funding-assured projects. For example, Eutechnyx has been given approx £10 million for their next project. You've got to assume that to secure that funding they used a model similar to what you proposed here.

It seems to me that the elephant in the room here is risk management. When a publisher funds a game's development, they assume the risk, so they get a lot of say over project management as a way to minimize that risk. That's reasonable; the problem is that the money people often also insist on meddling in creative decisions. Tragedy ensues.

In a new-company-for-each-game model, it looks like most of the risk gets shifted to the completion guarantor. So to understand how well this model might work for games, I think we'd need to see more than just one reference in one sentence to this part of the process.

For example, are there completion guarantee companies now who have worked with game development studios and understand the needs and difficulties of that business? How many such guarantors are there -- enough that they compete with each other, or just a few who can thus dictate "take it or leave it" deals to developers? How can developers contact guarantors, and what should they be prepared to present?

Finally, the question of iteration seems like exactly the right one to discuss with respect to shifting risk to completion guarantors. If they're going to say, "we'll insure you for X amount of money," then yes, that pretty much eliminates "we'll ship it when it's done and not before" as a development model. But while there surely are some developers who fritter away money on endless tweaks, we can also point to developers like Blizzard and Valve and Popcap whose determination to get it right has allowed them to make extremely high-quality games that appear to do very well for them financially.

Is it possible to structure a deal in the one-game-one-company model that allows for an extended polishing phase?

Bart, this thread isn't really about project management. It's about budgeting and the greenlighting process.

No one is saying there aren't milestones in this system. There just aren't "milestones" - as in the slang-term used in the game industry to mean payment that is only released upon certain stages achieved in the project schedule.

Remember: This system I'm espousing is much older than that used in the game industry. And if there was some defect of risk management in it, then how did all that innovation in film take place?

I really think this is where we're headed. There will need to be "studios" in the Hollywood sense; rooms full of computers and devkits for revolving projects. Like movies, I see a need for both permanent studio employees and nomadic contractors. Those "above the line" will be able to germinate projects at the planning stage and sell them to studios that make sense.

I think a lot of developers have a hard time seeing this as possible because it will be such a huge cultural shift. But it's really the best way to manage big entertainment projects as a business. Getting rid of killer burn rate will shift our practices toward better, high-concept designs and more established execution practices.

Your bit on "reducing burn" points out what I think is the most obvious flaw in this system - that it leads to massive job/income insecurity for all but a tiny handful of the people involved in a project. This may work for a few key creative/business types, but it is bad for just about everyone else. And I happen to hold this apparently strange opinion that it's a bad thing when people don't have a steady source of income. Short-term contract work (ie. temp work) works out poorly for most people who end up stuck with it.

Adam, I think that, instead, it will lead to is a system of modular outsourcing entities that can leave a project and move on to a different one fairly seamlessly.

First, this is an entertainment industry. Understand that. It runs on projects. Projects end.

Next, what is the best way to go from project to project? Either as: 1.) an employee of a giant studio who is laid off between projects and then has to go get a job in another similar entity, only to be laid off again when that project ends; or, 2.) as an employee of an outsourcing company that moves from project to project seamlessly, not getting laid off when Project A finishes, because Project B with a different core team and producer is already scheduled?

At the end no matter which way you chose to work, because we all know that games are never finished in time. As an old Orc once said: "work, work...".

On some levels I can see why project based companies would seem like an attractive proposal - transparency, greater publisher commitment and more creative control are all good things for the developer.

However when it comes to retaining value, growing your company and keeping the talent around required for success, you end up shooting yourself in the foot. I find it unrealistic that many individuals or small groups would freelance between projects within such a system because it would require disproportionate amounts of administration for each entity.

Essentially I see people/groups/companies following such a model barely surviving between projects, taking even longer than today to accumulate resources to be self-sustaining and treating their talent strictly as a resource, which in my opinion is an unhealthy place to be.

Also I could rant on and on about how planning ahead with the granularity needed to put together a tight budget along the lines described requires guesstimating to a unacceptable degree and reject rather than embrace the uncertainties inherent to game development. But i guess that discussion have been beaten to death many times over by now...

Well Thomas, this system has allowed core talent in film to thrive - Clint Eastwood, Steven Spielberg, Robert Redford, Paul Newman, Jim Cameron... (shall I go on?). This talent has not "barely survived" between projects. It has thrived, garnered fame, and gotten rich doing so. And the industry benefited from it.

What you describe is the old factory system, which is what film *used* to run according to, way back in the 1950s and earlier. That notion that you shackled your core talent inside a single studio, indefinitely, oblivious to hearing what *they* wanted to create, and instead beholden to serving the company's agenda of what was to be created.

I like the idea in theory, but this article misses the point of game development being a creative industry. The so-called Hollywood model works for them because you know there are a lot of items that are predictable in costs and can hire anybody from the Union who will work for those rates and the job will get done. Those are not extremely creative endeavors. The actors are the expensive parts, and those terms are negotiated separately. In the games industry, every artist, designer, and programmer is the equivalent of an actor in Hollywood, because their talents are what make the game good. Imagine what would happen to the Hollywood model if every grip and cameraman needed a separate contract negotiation where their costs varied as widely as your actors.

But taking a step back (from what little I know about Pixar), Pixar or any other CGI house seems to me much closer to a game development studio... they have programmers and artists that work project after project in-house. Their investment in tools and technology spans multiple releases. I have no idea what their budgeting process is, but I would imagine it's a blend between traditional Hollywood project-based budgeting and studio-based budgeting to cover overhead between projects. Of course, I would equate Pixar and so forth more with Electronic Arts and Ubisoft and other publisher/developers that have a lot of strength in their back-catalog income and have been very successful over their lifespan. MOST developers don't fit that description. The rest of us are younger (than EA), and are still working to secure our survival through the next project while building our core value, to secure larger deals in the future. Without a strong track record, investors (publishers) don't want to put money into a new game studio any more than they want to put money into a new director without a good reel. I think you can see how that hurts creativity and innovation, and is already largely responsible for the stagnation in both Hollywood and games. Going to a Hollywood model doesn't fix that, in fact, it probably makes it worse because track records in games are associated with organizations, when it's a group working together, not so much with individuals.

I think in a world where there is no residual investment in the technology--movie productions rarely require new cameras to be invented--the project driven budget makes sense. But once you step into the creation of value that exists OUTSIDE the final product, ie. physical inventions or intellectual innovation in code and technology, having a multitude of different entities for each project is a lot of red tape and licensing and paperwork, and does not improve the long-term viability of a team. Tight teams produce the best games--that should be a primary objective over anything else.

@ Jason: Only people who don't work in film believe that film people never invent new technologies. The reality is they do all the time.

Thanks for taking the time to respond Tim, it is always interesting to have an in-depth discussion of a relevant topic. Overall I agree with Jason's points and feel they complement my stand. But let's not keep it at that :)

Even though many game developers like to be inspired by the film industry - and they are arguably historically and technically related - there are some basic differences that by definition make development in the two areas differ too significantly for me to consider the project model you propose a "killer game biz model".

The most fundamental difference is that game developers are required to develop technical solutions in-house alongside the development of the aesthetic side of the game (the user experience if you will - LeBlanc & Hunicke & Zubek's "MDA Framework" defines it in terms of game mechanics, but aesthetics in games are hardly confined to interactivity, even though that is at the core of what make them stand out from "traditional" one-way communication media such as film, books and paintings). Even with 3rd party software you need code glue to hold things together and code to make gameplay related mechanisms work as intended. The combination of the aesthetic and technical rationales involved in the game development process is one of the things that make it really difficult to make a good game.

With film it is my impression that you *usually* don't base a project on the simultaneous development of one or more technologies. There are some obvious projects that stand out such as Lucas' Star Wars (early special effects), Cameron's Avatar (first with 3D galore) and Jason's mention of animation film studios such as Pixar. But *overall* I see studios invest in ready-made technical tools so they can focus intently on the development of their aesthetic output.

If you accept that there is a fundamental difference in the problems related to creating film and games, it seems clear that we must take that into consideration when defining how we work to solve said problems.

As Jason points out there is a vast difference in how many "super stars/prima donnas" you would hire for a film project and a game project. This could very well be related to the inherent difficulties/uncertainties introduced to game development by the struggle to combine and balance aesthetic and technical aspects of a project.

The point I was trying to make the first time around relies on the differences between making film and games. Divergent focus between opposing rationales make development problems more difficult to predict (apart from the fact that you are likely to have more of them with a higher complexity) which impacts the ability to plan your way out of longer development cycles. Great complexity and uncertain time scope inevitably affects budgets too (if neither is taken into consideration something is bound to go wrong anyway).

I don't pretend to have a ready-made solution to the problems I point out, but I stand by my opinion that project forms must take into consideration the characteristics of the media they intend to develop in order to be successful.

Thomas, you say the difference between games and film is X, and then describe something about games that works the same in film. It's clear you haven't worked in film.

The basic thing I would say is that if you want to see if something works, instead of hypothesizing why it won't - try it to see if it does.

Hi Tim. Do you have an email that you can be reached at? I have a question I'd like to ask you that isn't relevant to this article.