Tuesday, August 3, 2010

Thoughts on Development

A former colleague of mine recently asked me some questions based on my experience running a team.  I thought the answers might be useful for somebody else so I'm posting here.

-What are some of the key pitfalls that you have learned from creating original IP as part of a large company?

It’s always a challenge creating something original within a large organization. Large companies are inherently risk averse, creating new IP is inherently risky. Attempts to mitigate risk (focus testing, market analysis, creating rigid processes for the creative process) can often strangle the creative process. This is why being small, hungry and far away from the mothership is preferable.

Another challenge is assembling a team that can be comfortable with the loose, organic flow of early development. I have seen people who are great in early concept development and people who are fantastic in a well defined production phase but its tough to find people that can be a part of the whole phase.

Don’t scale the team too fast, keep the team as small as possible in the early stages to give yourself time to really figure out the foundation of the concept.

-In your opinion, what creates success when creating and executing on original IP?

Think of the old adage ‘plant a seed watch it grow’. There was a great talk on NPR last year about creativity. The guy discussing this question said something I really took to heart ‘when you plant a seed, you don’t dig it up every week to see how it is growing’. Creativity gets nurtured at the early stages by putting together a key group of talented people, have them work on something they are TRULY passionate about, provide them with the time & resources they need, and then being there to SUPPORT them, not impede them. When this basic foundation is set, magic will happen. Autonomy and ownership of what is being created is key. See Dan Pink’s TED talk on the science of motivation.

-From a production standpoint, I know creating original IP can get out of hand, what production constraints would you set up front? The scope of original IP can get so out of hand, that I feel it is important to take it one step at a time while ensuring that key stakeholders understand that this process is iterative.

Keep the team as small as possible for as long as possible. Make sure you are doing gameplay prototyping and using that as your ‘design exploration’ Don’t focus too heavily on documentation. Use wikis, basecamp, google docs or anything that can live ‘in the cloud’ so that it’s easily accessible to the whole team. I think it’s key that the partners (publisher, investor) understand that the early phase is for exploration and not expect you to make long term project commitments at this stage, you need to discover what the project is going to be.

Do regular playthroughs, make sure the whole (small team) sits down regularly and reviews the work that they are each doing, sharing ideas to improve. Work needs to be exposed as much as possible not stay hidden on someone’s machine. On my team we had daily standups, 2 week sprints and the team themselves really owned the work they were doing.

Plan the project so that you have a minimum set of features that you can ship with and an optimum set. Make sure that you clearly know what the goal is of the project, time, budget or quality. If it’s time/budget, schedule by prioritizing and delivering the core feature set earlier in development than the gravy, and be prepared to cut judiciously early so you have time to make sure that core set is awesome. You don’t want to ship a wide feature set where everything is mediocre. If it’s quality, then stakeholders need to know that scheduling is not exact and build in margins of error (i.e. 15% contingency) within the time & budget. If the stakeholder wants all 3 to be equally important...look out!

-What kind of milestones would you setup to ensure the success of creating original IP?

I think there are 2 phases that typically get stripped out of development that are critical for new IP.
  • Early stage world & character development plus gameplay prototyping.
  • Polish (true dedicated polish, not bug fixing)
Also, schedule lots of playtesting of the levels and integrate it into the level creation schedule. Make sure you have iteration pass time allocated for after each focus test.