Notes from Jason Frieds speech; How to make big things happen with small teams
4 guys, 3 products, 200000+ users
- Basecamp
- Backpack
- Ta-da List
Reducing mass
“The more mass an object has the longer it takes to change direction”
- Many employees
- Software lockin
- Longterm contracts
Embrace change
The right people
- Passionate, happy
- Well rounded - have to be good at most things
- Quick learners
- Trustworthy, reliable
- Good writer
- Communication is written!
Happy and average > Guru and disgruntled
Being small has advantages
- The customer is closer, less middle
- Everyone is on the frontline
- Change is easy
Act your size
You’re small, act like it, don’t pretend. Let the big guys be confusing.
- Less formalities (procedures)
- Less mass
- Less fear
- More flexibility
- More change
- More freedom
Embrace constraints
The only time you will really come up with the great stuff, is when you have constraints - things you can’t do.
Example: Basecamp
- Prior commitments (clients)
- 7 hour time difference in team
- Embraced this by turning those extra hours into “alone time”.
- Self funded
- Spend ressources more wisely.
- Less people
- With more people, you’re probably trying to do too much anyways. Take stuff out!
Make things manageable
Hard to stay manageable while small
Lowering the cost of change
Build half a product - not a half-assed product
There’s unlimited time in the future to add new stuff, and we’ll be more experienced
- Say no by default
- Listen to the product
- Let the customers remind you what’s important
- The feature requests you get many times are the ones that matter
- Ignore details early on.
- You will have plenty of time to polish the product after you launch.
- Improve what you have
- Only when you can’t improve existing features, add new features
- Decisions are temporary.
- You should always be able to come back to revisit a decision
- Nothing should be final (embrace change)
Build less software
Go for “good enough”, not “the biggest and the best”
- Less features
- Lower cost of change
- Simple software is easiest to change.
- Less room for error.
- Less support required
- Encourage human solutions
- Give people just enough to solve their problems their own way, then get out of their way.
- When features are general enough, they don’t lock the user into a certain pattern.
Get real, start with the UI
There’s nothing functional about a functional spec. The spec isn’t clear, it’s hard to describe what you want. Start with the customer experience. Textual specs are illusions of agreement.
- Start designing
- Start prototyping
- Start experiencing
- Start changing
- Rinse and repeat
Make decisions just in time
Less decisions to adhere to means less mass means easier change.
A problem isn’t a problem until it’s a problem.
Make decisions when you have real information.
Example: Basecamp
Basecamp couldn’t bill people at launchtime, since they had 30 days until the first billings had to happen. Excellent constraint.
Turn big into small
12 week project vs 12 1 week projects.
- Break into small chunks
- Finish
- Iterate
- Celebrate
Feel the hurt
Get the tech guys, builders, dev to support the product. You need to know what frustrates people. Feel the people when people complain. Shared annoyance. Annoyance leads to actions.
Publicity amplifiers
- Feature food
- Rave about kickass features, feed them to people, let them boast it over the Internet for you.
- Example: iCal subscriptions in Basecamp was served to Mac sites
- Example: RSS Feeds in Basecamp was served to blogosphere.
- Promote through education
- Tell people how you did things on your blog, on conferences, etc
- Example: Yellow Fade Technique.
- 30-day major upgrade
- Update soon after launch
- Show there’s momentum
- Make people feel they get extra for their moneys worth
- Not enough features? Hold something back!
- Transparency = trust
- Tell people when you suck, tell people when you rock