About ROD FERGUSSON – worked with Microsoft for 10 years, produced Microsoft training simulator (1 million in sales), Xbox games, most recently with Epic doing Gears of War for Xbox and PC
Martin Luther King Junior: “a genuine leader is not a searcher for consensus but a molder of consensus”
What is the definition of ‘Producer’ ?
- Trip Hawkins established the title when he founded EA in 1982
o Compared Producer as a meteorologist – aka paid to predict the future.
o Risk assessment, risk management.
o Barometer indicator of future state.
o Thermometer indicator of current state
o Producers sometimes don’t understand changes are based on where oyu are in the project.
o Preproduction establishes risks, allows you to add remove resources. (Covered the basic 101 iron triangle of scope, schedule and resources).
o It’s not all about the schedule. Schedule can sometimes slip months in a single day.
o If you have 12000 bugs in database, and you think you are on schedule, you are on crack. You are not on schedule!
In any schedule Rod creates, he pads a week within each milestone to sit back and identify issues and future risks.
Need to avoid the LA story effect (pre-recording the weather) – don’t lose site on the bigger issues such as the quality of your project.
Probability of Precipitation!
• Confidence that precipitation will occur and the percent of the area that will receive it.
• Use the past to try and predict the future.
• Evidence-based scheduling. Looking at the time it takes from start to end, account for all the distractions (meetings, sick days) and treat it as part of the task. After this they come back to a confidence factor. Eg. There is an 80% chance we’ll make our date.
Long Range Forecast
• Current state of the art for weather is 10 days. What about producer? For Producers it is 9-12 months.
• Longer the duration of a project, greater the uncertainty. That is why we use milestones to chunk it up into safer project achievements (eg 4-6 weeks).
• Buffer the heck out of the dates! Compared it to asking a weather man what the weather would be in 3 months time. He is adding padding to give room for error. Longer the duration of schedule you need to add buffer. Allow for variables.
Cut Early, Cut Often.
• Due to unknowns, teams tend to underestimate effort.
• Every feature takes polish time away from other features.
• If a game has a thousand features you have to concentrate on too many. If you reduce your feature set, you actually have time to concentrate on more features.
• Avoid sunk cost fallacy by looking at opportunity cost. Don’t get caught up in what it cost or how long something took to create, what is the cost of it in the future? Don’t keep a feature hanging on just because it was expensive or time intensive. It costs you in the future and across all the other features.
Fixed Ship Date Mindset
• Whole team must agree that this is an achievable date. Then hold it as unchangeable and fixed.
• It provides an END DATE for the team.
• This provides a clear goal, they knew it was going to end.
• This forces prioritization and creativity.
• The belief that the ship date is both realistic and unchangeable.
• Date can change but MUST be justified.
• Date not actually fixed until very confident.
Lead From the Front
• To be a leader means to be involved all the time and lead by example.
• When you first come to a project it is important how people perceive you.
• Some producers can be schedule nazis, too involved with red tape, but not actually helping the team.
• Your role is to do whatever is required of you to make sure the job gets done.
• Your job is to ensure the team can ship a game. A producer's job is NOT about overseeing a schedule.
• Never ask anyone to do something that you are unwilling to do. Eg. Don’t ask someone to work weekends if you aren't willing to be there on the weekend yourself.
• There is a difference between responsibility and authority. Lead with a carrot not a stick.
• Build respect and trust by adding value. Managing spreadsheets, helps you but not the team. This perceived value from the team is different. What can you do for a team that can add value to their job?
• Moral support counts. Simple as being there, getting them a coffee, ensuring they get sleep!
• Eg. Rod's policy they must spend at least 8 hours out of the office. Personally kick them out of office!
• There’s nothing too small you can do for your fellow team member.
Facilitate Communication
• Don’t facilitate meetings just for your needs, you must judge if it is valuable for others too – perhaps their time is better spent working than meeting.
• Get a base knowledge of other disciplines. Understand what they do.
• Listen more than you talk. Adapt to your team’s skillset, needs, what is going on in the team.
• How is this project needed to be run?
• Represent clearly what you know and what you don’t. Don’t be shy to say if you don’t know something. Helps you learn! Don’t pretend to be something you are not.
• Communication can be as simple as walking from desk to desk. Don’t assume your team are talking to each other! It’s not all about moderating large meetings, it's about facilitating communication.
• Adapt communication style accordingly. Not everyone communicates the same way. For example some people are fine in larger meetings, but others do better with one on one chat and analysis. Some people need information pulled out of them, others need to be corralled!
• “Straw Man FTW” – an attempt to pull information together with as much info as I have to distribute to other people
• Eg. Brainstorming – have base knowledge with conversations I have already had and then type up a proposal. Suggest the content. Email or propose to group then get them to discuss. At the same time they can comment and critique and you can learn.
Manage Expectations Along the Way
• Set clear expectations – difference between what you deliver and what you expect to be delivered.
• Don’t promise ten features when you actually deliver two.
• Eg. Don’t cut out 8 at once, manage expectations by cutting 2 at a time, so by the time the project is delivered people expect only two features. (aka keep people in the loop!)
• Constant two-way communication. Empowers the team to help make decisions.
• Choose milestone ‘spirit’ over ‘letter’. The whole reason for a milestone is to achieve a top level goal. By meeting the letter, it means you are paying it later and not now.
• Don’t exaggerate, stick to the facts. Exaggeration is the crying wolf syndrome. Focus on the facts and present those.
• Don’t overact. Always look for the compromise. Don’t be a chicken little!
Maintain Your Perspective
• One of your unique strengths is your ability to see the big picture.
• Don’t get passionate or flip out. You are the moderator, and your job is to resolve conflicts.
• Microsoft project and excel are your friends – they are your reality check.
Suggested read: “Dynamics of Software Development” by Jim McCarthy
Q&A's after presentation:
Question: Difference between technical project manager and producer?
A: One involved in creative, one involved in process. He says sometimes when you combine into one role only one of these perspectives will win. Epic only has one producer though.
Q: Ever looked at using Scrum?
A: Yes they can add sets to process. He bought a hybrid waterfall model.
Q: As an EP what’s your role in creative side? Are you moderating creativity or making game design decisions?
A: Epic is unique – recent game was created by a design committee – it’s about key stakeholders, and he is one of them currently. It comes down to trust and respect. You can’t assume you have that responsibility.
Q: Difficulties of upper management and producer?
A: He hasn’t – he likes staying hands on, not just watching other producers make games. Best way to form a bond with your team is to be in the trenches with your team.
No comments:
Post a Comment