How Google Works
“How Google Works” by Eric Schmidt and Jonathan Rosenburg
My Notes
- things a company should have
- you need a mission, mission ideas for CS
- products that make things “easy to sell, easy to buy”
- e-commerce, made easy
- buyer-centric vs. seller-centric vs. both?
- meetings buyers and sellers were they are
- increasing seller reach
- you need a workspace conducive for smart creatives
- open work spaces
- but it should also have places to retreat
- also some that are nice and quiet for focused work
- you need values
- listing comment sold cultural values
- no opinion is better than another because of tenure
- the best ideas win
- nothing is sacred
- the project you worked on all weekend may not be the one we need to use
- the project that made us 1st million might not be the one that will help us 10x
- listing comment sold cultural values
- you need a mission, mission ideas for CS
- finding direction, making decisions
- all product decisions should be based on a technical insights, not unfounded opinions
- open platforms grow faster, consider them for certain problems and areas — especially when challenging big incumbent companies
- don’t obsess over competitors
- know who they are and be “proud” of them, but don’t copy them — you’ll just me incrementally growing and not truly innovate
- hiring
- should be peer based
- set standards really high, find people’s passion in the interview, hire learners (growth mindsets and grit!)
- you want smart generalists not specialists!
- ask yourself, would you want to be stuck in an airport with this person for 6 hours? Hire interesting people! But not necessarily people you like
- you don’t have to hire someone with Java experience for a Java job, hire the best engineer you can and if they know Java that’s just a bonus
- hire someone who you know has high potential career trajectory
- during interviews ask questions that will show off candidate’s thinking, not their resume
- “How did you pay for college?” “If I was to look at your browser history what would I learn that isn’t on your resume?”
- don’t hire people that give generic answers
- good interviews should last about 30 minutes
- final hiring decisions should be left to committee
- don’t hire people who prefer to work alone
- don’t hire people who just want a job
- teams/careers
- to keep smart creatives, keep their jobs interesting by all means possible
- ex: have them shadow CEO for 6 months
- ex: set them on useful tasks (even if it is outside their comfort zone), don’t let them settle
- team building
- “Make managers trade away their MnMs, let them keep the raisins”
- people leaving a company usually isn’t b/c compensation, so listen carefully to what they are frustrated with
- career growth is like surfing
- you want to surf in the right industry/spot and try to catch the best wave/company and ride it as long as you can
- everyone should get good at statistics/data b/c it is going to be just as big as coding in the next decade
- everyone should think like a CEO and read!
- imagine you run into your CEO in the hallway and they ask what your working on. What do you say? Is it a solid pitch?
- to keep smart creatives, keep their jobs interesting by all means possible
- meetings
- all meeting rooms at google has two TVs, one for teleconference and one for data
- have a bias for data in making decisions
- also a bias for action when data/debate no longer yields any new learning
- meeting regularly for decisions that are fundamental to the company can be a good thing
- data and a speech don’t convince people, but data a speech and appealing to emotions do — “the Oprah rule”
- all meetings should have an organizer & decision maker
- organize/decision maker sends the invites, has the agenda 24hrs in advance, and sends summary/decisions made email afterward
- all meetings can be cancelled at any time for not having organizer or agenda
- all meeting rooms at google has two TVs, one for teleconference and one for data
- 80/20 rule: spend 80% of your time on 80% of your revenue
- data
- company should default to open; share as much information as efficiently as possible to everyone
- all OKRs, even personal ones, should be shared
- only things that have legal ramifications shouldn’t be shared
- presentations should have less slides and focus on quality charts/data
- company should default to open; share as much information as efficiently as possible to everyone
- company communication
- leadership needs to “know the details”, don’t be hands-off
- Eric S. keeps a list of questions for people on Apple Notes so he can bring them up when he sees those people
- having a system like Canny for internal company questions to ask upper management can be a good way to help those who have trouble asking questions in big open scenarios (as long as all questions are asked in good will, this can become toxic too)
- real conversation trumps chat
- write down your own user manual for yourself (interesting idea 💡)
- company “snippets” — a place where everyone writes down weekly short notes about what got done, short reflection, etc (like a big public diary for the company)
- over communicate the important big ideas or values and do it honestly, with varied media, and to the people you want to hear it
- don’t just “send all”, not give a reason you think it is relevant, and not direct the conversation to something productive
- “trip reports” are an effective meeting tool to get everyone talking “where did you go, what did you learn”
- if you are a manager, ask yourself: Would you work for you?
- answer emails quickly, keep your inbox clean, don’t waste time, don’t YELL, be concise
- continually be looking ahead, what can we be doing better, should we be looking at other products/solutions, are we the best in our field?
- business partnership comm. should be based on diplomacy, it is better when we work together even if our ideologies might differ
- when speaking with the press have insights not canned responses
- communication idea: don’t work on anything for a week and just talk to people across the company to learn about problems and processes
- leadership needs to “know the details”, don’t be hands-off
- innovation
- Innovation - radically new things; but also when 500 incremental changes combine to make something great
- innovative culture should support the crazy person trying something new and the first adopters who add fuel to the fire
- focus on the user, and the revenue will follow (user-first)
- 10x
- you aren’t thinking big enough, w/e you’re thinking about 10x it!
- doing 10% better requires tweaks, doing 10x better requires rethink
- ask how you could 10x a metric, it will require radical change!
- all green OKRs are bad, you didn’t set your targets high enough
- 70/20/10 resource allocation core/successful-ideas/crazy-ideas
- constraints force creativity, a lack of resources forces ingenuity
- don’t over-budget innovation, you can MVP something for real cheap!
- 20% time
- 20% time is for individuals time, not to be confused with 70/20/10 resource allocation
- 20% time really isn’t about time, it’s about freedom, it’s freedom to work on something even though you weren’t asked to — it’s just because you’re motivated about it
- You don’t get compensation bonuses for 20% projects, compensation will dilute innovation
- “Demo days” are a way that google encourages people to build prototypes; most won’t go on to be a product, but everyone will learn something, collaboration, and be able to bring that back to their work
- Ship and Iterate quickly and use data to feed the winners and starve the losers
- Don’t ship crappy products tho, ship MVPs that are quality and could get to the wow moment within a few iterations
- the future…
- how will you go to smart creatives or get them to come to you?
- What are your competitive advantages? What would you do if you lost them? Would your customers stay?
- How will emerging technology effect your business?