Tuesday, October 15, 2013
I Hate Meetings - Michael Lopp - Palantir
A bad meeting = Nth discussion on the same topic and no one owning the decision.
books by Michael
Bored people quit - that's what they do.
You are (at most) three years away from building something new. Look at resumes you'll see that people do this.
Irrelevancy is always just around the corner. And shows up usually when you're happiest.
Meetings exist to scale communication and to try to keep track of what might be about to go off the rails.
The process for making something happen = idea -> analysis -> execution
This process happens all in your head when it's just you. When it's two people you need to discuss it and sometimes it takes debate. What this is is a meeting.
The first meeting = it's not just you, together working to solve a problem, in a finite amount of time, and done as needed
All ideas get better with more eyeballs.
But now we have 30 people, not just two. Ideas are coming from every which way. Rapid and organic error correction. Low cost situation awareness. The most random shit becomes culture. Everyone knows everything and does everything.
This first round of 30 people represent the Old Guard. They don't write anything down and keep everything in their heads. They will likely have disproportionate power. Their instincts say that everything they learned at this size will work at a larger size. They run meetings based pretty much on the "first meeting" rule.
Now the company is larger and has 300 people. The company is working. Ideas are still showing up but there's lots of them, decisions are happening slower because it's hard to get input from everyone, and execution is now becoming stove piped. People are a little sad because they used to get to do everything put now roles are solidifying. Stories are retold and becoming myth. Situational awareness is now expensive. Increasing communication friction. Learning can no longer occur via osmosis (a clue to this is when people ask for a wiki). The New Guard is now here. It's stranger town.
The Old Guard can handle problems and fix things based on their political knowledge. They keep doing things the old way but this is where things go off the rails. The issue is that groups of people do not communicate at scale. We think we do, but we don't. The Old Guard just talks to the right people and things get done.
In this setup the Old Guard has no ability to anticipate problems because communication isn't scaling. Then inevitably something bad happens and someone says we should meet.
Meetings go viral when: the perceived value of the meeting goes up, it begins to seen as the only hammer available, and certain people are really good at running a meeting (they are meeting infectors and it is in their interest to exploit this talent).
The Mutated Meeting = Random people who aren't contributing, a lack of agenda, that doesn't respect attendee's time, that goes on forever and ever
Apple called people out of meetings who didn't contribute. And noise didn't count, it had to be signal.
What you need is consistent useful communication in all directions. There are three essential and good meetings that avoid things going off the rails. Preventative maintenance meetings avoid stuff exploding. The reward is silence which does take the fun out of saving the world when the sky is falling.
The one to one meeting
the update meeting = every week, no matter what, 30 minutes at least (when you skip it trust erodes). Cheat sheet - three prepared points, a performance review, the supervisor's current disaster. The topic should be what you care about. Assume someone has something to teach you in the one on one
the vent meeting = just nod and listen (don't get sucked in or commiserate, but lend an ear)
the disaster = you don't want to have this
Staff meeting - every week (early), no matter what, 30 minutes (at least)
Dashboard: At apple there is the ET which is about everything going on in the company. It sets the tempo for what will happen next week. It handles the state of business
Special guest meeting: information from others and there is serendipity where you learn more
Tapestry meeting - frequently, not just when it's on fire, possibly over poker
trying to build improbable relationships, culture, to get people to bond
serendipity focus
All other meetings except those above need an expiration date. Never have it recur forever. You need an expiration date.
Success is (confusingly) silent. These 3 types of meetings are a way to nip problems in the bud. You should find out who is bored and who needs honest feedback.
You can only manage 3 - 7 people. Beyond that someone is suffering.
Meetings should frequently justify themselves. Vote with your feet on whether a meeting is important. Prune aggressively and get rid of unnecessary meetings.
Leveraging Logic as a Leader - Patty McCord - Netflix
It's not what you aspire to do - It's what you actually do
Fundamentals
Company First (if you don't have a successful company you don't need a culture)
Judgment trumps everything (intelligence, speed, intuition)
You can just tell the truth
Teams: Visualizing Greatness
Imagine in six months if everything was amazing and you had the perfect team. What would be occurring then?
What would you measure (metrics, deliverables, reporting, cross functional alignment, timeline, deadlines, due dates)?
What would it look like (winning debates, winning tests, customer feedback, data & analysis, influence)?
What would people have to know how to do in order to pull it off?
What kinds of skill and experiences would lead you hear?
And only after that do you think about who is on the team.
To build the right team first you need to know what you are going to build.
If you start with who you have, you may be limited in what you can do.
You now can tell the truth.
Which means people can decide if they are motivated by the work.
They know what they can do.
They can own their careers.
They can opt in or opt out.
The steps above would allow for a world where there would be no need for performance improvement plans. In truth a PIP just exists to justify letting someone go. It is better to be honest about the work and needs of the employees from the start. Also the steps above would keep interviewing from being a sin. It allows people to be honest and for people to do their best work.
If an employee is unhappy you should tell them to go interview. It will make it easier for them sometimes to articulate to a complete stranger regarding what they want. Also it will let them know what market value is for their job. And they may realize that the grass isn't greener and they'll come back more motivated. The catch is that if someone comes back you need to say "what did they offer you" and you need to offer it because that's what "market comp" means.
Good interview question: tell me about the day you went home and said to yourself "damn I'm good"
Netflix didn't have performance reviews. They had gantt charts and knew who was performing and who was lagging behind.
Pattern recognition - how to lean people. Pay attention to patterns. Learn to observe. People are rewarded differently, wired differently, and need to be treated differently.
Learn to observe how people react to: change, challenge, the unknown, each other, unexpected interruption, learning
What Patty did to innovate within HR is to just stop doing things. No more PIP. No more vacation policy.
Humans are organized for efficiency. You cannot avoid organization, it will form.
Fundamentals
Company First (if you don't have a successful company you don't need a culture)
Judgment trumps everything (intelligence, speed, intuition)
You can just tell the truth
Teams: Visualizing Greatness
Imagine in six months if everything was amazing and you had the perfect team. What would be occurring then?
What would you measure (metrics, deliverables, reporting, cross functional alignment, timeline, deadlines, due dates)?
What would it look like (winning debates, winning tests, customer feedback, data & analysis, influence)?
What would people have to know how to do in order to pull it off?
What kinds of skill and experiences would lead you hear?
And only after that do you think about who is on the team.
To build the right team first you need to know what you are going to build.
If you start with who you have, you may be limited in what you can do.
You now can tell the truth.
Which means people can decide if they are motivated by the work.
They know what they can do.
They can own their careers.
They can opt in or opt out.
The steps above would allow for a world where there would be no need for performance improvement plans. In truth a PIP just exists to justify letting someone go. It is better to be honest about the work and needs of the employees from the start. Also the steps above would keep interviewing from being a sin. It allows people to be honest and for people to do their best work.
If an employee is unhappy you should tell them to go interview. It will make it easier for them sometimes to articulate to a complete stranger regarding what they want. Also it will let them know what market value is for their job. And they may realize that the grass isn't greener and they'll come back more motivated. The catch is that if someone comes back you need to say "what did they offer you" and you need to offer it because that's what "market comp" means.
Good interview question: tell me about the day you went home and said to yourself "damn I'm good"
Netflix didn't have performance reviews. They had gantt charts and knew who was performing and who was lagging behind.
Pattern recognition - how to lean people. Pay attention to patterns. Learn to observe. People are rewarded differently, wired differently, and need to be treated differently.
Learn to observe how people react to: change, challenge, the unknown, each other, unexpected interruption, learning
What Patty did to innovate within HR is to just stop doing things. No more PIP. No more vacation policy.
Humans are organized for efficiency. You cannot avoid organization, it will form.
Leading from First Principles - Scott Chacon - GitHub
Five years ago the four founders of GitHub were open source developers without business credentials. They worked from their homes. They ran the company on business minimalism which means they ran until they had problems and then fixed them. They ran an open source business which means:
- no meetings
- no offices
- no work hours (work when you feel productive)
- everything is asynchronous
- people self-assign work
- maintainers, not managers
- everybody contributes ideas (maybe only a couple provide vision)
The entire company works this way (even the legal department). They call these ideas their first principles (borrowing the concept from philosophy). There's no dogma that says "no managers" -- there's the argument of "is a manager the best way to get the value we want".
The 3 year old prosecution = "why" - your ideas should survive this. Direct the three year old prosecution at your business. Same idea as the Toyota "five whys" concept.
Examples of 3 year old prosecution
OFFICE - is it the same as the factory 100 years ago. Are you doing something because it has been done in the past or because it works for you?
First principles of the office:
- a place to work (maybe you have kids and you need a space you can get away to)
- creative collaboration
- serendipitous interactions
- meet the suits (auditor meetings etc)
- interviews and onboard people
- physically ship things to and from
- event space (for public or internal)
Note that at the office GitHubbers don't talk to each other because it's more fair to those not at the office and are on chat.
THE SCHEDULE - most people work about the same hours - 9 to 5 is "core hours" for meetings - why?
How often do you need to meet in person? Only for one of these things: strategy and brainstorming and serendipitous interactions OR heads down, uninterrupted implementation
Think about the university environment: you go to a couple classes a day and then you make up your study time on your terms. You decide who you work with and when you work on what. On the other hand kindergarten is someone watching you all day. Which does your office represent?
- set your own daily schedule
- mini-summits a few times a year (hack-house) to come up with a vision - it's up to the team to decide based on need for vision of communication issues
- whole company gets together once a year
- autonomous - there are benefits to allowing your people to work on their own. This also allows you to bring more diverse people into your organization. It's easier since good people will leave other companies to work for yours. For instance GitHubbers don't miss dance recitals.
MEETINGS - what is the output of a meeting? Who generally comes out ahead in these meetings? Does it support the best ideas or certain personalities?
- chat rooms
- discussion lists (@ mention people)
- as long as it is logged someone else can find out about it on their time. you never leave people out
MANAGEMENT - what does management do? Managers solve a lot of problems, but they are not the type of problems that GitHub has right now.
Why would you have managers?
Assign tasks, coordination, facilitating communication - at GitHub they assign for themselves (everyone is a mini-manager)
Hiring and firing - informed by team, but the four founders make the decisions
Venn Diagram: your skills and interests - problems GitHub has = overlap: what to work on
This is what they tell new employees to work on
By using software to allow cross company communication they have not put personal bias or selective memory in the mix.
Pay structures - why are you doing bonuses? is it maslovian based or reward based?
Long hours - do you implicitly or explicitly look for this and is it important for the business? Is it a badge of honor or a badge of foolishness
Benefits and perks - they've decided against free food because they want to be good community citizens and have people dine at local establishments
Retention counteroffers - in the long term is this in the interest of the company to have people there only for money?
With everything go to first principles and see if the thing is supported.
For their IP assignment legal document they passed it around and let employees fork it and give input which resulted in a better legal document in the end.
Ask these three questions on everything:
What are our values?
What are we trying to accomplish?
What is the best way to get there?
Are you willing to put kittens on your factory floor?
How to Create a Culture of Shipping Product Continuously - Hiten Shah - KISSmetrics
Founder Bombs - when someone high up gives ideas to others and they have no idea how to deal with this thing that came out of the blue and seems to be a request, but it isn't clear. One person realized that a founder bomb only matters if in a seven day period it is mentioned three times
The solution to founder bombs has been for founders to be more mindful and also to find someone who can handle that kind of input. The challenge is to effectively balance all the forces.
What are you working on? Why are you working on it? - ask these questions to a lot of people and you'll learn a lot about the company culture. What's most important is to watch how they articulate what should often be the same message for everyone. It's important that people understand why the company exists and what they are doing and how it fits in.
Closed door decisions - you shouldn't hear "they walked out of the room with a decision about what we are going to do". You should be sure everyone on the team understands the goal.
"Build products people want" - Paul Graham
Shipping continuously is not enough. You have to ship the right things. Great products are created by people for people. Don't ship shit. Make the right product.
"Hire the most amazing people you can. Communicate goals. Turn them loose. Profit." - Sam Schillace, Box
Amazon's Two-pizza teams (2PT) - 6 to 10 people - this is the maximum size of a team. This reduces communication barriers and group think. Kahn likes the one pizza rule of 3 to 5 people.
At Apple they have cauldrons, 3 to 5 people. During a 2 to 3 hour meeting you put all of the ideas into the cauldron. The idea of it is that the best ideas can bubble up from the cauldron without people thinking about who had the idea since all of the ideas go in blank into the cauldron. Don't worry about who's idea it is, value the idea on its own merit.
"The ultimate goal of building a company is to have the right product thesis at the right time." - Josh Elman
A product thesis is problems, use cases, and examples
Working backwards - "We try to work backwards from the customer rather than starting with an idea for a product and trying to bolt customers into it." - Ian McAllister Amazon
Amazon will create a fake press release first. KISSmetrics will write an internal blog post in the same sort of way. It forces you to really focus on what you want to build and why. It gets everyone on the same page about the project.
Cust Dev + Research + Metrics = Thesis -> solution team review (cauldron) -> Q&A on thesis -> plan solution -> defend the solution (iterate where necessary)
Golden motion - pick one metric or goal and build around that - happens with the solution team review
Seek feedback from your team. Have a postmortem.
Send a survey to employees - ask who hit it out of the ballpark last month, what feedback do you have. They are anonymous and optional. Find ways to get feedback.
KISSmetrics recently hired a happiness manager - half of her job is like an office admin/ executive assistant - the other half was helping the whole company be happy. She handles the monthly anonymous surveys, when is someone's birthday, what is someone's favorite beer. Someone focused internally and on the whole team.
The solution to founder bombs has been for founders to be more mindful and also to find someone who can handle that kind of input. The challenge is to effectively balance all the forces.
What are you working on? Why are you working on it? - ask these questions to a lot of people and you'll learn a lot about the company culture. What's most important is to watch how they articulate what should often be the same message for everyone. It's important that people understand why the company exists and what they are doing and how it fits in.
Closed door decisions - you shouldn't hear "they walked out of the room with a decision about what we are going to do". You should be sure everyone on the team understands the goal.
"Build products people want" - Paul Graham
Shipping continuously is not enough. You have to ship the right things. Great products are created by people for people. Don't ship shit. Make the right product.
"Hire the most amazing people you can. Communicate goals. Turn them loose. Profit." - Sam Schillace, Box
Amazon's Two-pizza teams (2PT) - 6 to 10 people - this is the maximum size of a team. This reduces communication barriers and group think. Kahn likes the one pizza rule of 3 to 5 people.
At Apple they have cauldrons, 3 to 5 people. During a 2 to 3 hour meeting you put all of the ideas into the cauldron. The idea of it is that the best ideas can bubble up from the cauldron without people thinking about who had the idea since all of the ideas go in blank into the cauldron. Don't worry about who's idea it is, value the idea on its own merit.
"The ultimate goal of building a company is to have the right product thesis at the right time." - Josh Elman
A product thesis is problems, use cases, and examples
Working backwards - "We try to work backwards from the customer rather than starting with an idea for a product and trying to bolt customers into it." - Ian McAllister Amazon
Amazon will create a fake press release first. KISSmetrics will write an internal blog post in the same sort of way. It forces you to really focus on what you want to build and why. It gets everyone on the same page about the project.
Cust Dev + Research + Metrics = Thesis -> solution team review (cauldron) -> Q&A on thesis -> plan solution -> defend the solution (iterate where necessary)
Golden motion - pick one metric or goal and build around that - happens with the solution team review
Seek feedback from your team. Have a postmortem.
Send a survey to employees - ask who hit it out of the ballpark last month, what feedback do you have. They are anonymous and optional. Find ways to get feedback.
KISSmetrics recently hired a happiness manager - half of her job is like an office admin/ executive assistant - the other half was helping the whole company be happy. She handles the monthly anonymous surveys, when is someone's birthday, what is someone's favorite beer. Someone focused internally and on the whole team.
Nothing to Hide: Living with Complete Email Transparency - Patrick Collison - Stripe
Had unconventional ideas: no office, pay everyone the same, not having a phone number, not having a receptionist
Later realized an office is needed if you have a few employees. Also that compensation has to vary.
Believer in ambient transparency, where you can overhear things in the office and join in if it is important to you.
At first it was just a couple brothers and they shared an email address so they'd be on the same page. With a third employee they configured gmail so all 3 could log into each other's inboxes. Later they configured the server to automatically BCC everyone on every outgoing message. This scaled to around 6 people, but of course they needed a new solution to scale further. For each functional area there was an email address. So for instance there was "ops" and "finance" which you could CC on any email if it was relevant. A new employee configures which of many accounts they will subscribe to (there's a script for it). It gives an ambient feel of what's going on in the organization. This means they aren't surprised by things that are happening. Employees are able to discover a lot just by searching through the stream. Also people get to learn about things outside their area, like for instance "legal" is a popular list for many who have nothing to do with it. A benefit is that this system seems to reduce politics. Some of politics is controlling the flow of information, but no one can do that. Some issues are that people get confused by things or a little surprised, but they seem to get used to that. Also it can be used as a crutch for implicit agreement by merely CCing and not getting authorization. People can loose too much time in email if they don't have good time management.
Overall there is a hive mind mentality where everyone can be involved. Other tools that can help for this are HackPad. It allows for transparency. By having this overall transparency meetings can be more efficient. Especially it calms fears about the board meeting. At Stripe a random employee is always invited to the board meeting to take notes and pass them out to everyone.
Lessons learned
1. get an office
2. it is net better to not have project manager. better to have engineers to manage. there is a faster iteration cycle when this happens. you get twice the amount of iterations
3. self discipline matters most in hiring people
4. hiring individual contributors who can be highly useful (have people do real life activities like find bugs or refactor code instead of text book problems)
5. a change of environment can have an amazing impact. hack trips to foreign countries are extremely impactful
6. happy culture is built from hiring happy people
7. don't underestimate how much people will want to learn about other areas of the company. bring them to important meetings, be transparent
8. have as much discussion about your code as possible - ways of doing things - idioms
9. have a build and run distinction in your team. every week have 25% of the team handle run while the other 75% handle build
10. make sure that group chat happens in your company. its a communication tool and a culture tool
11. have a shipped@ email address for people to send to when you have shipped something. it feels good to send to that email address and allows people to see what has happened. do pay attention though to the results of shipping (it's not a good thing to ship a bad product)
12. have a social@ email address to allow people to easily set up group outings (like, I'm going out for a drink, who wants to come along)
13. at around 25 people have a company chef it actually makes financial since versus eating out and people love it
14. visit your users on a frequent basis and get the rest of the company to do it too
15. offices are often not good for working late into the evening. the lights are often too florescent for evening hours. get ikea floor lamps to give a warmer feel at night. it helps people do their best late at night. (this lesson has been proven each time Stripe moves into a new office as they immediately see few people staying late until they correct the lighting)
16. tell everyone why you've fired someone. it helps reassure others about what happened and why.
17. thank people profusely if they give a criticism about the company (it's hard to speak up, so reward it)
18. constantly find ways to praise those who do the undervalued work.
19. preserve values more than culture.
20. don't talk about someone who's not present and incourage others to do the same
get involved: hack pad at bit.do/cultivate - https://hackpad.com/Culture-Hacks-8WwaHfFfZfl
Things that would be helpful - group chat lines (searchable logs, all OS, inline images) they use IRC - screen sharing collaboration for code (both terminal and screen)
Later realized an office is needed if you have a few employees. Also that compensation has to vary.
Believer in ambient transparency, where you can overhear things in the office and join in if it is important to you.
At first it was just a couple brothers and they shared an email address so they'd be on the same page. With a third employee they configured gmail so all 3 could log into each other's inboxes. Later they configured the server to automatically BCC everyone on every outgoing message. This scaled to around 6 people, but of course they needed a new solution to scale further. For each functional area there was an email address. So for instance there was "ops" and "finance" which you could CC on any email if it was relevant. A new employee configures which of many accounts they will subscribe to (there's a script for it). It gives an ambient feel of what's going on in the organization. This means they aren't surprised by things that are happening. Employees are able to discover a lot just by searching through the stream. Also people get to learn about things outside their area, like for instance "legal" is a popular list for many who have nothing to do with it. A benefit is that this system seems to reduce politics. Some of politics is controlling the flow of information, but no one can do that. Some issues are that people get confused by things or a little surprised, but they seem to get used to that. Also it can be used as a crutch for implicit agreement by merely CCing and not getting authorization. People can loose too much time in email if they don't have good time management.
Overall there is a hive mind mentality where everyone can be involved. Other tools that can help for this are HackPad. It allows for transparency. By having this overall transparency meetings can be more efficient. Especially it calms fears about the board meeting. At Stripe a random employee is always invited to the board meeting to take notes and pass them out to everyone.
Lessons learned
1. get an office
2. it is net better to not have project manager. better to have engineers to manage. there is a faster iteration cycle when this happens. you get twice the amount of iterations
3. self discipline matters most in hiring people
4. hiring individual contributors who can be highly useful (have people do real life activities like find bugs or refactor code instead of text book problems)
5. a change of environment can have an amazing impact. hack trips to foreign countries are extremely impactful
6. happy culture is built from hiring happy people
7. don't underestimate how much people will want to learn about other areas of the company. bring them to important meetings, be transparent
8. have as much discussion about your code as possible - ways of doing things - idioms
9. have a build and run distinction in your team. every week have 25% of the team handle run while the other 75% handle build
10. make sure that group chat happens in your company. its a communication tool and a culture tool
11. have a shipped@ email address for people to send to when you have shipped something. it feels good to send to that email address and allows people to see what has happened. do pay attention though to the results of shipping (it's not a good thing to ship a bad product)
12. have a social@ email address to allow people to easily set up group outings (like, I'm going out for a drink, who wants to come along)
13. at around 25 people have a company chef it actually makes financial since versus eating out and people love it
14. visit your users on a frequent basis and get the rest of the company to do it too
15. offices are often not good for working late into the evening. the lights are often too florescent for evening hours. get ikea floor lamps to give a warmer feel at night. it helps people do their best late at night. (this lesson has been proven each time Stripe moves into a new office as they immediately see few people staying late until they correct the lighting)
16. tell everyone why you've fired someone. it helps reassure others about what happened and why.
17. thank people profusely if they give a criticism about the company (it's hard to speak up, so reward it)
18. constantly find ways to praise those who do the undervalued work.
19. preserve values more than culture.
20. don't talk about someone who's not present and incourage others to do the same
get involved: hack pad at bit.do/cultivate - https://hackpad.com/Culture-Hacks-8WwaHfFfZfl
Things that would be helpful - group chat lines (searchable logs, all OS, inline images) they use IRC - screen sharing collaboration for code (both terminal and screen)
What *Do* You Do All Day? - Kate Matsudaira - popforms
praise isn't always proportional to effort
how do you measure performance? hours (this isn't great), lines of code (supports inefficiency), bugs & tests (doesn't work), features (how do you measure the effort though)
the better you are at your job, the less people know about it because you don't ask for help or lean on others -- you gain autonomy
in teams without managers each member is actually expected to be a manager -- someone still has to lead
we need leadership, not likership
Abraham Lincoln lost almost all elections and failed at what he did until he was 51 and became president
leadership is a decision not a position
positive use of power is good leadership
sources of power: formal & informal
informal power is charisma, expertise, or relationship based
charisma - you have it or you don't. luckily it's not required
expertise - educate yourself and you'll gain it
relationship - build trust (it is essential)
what is trust? your relationships with everyone are important (it is a weighted graph)
elements of trust: relationship architecture, reputation, contribution
contribution - what you do, setting an example, practicing what you preach (never ask something of someone you wouldn't do yourself)
have integrity - even when people aren't watching
ask for engineering integrity - how does the system fare a few years after go live
know your timing - analyze (incorporate) -> estimate -> (confidence) -> actual -> (validate) -> analyze
being on time shows you think other people's time is as important as your time
... and if you show up on time you've predicted the future accurately
think about your coworkers - who is the person you admire the most? what qualities? Now, think about yourself, do you have those qualities?
make sure others feel that you feel they are important - when you walk around imagine everyone is wearing a sign that says "i want to feel important today" - how can you help them with that
celebrate wins - let people know when they've done a job well - help people celebrate
be open to ideas that sound wrong. ideas are fragile. take an interest in other's ideas. take a moment before responding... then explore the idea
empathy and attitude - when you come to work you should have sunshine not rain clouds. be cautious about commiserating with people. listen with empathy, but try to have good attitude and not take sides, but instead reframe (are we sitting on a boat in the rain, or are we on a great adventure?)
relationship architecture - it's the trust graph of your relationships with others. Make 2 ordered lists: 1. the most important people on your team (those that get things done) 2. the people you have the best relationships with
there should be a lot of overlap in these lists. you need to be more intentional with your relationships
you are the average of your five best friends
start with the leaders. don't' focus on title, look for the people that everyone follows. the people who stand up and everyone follows.
relationship troubleshooting
- relationships are like filmstrips
- apologize swiftly and sincerely (don't go to boss or HR)
- don't deflect when there is confrontation
archetypes - the jerk - dealing with difficult people - pause - a lot of times people are difficult because they need recognition or to feel important
- sharkasm - the passive aggressive - seek resolution with someone's interests not their position - people hear what they want to hear
- the outspoken one - has opinion on anything - sometimes experts - repeat back what you hear to them - list options and vote (put what people say on the whiteboard so they know they are heard)
- the strong silent type - using post it notes to vote helps - talk to introverts one on one after meetings and pay attention to them, focus - ask questions wait for answers
- the complainer - whine and act defeated - stay positive but realistic - just state the facts and how they'll be addressed
Focus on the long term - be deliberate with who you have relationships with. difficult people change if you change. with your support and interest people can change.
where does success come from? it's not projects or work, it comes from people
how do you measure performance? hours (this isn't great), lines of code (supports inefficiency), bugs & tests (doesn't work), features (how do you measure the effort though)
the better you are at your job, the less people know about it because you don't ask for help or lean on others -- you gain autonomy
in teams without managers each member is actually expected to be a manager -- someone still has to lead
we need leadership, not likership
Abraham Lincoln lost almost all elections and failed at what he did until he was 51 and became president
leadership is a decision not a position
positive use of power is good leadership
sources of power: formal & informal
informal power is charisma, expertise, or relationship based
charisma - you have it or you don't. luckily it's not required
expertise - educate yourself and you'll gain it
relationship - build trust (it is essential)
what is trust? your relationships with everyone are important (it is a weighted graph)
elements of trust: relationship architecture, reputation, contribution
contribution - what you do, setting an example, practicing what you preach (never ask something of someone you wouldn't do yourself)
have integrity - even when people aren't watching
ask for engineering integrity - how does the system fare a few years after go live
know your timing - analyze (incorporate) -> estimate -> (confidence) -> actual -> (validate) -> analyze
being on time shows you think other people's time is as important as your time
... and if you show up on time you've predicted the future accurately
think about your coworkers - who is the person you admire the most? what qualities? Now, think about yourself, do you have those qualities?
make sure others feel that you feel they are important - when you walk around imagine everyone is wearing a sign that says "i want to feel important today" - how can you help them with that
celebrate wins - let people know when they've done a job well - help people celebrate
be open to ideas that sound wrong. ideas are fragile. take an interest in other's ideas. take a moment before responding... then explore the idea
empathy and attitude - when you come to work you should have sunshine not rain clouds. be cautious about commiserating with people. listen with empathy, but try to have good attitude and not take sides, but instead reframe (are we sitting on a boat in the rain, or are we on a great adventure?)
relationship architecture - it's the trust graph of your relationships with others. Make 2 ordered lists: 1. the most important people on your team (those that get things done) 2. the people you have the best relationships with
there should be a lot of overlap in these lists. you need to be more intentional with your relationships
you are the average of your five best friends
start with the leaders. don't' focus on title, look for the people that everyone follows. the people who stand up and everyone follows.
relationship troubleshooting
- relationships are like filmstrips
- apologize swiftly and sincerely (don't go to boss or HR)
- don't deflect when there is confrontation
archetypes - the jerk - dealing with difficult people - pause - a lot of times people are difficult because they need recognition or to feel important
- sharkasm - the passive aggressive - seek resolution with someone's interests not their position - people hear what they want to hear
- the outspoken one - has opinion on anything - sometimes experts - repeat back what you hear to them - list options and vote (put what people say on the whiteboard so they know they are heard)
- the strong silent type - using post it notes to vote helps - talk to introverts one on one after meetings and pay attention to them, focus - ask questions wait for answers
- the complainer - whine and act defeated - stay positive but realistic - just state the facts and how they'll be addressed
Focus on the long term - be deliberate with who you have relationships with. difficult people change if you change. with your support and interest people can change.
where does success come from? it's not projects or work, it comes from people
Cracking the Culture Code - Elaine Wherry - Meebo
 |
| Your reason for everything (even free food) should tie into your vision |
The Meebo policy is to pay for lunch if you have it with two or three employees since that fosters community.
Meebo was hyper communicative, but people were feeling out of the loop. The issue was that the mission and vision weren't clear so people didn't know how things fit together and had no larger perspective.
Culture = Artifacts + systems and values + truth
For our companies truth is defined by the market
This has problems since it changes and is short term
It is also unfair
Salaries are based on what the market pays not what you deserve
 |
| There are some uncomfortable truths you need to face and be both honest and open about. |
Besides the market we rely on vision and data for truth
Look back to shape your vision -- find what people said about you in the past
Making digital life surprisingly simple = Meebo's vision
Moving towards truth is via trust. It is either given or earnedCulture tends to go from given trust to earned trust to a mix as the company gets on its feet.
Why does Zappos offer 4k to new hires to quit?
Badgeville gives stickers that people put on their laptops when they do something well. It is an example of trust earned.
Staff should be exceptional people who give 100% and deal well with change.
Culture can be groomed
Hiring is really really hard to change
Culture is top down
Leadership is same values, different actions
Adaptability is huge
Culture is not a cult -- it's a way of doing business
We are not reinventing the wheel
How I failed - Tim O'Reilly - O'Reilly
Failure: Not making sure that my team was hearing what I was actually saying.
You don't know what book you wrote until you know what people are reading
[TK:Pyramid pic]
"You can leave anything out as long as you know what it is" - Hemingway
Good fiction has way more depth than what you are given. The author knows more than they consciously reveal.
[tk Decision filter pic]
O'Reilly learned late of the sponsorship model for events
O'Reilly has strength in doing good ideas just to do them, but sometimes goes wrong because they don't know how it fits into their vision
Failure: listening to "that's how it's done"
Insight and fresh thinking should be applied both inside and outside. Just as much as you try to understand the market and customers you need to understand your employees.
Failure: lack of financial and operational discipline
Cash flow positive = putting money in the bank. O'Reilly was profitable but not cash flow positive for too long
"Happiness is positive cash flow" - Fred Adler
Amazon says, we don't spend money on anything that doesn't help customers
Treat the financial team as co-founders
Hold teams accountable to their numbers (financial or otherwise)
Tim thinks he failed when he let people off the hook on their numbers
Run lean, reinvent tirelessly
Run lean and when you pivot reassess your staff
Work hard on your internal culture
Drive by Dan Pink - autonomy, mastery, & purpose = what employees need to be motivated
Failure: tolerating mediocrity
Sometimes you need to cancel a project at the last minute because it isn't good enough. It might be disruptive but you can try to have processes that help mitigate it
"Keep stretching the bow, you repent of the pull" Lao Tzu
If you keep pulling back on the bowstring your arm will start to shake and you will progressively ruin your aim
Stand up for greatness
Failure: hiring supplements not complements - hire people that are not like you and bring more and bring diversity to the table
As a leader you should not be the best person in the company at any one thing. Applies to both strengths and weaknesses
Failure: I'll take care of that - compensate for the employees weakness is not good
A leader can be a handyman that just deals with what's lying around rather than an engineer.
The organization you build is more important then the product you build in the long term.
O'Reilly has the vision of being an education company. That's why it can hold events like Cultivate and not stray from the vision. They aren't just a book company.
You don't know what book you wrote until you know what people are reading
[TK:Pyramid pic]
"You can leave anything out as long as you know what it is" - Hemingway
Good fiction has way more depth than what you are given. The author knows more than they consciously reveal.
[tk Decision filter pic]
O'Reilly learned late of the sponsorship model for events
O'Reilly has strength in doing good ideas just to do them, but sometimes goes wrong because they don't know how it fits into their vision
Failure: listening to "that's how it's done"
Insight and fresh thinking should be applied both inside and outside. Just as much as you try to understand the market and customers you need to understand your employees.
Failure: lack of financial and operational discipline
Cash flow positive = putting money in the bank. O'Reilly was profitable but not cash flow positive for too long
"Happiness is positive cash flow" - Fred Adler
Amazon says, we don't spend money on anything that doesn't help customers
Treat the financial team as co-founders
Hold teams accountable to their numbers (financial or otherwise)
Tim thinks he failed when he let people off the hook on their numbers
Run lean, reinvent tirelessly
Run lean and when you pivot reassess your staff
Work hard on your internal culture
Drive by Dan Pink - autonomy, mastery, & purpose = what employees need to be motivated
Failure: tolerating mediocrity
Sometimes you need to cancel a project at the last minute because it isn't good enough. It might be disruptive but you can try to have processes that help mitigate it
"Keep stretching the bow, you repent of the pull" Lao Tzu
If you keep pulling back on the bowstring your arm will start to shake and you will progressively ruin your aim
Stand up for greatness
Failure: hiring supplements not complements - hire people that are not like you and bring more and bring diversity to the table
As a leader you should not be the best person in the company at any one thing. Applies to both strengths and weaknesses
Failure: I'll take care of that - compensate for the employees weakness is not good
A leader can be a handyman that just deals with what's lying around rather than an engineer.
The organization you build is more important then the product you build in the long term.
O'Reilly has the vision of being an education company. That's why it can hold events like Cultivate and not stray from the vision. They aren't just a book company.
Subscribe to:
Posts (Atom)