When it comes to idea generation, particularly in teams, we immediately think of brainstorming as the go-to technique to generate lots and lots of ideas. So far so good. But surprisingly few people know even the basic rules of brainstorming, let alone how this affects groups and is affected by groups in turn. To give you some examples that may sound familiar, here are things I've seen happening many times:
- Teams easily spend several hours meetings, particularly the first ones
- Discussion is open and wide but often jumps and back and forth between broad vision and minute details of features
- Teams saying that it's very hard to come up with new and original ideas for the same challenge and business case
- A lot of “Yes, but…” phrases
Sounds familiar? Well, I started reading and researching and it turns out there's a simple scientifically proved reason for it: teams suck at brainstorming. Badly.
But there is good news. First, brainstorming and idea generation is only one step in the whole creative problem solving process and teams are (or at least can be) substantially better at some of the other steps. Second, there are simple ways to mitigate the obstacles teams face in brainstorming and make them both more effective (more, better, variety of ideas) and efficient (faster) at this.
Given that this is a very common activity in the games creation process, I thought it might be useful also for others to summarize what I've learned about it. Not the least, because I am curious what others have learned, and discuss.
So buckle up and let's go!
Creative Problem Solving
Let's first make sure we are talking about the same thing - particularly when it comes to anything related to the terms “creativity”, “ideas”, and “innovation”, things easily get fuzzy. So here's my definition of creative problem solving (CPS):
- The goal is to develop the best solution to a creative problem (or challenge).
- Creativity is the production of original ideas.
- Innovation is the execution of these ideas.
- The best creative solutions are the ones that are both original and useful.
The word “useful” in this context means basically that it solves the actual problem and that it is feasible.
It all starts with a clear problem that you are trying to solve. Just think back for a minute how many times you have been in “brainstorming sessions” which ended up being a bunch of people talking about different things all along?
The whole CPS process goes through 4 distinct phases, all of which have different challenges and thus also different tools, advantages, and disadvantages for teams. The phases are:
- Problem definition
- Idea Generation
- Evaluation and selection of ideas
Let's go through them one by one.
Problem Definition in Creative Problem Solving
Simply put, teams need to spend time first on agreeing what it is that they are actually trying to solve. In reality, there are often multiple problems to solve. That's just fine. The important thing is to clarify what they are, make sure everyone understands and agrees on that, and write them down.
The whole thing thus boils down to these steps:
- Identify the problem.
- If more than one, break it down it several smaller problems till each one is tangible
- Write down a clear, concise problem statement for each
- Prioritize them
Writing them down is important: first, it makes sure that everyone understands the same thing; second, you might not get to each of them straight-away and you don't want to repeat the whole shebang three days later.
Once done, take a break! Often, even if teams do a proper problem definition, they jump straight to idea generation. Sometimes, that might work, e.g. when it's a fairly small and straightforward thing. Most times, it's not. Taking a deliberate break gives everyone the chance to actually think about the problem first, do some research and exploration, and come to the following brainstorming with a fresh mind loaded with inspiration.
I'd also suggest to time-box this session. Talking longer than 45m just about problem definition is a sure indicator of other communication problems. It also is a good way of gently pushing people back to the matter at hand whenever discussion derail.
A final thought: what matters in the end is finding a great solution to a given problem - not the problem itself. So as banal as it may sound, it might be better to not emphasize the word “problem” too much but instead defining a clear goal/objective statement. Especially when it comes to games, we don't really want to “solve a problem” - we first and foremost want to create unique, fun, awesome experiences - not just solve and tick off technical problems. Focussing on the positive helps.
Idea Generation (“Brainstorming”)
The Four Basic Rules of Brainstorming
The term “Brainstorming” was coined by Osborne in the 50ies. Weirdly enough, even though everyone does it all the time, most people I've asked are not aware of the basics that Osborne himself laid down 70y ago. So let's repeat those first.
- Defer judgement
- Freewheeling welcome
- Quantity desired
- Combining and improving encouraged
Not deferring judgement is the most common mistake I've seen (and frequently do myself!) in groups. Someone makes a “crazy” suggestions and milliseconds later sure as hell someone will say “that's bonkers”. In fact, practically all ideas are met almost immediately by a “Yes, but…". Rule 1 is super clear: don't judge while generating ideas. What we actually want from brainstorming is divergent thinking.
Freewheeling is something that we usually do well in games, but it's good to remind teams and individual that there is no such thing as a crazy or stupid idea. The important thing is to get them out.
And many of them! Quantity is desired. Why? Because research shows that quantity does in fact highly correlate with also finding a larger variety of possible solutions as well as a higher quality. So if you hear the argument “well, but quality matters and we strive for quality from the start”, kindly point out that quantity is proven to breed quality.
Finally, it's great to remind teams that it's not only fine but a great thing to just take someone else's idea, add something to or modify it, merge with another idea etc.
The Problem with Brainstorming in Groups
Back to the real problem: research has shown many times that just putting people in a room and let them “brainstorm” produces significantly less ideas of less quality than having people do it separately and then joining ideas (in research called “nominal groups”). How's that possible? There are four main issues that crop up in groups:
- Production blocking: in a nutshell, you cannot think, listen, and talk at the same time. While you are trying to think of an idea, someone else talking will inevitably derail your thoughts.
- Social Loafing: in groups, individuals can just cruise along assuming someone else will pick up the slack
- Conformity: imagine you brainstorm about a new puzzle game theme and the first person shouts out “Pink Elephants!” Your brain is immediately primed and it's much harder for everyone to think of anything else but pink elephants.
- Downward norm setting: especially when the same group comes together many times over long time, performance of the group is usually defined by its weakest member.
There is plenty of research demonstrating these effects over and over again. Some issues with the research remain though: first, research is usually done in controlled environments with very homogeneous groups (e.g. male students around 20y old), the problem the groups solve is often irrelevant by that group who also does not have to implement the solution. In addition, most research only studies fluency, i.e., the amount of ideas generated.
But let's not kid ourselves: the issue remains that unguided, left-to-their-own devices groups produce significantly less, less diverse, and lower quality ideas than the same individuals doing it separately.
Let's talk about how we can mitigate the issues we've now identified.
How to Improve Idea Generation in Groups 10x
Some of the literature is quite extensive on the subject but for the sake of brevity and clarity, I've filtered down the various recommendation to nice short list of five basic actions that I believe are most relevant and practical for games teams:
- Use a trained facilitator
- Follow and reinforce the basic rules of brainstorming
- Always use small & diverse teams
- Always prepare meetings
- Use Brain-writing (write & read)
A trained facilitator is the single most important thing you can do. He or she will make sure Points 2-5 happen. Here's the good news: “trained” does not mean “professional”, and the facilitator can be a member of the group itself (it's likely even better). Even spending half an hour talking with one member of the team through the rules of brainstorming and providing tips & tools can make a huge difference. It's also something people get better at quickly, since they have the chance to practice often. Doing short retros on what worked well and what could be done differently with a facilitator helps.
Following and (gently) reinforcing the four rules of brainstorming is particularly important when it comes to defering judgement. The facilitator should always be on the lookout for those pesky “Yes, but's” and gently, but firmly, remind everyone to not criticize ideas in that moment.
Using small and diverse teams always helps. Diverse teams will find it easier to build on top of other people's ideas and get inspired by a completely different way of thinking. So mix those developers, artists, and designers. Diversity is also more than just functions: men, women, nationalities, ethnics, ages etc. all add to diversity. When it comes to size, try to stay smaller than 8 people in the room. Communication difficulty grows exponentially and social loafing gets more likely the larger the group.
Preparing meetings sounds like such a no-brainer, doesn't it? So why does it happen so rarely? Here are some things you (or the facilitator or really any team member) can and should do before a brainstorming session:
- Make sure there is a clear concise problem/goal/challenge statement and that everyone who'll be at the meeting knows this beforehand
- Prepare a list of idea stimulating questions, analogues, inversions, etc. - anything really that can help the whole group to activate their associative memory to come up with more and more diverse ideas (particularly helpful when groups are stuck)
- Prepare any tools, methods, materials you intend to use in the meeting and be able to explain the what & why of it quickly
Finally, brain-writing. If you google the term, you find a variety of fancy this-is-the-one-way-to-do-it explanations. In reality, it boils down to a simple thing: have everyone write down there ideas on a piece of paper before talking or shouting them out. Whether this is done just separately in a round-robin fashion (write an idea on a paper, hamd to person to the right, get one from left, read, add) or any other way is, in my experience, not that important.
Using these five actions can easily tenfold the amount and variety of ideas - so research says. But even if it “just” doubles the amount of ideas, it's a huge boost for very little effort. As a manager, there are really only two actions:
- Give teams an overview of the issues with and methods how to improve brainstorming, and
- Train one (or several) members of the team to become better facilitators.
Final note: just as the problem definition session, it makes little sense to do brainstorming sessions longer than 30 to max 45m. Even if the team feels it's has more and better ideas to give, it's better to take a break. Fatigue, hunger, thirst are really creativity's worst enemy.
Evaluation and Selection of Ideas
A small, diverse team that is well prepared, guided by a facilitator, and using brainwriting can easily come up with dozens of ideas. Lots to digest and filter.
First, again it's recommendable to take a break between idea generation and selection. As a rule of thumb, I'd say that the longer the brainstorming session, the better (and longer) to do a break. It keeps the mind fresh and gives ideas time to sink in.
If there are a lot of ideas, it helps to first group, combine related ones, and discard close doubles. Doing a negative selection first might also be helpful, i.e., quickly removing those ideas the teams thinks aren't and cannot be made feasible at all. Even so, there's likely still a lot left to process. Before doing so, it helps to go back to the problem statement and remember that the best creative solution should be.
Ideas can range from very creative (original, new) to very conservative (common, known), and from realistic (connected to current ideas and knowledge) to unrealistic (or idealistic, i.e., unfamiliar and disconnected from current knowledge). The second axis is often overlooked.
The best ideas are the ones that are both highly creative and realistic. This will ensure something highly original and feasible. The ones we surely do not want to waste time on are ideas that are neither original nor realistic (dubbed conservative conservatism).
In between, we have conservative realism and creative idealism. In the first case, it's ideas that are definitely feasible but which don't appear very original. In this case it helps to ask which particular aspect makes it so boring and common - and what woud be the simplest thing we could do to give the idea a twist and make it original. It often does not need much.
In the case of creative idealism, the idea clearly very original but nobody thinks it's feasible; or it can be something that just feels very unfamiliar to players. I've noticed that many such ideas come up when teams try to come up with “something different” and forget that the goal is to get something meaningfully different. In any case, here the question should be what the biggest blocker for implementation would be and how we coud make it a simpler solution without killing it's originality. Likewise, for ideas that really original but somehow feel wrong (remember the California Role Rule!), ask how you could do the same concept in a more familiar way. To illustrate later, imagine that someone puts forward the idea of a new blocker in Candy Crush described as a “black hole with laser guns (in 3D)". I think we can agree that this would be highly original for a switcher game - but at the same time woud feel very weird and unfamiliar in the Candy world.
Given this image above, here some tips how to quickly sort and filter through lots of ideas (usually on Post-Its) using swap sorting:
- Focus on one axis first, e.g. creativity
- Take a first post-it and, as a group decide where to put it on the horizontal axis. Don't worry too much about scale at the start
- Take the second post-it and compare to the first one: is it more or less creative? Again, take a quick group discussion
- Take the next post-it and starting from the most creative until the team agrees where to put. Same rating is fine, just stick them on top of each other
- Repeat till you've ranked all post-it's
- Never spend more than 30s per post-its. What matters is identifying on which end of each scale an idea is put, not whether it's #2 or #3.
Once done, repeat the same process for feasibility (i.e., rank from unrealistic to realistic), this time moving the post-its up and down but leaving them where they are on the x-axis.
Done? Then first immediately discard all post-its in the lower left quadrant. They are just mental ballast.
Second, if you already have enough great ideas in the upper right quadrant, I'd only focus on these. In reality, teams often rank “creativity” and “feasibility” against each other and you find post it's on a diagonal going from top left to lower right. If so, start with the ones closer to the upper right and ask the questions on how to improve the idea to make it either more original or more realistic. Use, combine, mix & match other ideas whenever possible for this.
I don't think there is much need for a deep process once you get down to the last few ideas as long as eventually the team makes a decision. Execution will quickly show whether something works or not and it's good to have alternatives identified.
Of course, there are many different ways to facilitate filtering and selection. What's important is that you prepare one and keep trying out new things to see what works for a given team.
To wrap up, I'd again recommend keeping this meeting time-boxed. This focuses everyone to make decisions. It's very easy to enter into long, painful discussions of minute details! This does, in my experience, rarely improve the selection process. In fact, some research even suggests that groups might be slightly better than individuals in evaluating and selecting great from bad ideas - but not better than a random selection. In other words: a monkey could pick post-its blindly and you'd get the same quality!
But the jury probably is still out. In my experience, a small, multidisciplinary team is actually pretty damn good at spotting quickly what's feasible and what's something really cool and original. In fact, other research related to inventions and patents has also shown that “lone inventor” is a myth. So Go Team.
If you reach that point, then there is little left to say. Except that hopefully now, the team (hopefully) has
- a very clear, common view of what the goal is that they want to achieve
- high confidence in the originality and feasibility of their concrete idea
If so, best to get out of their way and let them do their magic. If not, I'd start again at step 1.
So far, I've heard two typical counterarguments against the whole process.
“But we don't have time to do X meetings!!”
This is a weak argument if you've just seen a team coming out of a 3h long session. In turn, if teams only spend 15m on the whole process, I'd argue they are either dealing with a very trivial problem or aren't making any effort to find a truly great solution and just go for the easiest way out. Or maybe it's time to reflect whether you are pushing and rushing the team when instead they should be given more time to do it properly.
“But we already do have and come up with many great ideas!!”
Good for you! All I am (and research is) suggesting that you can get even more, better, and more varied ideas in less time. Give it a try.
I hope that most of the above is as much common sense as it did to me. For me, it was great and useful to reaffirm and remind myself of some basic principles and simple actions that can hopefully help the small pitching teams to get even better at what they are already doing so well: coming up with great ideas for our new game. I am looking forward to seeing how my team will apply the above, what they learn and how they will improve it. And I am looking forward to comments, discussions, and suggestions.
Below some reading and resources I found particularly useful and interesting - in case you want to read more.
- Nice, practical tips and tools: “Creating Minds - tools, techniques, methods, quotes and quotations on all matters creative."
- “Using Brain-writing For Rapid Idea Generation”
- Isasksen and Gaulin, A Reexamination of Brainstorming Research: Implications for Research and Practice.
- Leigh Thompson, “Improving the creativity of organizational work groups”. Academy of Management Executive, 2003, Vol. 17, No. 1.
- Stroebe, “Relative accessibility of domain knowledge and creativity: The effects of knowledge activation on the quantity and originality of generated ideas”
- Nijstad and Stroebe, “How the Group Affects the Mind: A Cognitive Model of Idea Generation in Groups". Personality and Social Psychology Review, 2006, Vol. 10, No. 3, 186–213.
- Diehl and Stroebe, “Productivity loss in brainstorming groups: Toward the solution of a riddle". Journal of Personality and Social Psychology 1987, Vol. 53, No. 3, 497-509.
- Singh and Fleming, “Lone Inventors as Sources of Breakthroughs: Myth or Reality?". Management Science Vol. 56, No. 1, January 2010, pp. 41–56.
- Lehmann-Willenbrock1, Chiu, Lei, and Kauffeld4. “Understanding Positivity Within Dynamic Team Interactions: A Statistical Discourse Analysis". Group & Organization Management 1–40.
- Paulus and Yang, “Idea Generation in Groups: A Basis for Creativity in Organizations". Organizational Behavior and Human Decision Process, 2000, Vol 82, No. 1, 76-87.
- Kerr and Tindale, “Group Performance and Decision Making". Annu. Rev. Psychol. 2004. 55:623–55.
- Boudreau, Lacetera, and Lakhani, “Incentives and Problem Uncertainty in Innovation Contests: An Empirical Analysis". MANAGEMENT SCIENCE Articles in Advance, pp. 1–21.
- Goncalo and Staw, “Individualism–collectivism and group creativity". Organizational Behavior and Human Decision Processes 100 (2006) 96–109.
- Brown and Paulus, “Making Group Brainstorming More Effective: Recommendations From an Associative Memory Perspective". Current Directions in Psychological Science, Vol. 11, No. 6, 2002.