Work Rules! (Notes)

Work Rules! Insights from Inside Google that will Transform How you Live and Lead

My notes

Author: Laszlo Bock
Recommended by: Doug Meiser

Overview

Laszlo Bock is the VP of People Operations at Google. In this book he shares his experiences of working at Google and what it takes to create a workplace that is frequently voted as the best company to work for. His main advice is to hire smart people, give them a common vision/goal, and then get out of the way and let them work. If something is an obstacle to them getting their work done or achieving the agreed upon result, it is a manager's job to break down that obstacle. The goal of People Operations is to create a workplace where people are encouraged to speak freely, to point out obstacles and remove them, to try new things, and to make data-driven decisions about what works and what doesn't work. You'll hear Laszlo talk about data-driven experiments a lot. Google is running hundreds of experiments a year on its own employees. Little things like changing the placement of healthy food in the breakrooms to incentivize healthier eating or changing the size of plates in the lunchroom (research showed people eat less and feel more full with smaller plates). Also big things like seeing what happens if we eliminate all managers (they quickly brought them back after admitting that didn't work out so well). The book is full of real-world, practical advice. This page will capture notes on each chapter and some of the take-aways I had from it.

Chapter 1: Become a Founder

Choose to think of yourself like a founder

Now act like one

Encourage people to take control of their workplace to achieve their desired results. Don't micromanage. Give them autonomy. Make sure that their work life is something they can take pride in and know that they are contributing to something valuable. Laszlo quotes Henry Ford on page 23:

The kind of workman who gives the business the best that is in him is the best kind of workman a business can have. And he cannot be expected to do this indefinitely without proper recognition... If a man feels that his days' work is not only supplying his basic need, but is also giving him a margin of comfort, and enabling him to give his boys and girls their opportunity and his wife some pleasure in life, then his job looks good to him and he is free to give it of his best. This is a good thing for him and a good thing for the business. The man who does not get a certain satisfaction out of his day's work is losing the best part of his pay.

He also gives the example of a Mervin J. Kelly at Bell Labs who built "Franken-teams, combining 'thinkers and doers' as well as disparate experts on single teams." One such team created the transistor. This team contained "physicists, metallurgists and electrical engineers; side by side were specialists in theory, experimentation, and manufacturing."

Mr. Kelly believed that freedom was crucial, especially in research. Some of his scientists had so much autonomy that he was mostly unaware of their progress until years after he authorized their work. When he set up the team of researchers to work on what became the transistor, for instance, more than two years passed before the invention occurred. Afterward, when he set up another team to handle the invention's mass manufacture, he dropped the assignment into the lap of an engineer and instructed him to come up with a plan. He told the engineer he was going to Europe in the meantime.

All great teams start with a founder mentality.

Chapter 2: Culture Eats Strategy for Breakfast

Start by giving people "a mission that matters." Don't surround your mission statement with too many words or talk about shareholders and profits. Give people something simple and purposely open-ended. Google's mission is "to organize the world's information and make it universally accessible and useful." How many thousands of projects and ideas could be launched toward this goal? Make the mission a moral statement rather than a business goal. Make it big and open-ended. Something that can never be completely finished. Let team members be creative and innovative in how they pursue that mission.

Transparency is the second cornerstone of Google's culture.

If you believe that people are good, you must be unafraid to share information with them

"Default to open" is their belief. "Assume that all information can be shared with the team, instead of assuming that no information can be shared. Restricting information should be a conscious effort, and you'd better have a good reason for doing so." Google's code base is almost completely open to all engineers on day 1 instead of being locked down. The Google intranet includes product roadmaps, launch plans, status reports, and both employee and team goals ("OKRs" or "Objectives and Key Results") so that everyone can see what everyone else is working on. How can people act like founders and find where they can best help out if they don't know what is going on?

There are about 1 or 2 major incidents per year when someone violates this trust and releases information outside of Google. People are immediately fired for such violations and everyone in Google is told about these incidents. In one case, an employee leaked product specifications about a new phone being built. Instead of tightening down on information, the VPs showed the specs to everyone at an all-hands meeting. Employees knew they were trusted, and they also knew that they would be immediately dismissed if they violated that trust.

Give people slightly more trust, freedom and authority than your are comfortable giving them. If you're not nervous, then you haven't given them enough.

Culture also means the most when it is put to the test. Google pulled out of China when their search results were being censored. They put up a page notifying people of the censorship and explaining why they couldn't continue to operate there. They sacrificed profits for core values. That's when you know your values and your culture really matter. Google employees constantly worry about losing their culture as they grow bigger and bigger. One of the big goals of People Operations and the cultural teams running throughout the company is to intentionally focus on culture to make sure it is not lost.

Chapter 3: Lake Wobegon, where all the new hires are above average

Hiring is the Single Most Important Activity in an Organization

Most companies hire average people and then try to train them to be above-average. Instead, you should focus on hiring the very best. Hire the 90th percentile who can begin doing great things right away without extensive training. It's going to cost you more, but the average company spends more on training than it does on hiring. With less training needs (or training covered by internal talent - as discussed later) you can put more money into hiring. The goal is to be more like the Yankees: pay well, hire the best, and win more World Series than anyone else. (Google didn't always have the money to pay people top salaries - in which case you'll have to draw people in with your mission and your culture.) You'll have to wait longer and work harder to find the top performers in your industry, but don't settle for less. As one SVP said: "A top-notch engineer is worth 300 times or more than an average engineer... I'd rather lose an entire incoming class of engineering graduates than one exceptional technologist."

"Only hire people who are better than you."

Find people who are better than you in at least one area. Otherwise, why are they there? You want to surround yourself with people that have skills you don't have. Then trust them to use those skills. As Steve Jobs famously said (I don't think this was in the book):

"It doesn't make sense to hire smart people and then tell them what to do. We hire smart people so they can tell us what to do."

Google also doesn't let managers make hiring decisions for their own teams. Managers are biased and will be the first to compromise quality if they really need to fill a crucial position quickly. Managers can make recommendations about who to hire, but at Google a separate hiring board will interview candidates and make sure they are a good fit for the company and the position. This board then makes a recommendation that goes all the way up to senior VPs. At one time, the CEO approved every single person who was hired at Google. Imagine how good it feels on your first day knowing that senior leadership has personally approved and committed to you and believes you can succeed. New managers tend to really hate that they can't hire people on their word alone, but they do admit that the quality of teams improves with this process.

Chapter 4: Searching for the Best

Google has done a lot to speed up the rate at which they hire people. At first, it could take months to get someone in the door. Employee feedback identified this as an area to improve - especially in giving people feedback as they progressed through the process. The process still takes time, but they have done several experiments and changes over time to try to speed things up. They noticed that their best employees were usually referrals from current employees. However, the employees eventually used up their social networks and didn't know any other great people to hire. At that point they turned to the internet and websites liked LinkedIn where you can basically get a database of every IT professional in the country (or in your city, if you prefer) and filter through them with algorithms to find the ones that really stand out. The best people are typically happily employed and well-paid by other companies. They aren't even looking for jobs. Proactively recruiting these people is the best way to get them. Posting jobs and hoping that people find you doesn't work. Posting billboards with math puzzles or other unusual advertising was tried at Google but also found to not work. Don't wait for good people to find you (this doesn't work even when you have the brand name and awareness that Google does). Go and find the best people proactively.

Chapter 5: Don't Trust Your Gut

People are incredibly biased in more ways than they know - especially when it comes to interviewing and making judgment calls about people. It is important to use data- and science-based methods to eliminate this bias and get at better decisions. Frank Schmidt and Jon Hunter did extensive research to measure the best way to predict success in a new hire. They measured the R-squared or the percentage of variance explained by each method and ranked the following interview techniques from best to worse:

Sample work tests (29%) - ask someone to solve a real-life problem similar to one they will do on the job
General cognitive tests (26%) - look for intelligent people (but watch out for bias in tests. Even SAT/ACT tests show signs of bias)
Structured interviews (26%) - ask the same questions to all candidates with a consistent rubric for evaluation
Unstructured interviews (14%) - ask whatever questions you want / just have a conversation with a candidate
Reference checks (7%)
Years on the job (3%)
Handwriting analysis (0.04%)

Google focuses on the top 3 in their interviews. They have a tool that will help interviewers select pre-defined, structured questions from a bank of pre-approved questions based on the attributes you are looking for in a new-hire. For those outside Google they note:

If you don't want to build all this yourself, it's easy enough to find online examples of structured interview questions that you can adapt and use in your environments. For example, the US Department of Veterans Affairs has a site with almost a hundred sample questions at http://www.va.gov/pbi/questions.asp. Use them. You'll do better at hiring immediately. (page 97)

They also look specifically for the following 4 qualities in every new hire:

General Cognitive Ability - focus on how people have solved hard, real-life problems
Leadership - specifically "emergent leadership" that ignores formal roles and titles
"Googleyness" - creativity, fun, intellectual humility, conscientiousness, comfort with ambiguity, and evidence that you've taken some courageous or interesting paths in life
Role-Related Knowledge - prior experience/knowledge of the job/problem

Interestingly, the author notes that role-related knowledge is "by far the least important attribute." "Our experience, is that curious people who are open to learning will figure out the right answers in almost all cases, and have a much greater chance of creating truly novel solutions."

Google also does extensive analytics and assessment of their hiring process. They look at how candidates are rated in each of the areas to see which are the best predictors of success. They also look at how well each interviewer does. If an interviewer recommends someone to be hired, are they actually hired? Are the new hires consideed to be performing well after some amount of time on the job? Analytics like this can help them find the best interviewers and judges of character in the company and make sure they are doing more interviews while weeding out those who might should not be interviewing. The average Googler spends 1.5 hours a week involved some way in the hiring process. In short: Google puts a lot of time, focus, and science into making sure they hire the right people.

The author summarizes his advice with:

Set a high bar for quality
Find your own candidates
Assess candidates objectively
Give candidates a reason to join

Chapter 6: Let the Inmates Run the Asylum

Google tries to avoid top-down control and create a hierarchy-less environment. They eliminate status symbols and try to put managers and executives on the same level as everyone else. They even go as far as to eliminate words like "global" and "strategy" from executive titles. Executives get the same benefits as everyone else. They want people to run the company collaboratively as stakeholders.

They also strive to make data-based decisions rather than go on opinions. They pull out one of my favorite quotes:

"If you have facts, present them and we'll use them. But if you have opinions, we're gonna use mine" - Jim Barksdale (CEO, Netscape)

They work very hard for people to understand the reasons why decisions are made and to make sure people see the decision-making process as open and credible. Promotions are a very sensitive issue. People see lists of who gets promoted and make assumptions about the reasons why. Maybe they were friends with other managers. Maybe they were in high-profile projects or areas. People Operations did extensive analysis and found that these types of things had negligible predictive power in explaining promotions. They regularly post information about these types of analyses online and will walk anyone through the reasons why someone was promoted if they come ask - all to be open and make sure that rumors and infighting don't spread.

Google also wants to offer employees a way to change the company. Googlers can submit ideas for making the workplace better, and most of the ideas they try come from employees. One idea that became big was the Googlegeist survey - an annual survey similar to our Associate Engagement survey (except Laszlo says that Google specifically avoids anything as nebulous as "engagement"). They poll Googlers on how the company is doing and how people feel about working for the company and look for ideas for improving the company. Results are publicly available so everyone can see how every team, manager, and department are doing. Many of the ideas and opportunities that are called out are picked up by associate-led groups to see if they can do something about it. These aren't formal cultural committees or groups - just people who see a problem and want to help fix it. Googlers have "20% time" to work on any ideas or products that might add value to the company and much of that work is done in this time. These grassroots-led efforts are much more effective than any top-down edict from senior management.

He ends by saying that if you expect little from people, you will get little. If you expect a lot, you will get a lot. (My view on this is that people usually know what the real problems are in a company. The trick is to empower associates to fix those problems and to either have management get out of the way or help them where they can.)

Chapter 7: Why Everyone Hates Performance Management, and What We Decided to Do About It

Both managers and employees hate performance reviews, but when you try getting rid of them people get upset. When you try changing them people get upset. HR VPs at multiple companies have observed that while everyone agrees they hate the current system, they also agree they hate whatever new system you are proposing even more. There doesn't seem to be a way to win with performance management. Google used to have a very complicated rating scale where employees were rated every quarter with a score between 1.0 and 5.0 with one decimal of precision. Research showed that this scale was too precise to calibrate and really didn't add any value. Instead, they moved to a simple 5-level scale:

Superb (top 5%)
Strongly Exceeds Expectations - 20%
Exceeds Expectations - 30%
Consistently Meets Expectations - 40%
Needs Improvement (bottom 5%)

If people are rated a 2 or better they are doing fine. The real goal of this is to find the to performers and the bottom performs - the two tails. Eliminating the stigma of being rated a 2 made managers more honest in their ratings and helped identify those who deserved recognition and those who needed help. (Maybe Kroger could do something like this?)

Laszlo emphasizes the importance of calibrating ratings across managers so that you know they are being consistent. They also train managers to avoid the typical types of bias. They even go as far as to print out information cards with common types of bias and ask that managers use these in their calibration sessions. An excerpt from one of these (shown page 166) defines the biases:

Horns & Halo Effects - When the overall impression of someone as generally amazing/terrible clouds judgment against new evidence that might point to the contrary
Recency Effect - Tendency to remember the last few things someone did and to eight them disproportionately
Fundamental Attribution Error - Either paying too much attention to a person's "ability" and not enough to the situation/context that impacted their performance, or vice versa
Central Tendency - "Playing it safe" by rating close to the midpoint
Availability Bias - Mistaking what's easy to bring to mind with what's more frequent

He ends by suggesting the following 4 keys:

Set goals correctly - Goals at Google begin with the CEO's OKRs (Objectives and Key Results) and others below then defining their goals to make sure they contribute to these. Goals should be ambitious and difficult (or even impossible) to obtain. If you are hitting all of your goals all of the time, they are not ambitious enough. Push for more.
Gather peer feedback - Make sure you are getting 360 reviews from peers working with someone. They know what that person is really like.
Use a calibration process to finalize ratings
Split rewards conversations from development conversations - Don't have a pay or salary discussion with someone as part of the performance review. Focus the performance review on how someone is doing and finding ways for them to challenge themselves and achieve more. Then have a separate conversation (at least a month later) about pay. Tying pay increases to performance reviews and having one conversation about both is a great way to make people defensive, antagonistic, and unwilling to grow.

Chapter 8: The Two Tails (The biggest opportunities lie in your absolute worst and best employees)

The goal of effective performance management is to identify your extreme good performers and bad performers. Find out what those best performers do well that makes them succeed and find ways to teach this to the lower performers - especially those at the bottom. Google identifies its lowest 5% performers each year (this is done separately from the performance review) and has a discussion with them that goes: "You are in the bottom 5 percent of performers across all of Google. I know that doesn't feel good. The reason I'm telling you this is that I want to help you grow and get better." Don't make poor performance the "kiss of death" or immediately fire the poor performers. Instead make it a learning opportunity where the person can get increased attention and training. They will either appreciate this and grow or decide that this isn't the right place for them to work. Either way, it's better than making someone think they are doing well by ignoring the problems they are having and then having to fire them suddenly when times get tight and a budget gets cut.

Then you need to put your best performers under a microscope. Google does this through their People and Innovation Lab (PiLab). Some of this groups first studies included:

Project Oxygen - Setting out to test the hypothesis (shared by many engineers) that managers don't matter, and ended up demonstrating that good managers were crucial
Project Gifted Youngsters - Trying to understand what people who sustain the highest performance for the longest periods of time do differently than everyone else. They looked at the top 4% of performers compared to everyone else and then the top 0.5% of performers compared to the top 4%)
The Honeydew Enterprise - Understand what behaviors and practices most foster and inhibit innovation among software engineers

A deep dive on Project Oxygen was presented. This project began by finding the best and worst managers in Google. They used a manager's performance rating and their Googlegeist score (a measure of how happy the team was with the manager). They then identified the managers who scored the best on both of these metrics and the worst on both of these. They also were able to identify the key questions on the Googlegeist survey that really predict whether a manager is good or bad. Then they did an experiment to see if good managers and bad managers make a difference. They were able to watch people changing positions from the good managers to the bad managers (or vice versa) and see how this impacted those employees responses on the Googlegeist survey. They noticed dramatic changes in these employees' happiness and satisfaction with the company. Those moving to a good manager saw huge improvements while those moving to a bad manager saw increased dissatisfaction. "The biggest changes were on questions that measured retention, trust in performance management, and career development. Switching to a worse manager was - by itself - enough to transform someone's experience of Google - chipping away at their trust in the company and causing them to consider quitting." (page 194).

Google used the attributes of the good managers to build a "prescription for building great managers":

The 8 Project Oxygen Attributes
Be a good coach.
Empower the team and do not micromanager.
Express interest/concern for team member's success and personal well-being.
Be very productive/results-oriented.
Be a good communicator - listen and share information.
Help the team with career development.
Have a clear vision/strategy for the team.
Have important technical skills that help advise the team.

They also provided info sheets with details on how managers could improve in each of these areas. They created a semiannual "Upward Feedback Survey" asking teams to give anonymous feedback on their managers. Some sample questions from the survey include:

My manager gives me actionable feedback that helps me improve my performance.
My manager does not "micromanage" (i.e., get involved in details that should be handled at other levels.)
My manager shows consideration for me as a person
My manager keeps the team focused on our priority results/deliverables
My manager regularly shares relevant information from his/her manager and senior leadership
My manager has had a meaningful discussion with me about my career development in the past six months.
My manager communicates clear goals for our team
My manager has the technical expertise (e.g., coding in Tech, accounting in Finance) required to effectively manage me.
I would recommend my managers to other Googlers

The author cites a book, The Checklist Manifesto, that explains how powerful checklists are in helping people remember what to do. For managers, this survey is the checklist. Managers receive their survey results in a web page that they can choose to share with their team or not. (They are encouraged to share and to set a personal example by sharing, discussing, and acting on feedback to get better.) They also separate this survey from the manager's performance review - just to make sure this is seen as a tool for growth and not something that should be gamed or manipulated. Google also offers classes such as "Manager as Coach" or "Career Conversations" that they can show improve manager's ratings on these qualities. (In fact, they even know the exact percentage increase on each question that you should expect after taking the class.) These classes are led by other Google managers - the ones who score the highest in each of these areas. This program is a model for what you can do with any type of job role to identify high performers and those who need improvement and then to build checklists, feedback loops, and curriculum for getting better. This is how you keep people focused on personal growth.

Chapter 9: Building a Learning Institution

Your best teachers already work for you... Let them teach!

As mentioned in the previous chapter, once you have identified your best performers and the areas where they have mastery of skills that others need - you have identified the best instructors you could ever find. Don't hire outside instructors. Have the people do the teaching who know not only the general knowledge but also how to apply it within your company. Also, don't be worried about taking your best performers off of the front lines for a little while. If they spend 10% of their time teaching, the rewards you will get from everyone else applying those skills 100% of the time for years to come will more than offset the time you lost from that one person. As Andy Grove (CEO, Intel) said:

Training is, quite simply, one of the highest-leverage activities a manager can perform. Consider for a moment the possibility of your putting on a series of four lectures for members of your department. Let's count on three hours of preparation for each hour of course time - twelve hours of work in total. Say that you have ten students in your class. next year they will work a total of about twenty thousand hours for your organization. If your training results in a 1% improvement in your subordinates' performance, your company will gain the equivalent of two hundred hours of work as the result of the expenditure of your twelve hours.

Of course, you have to train people in things that actually will make a difference in productivity. Google emphasizes only having formal classes if you can measurably show the value of taking them (as with the courses for managers that showed improvements on their team feedback scores). Donald Kirkpatrick put together a framework for measuring the effectiveness of training at 4 levels (each more powerful than the last):

Reaction - Ask the student right after the course if they felt like it was value add
Learning - Assess the change in the student's knowledge or attitude
Behavior - Can you measure any changes in behavior that resulted from the training?
Actual Results - Can you measure changes on key performance metrics like sales or code quality?

A good way to measure these last two levels is to find two similar groups and give one group training. Then see if there are any noticeable differences. If not, don't waste your time rolling out the training to more groups or even worse, launching a company-wide initiative.

Chapter 10: Pay Unfairly

What you pay people is a very complicated matter. Your best people are probably worth a lot more than what you pay them. Numerous studies have been done that show employee performance tends to follow an exponential distribution rather than a normal one. You have a few high-performers who are driving most of the innovation and expertise at your company.

How many people would you trade for your very best performer? If the number is more than five, you're probably underpaying your best person. And if it's more than ten, you're almost certainly underpaying.

The ability of individuals on a team to contribute to the company's goals regularly varies by as much as 300 to 500%. In some cases, two people doing the same work can have up to a 100 times difference in their impact, and in their rewards. Google tries to pay people based on their contribution, so you might have people on the same team making 5-20 times more than the lowest performer on a team. (Personal note: it's probably easier to do this with discretionary bonuses rather than fixed salaries that are less flexible year-to-year.) You have to be careful doing this though. You have to be able to explain why the pay difference exists and it has to be fair. This is probably why most companies avoid doing this and instead focus on normalized salary bands with not more than 30% variation in each band. However, this system will incentive your very best performers - those worth the most to the company - to quit and go work somewhere that values them closer to what they contribute.

The author also suggests the following:

Celebrate accomplishment, not compensation - Google tried to reward special teams with a Founder's Award and big cash payouts. This caused more problems than it created and was eventually abolished. Now they just recognize teams publicly with praise and attention. Any additional bonuses or compensation are paid privately and are based just as much on how you achieved your goals as well as the goals themselves.
Make it easy to spread the love - Make it easy for anyone to publicly thank or appreciate someone else in the company. Google has an internal app for this and posts thank you's and spotlights on walls
Reward thoughtful failure - If a team aimed high and missed, reward them for taking the risk and doing the right thing. Google rewarded a team that created a new, innovative product that totally failed because the team had done all the right things and succeeded in the areas that Google values (innovation, courage, and teamwork)

Chapter 11: The Best Things in Life Are Free (or Almost Free)

There are two incentive programs at Google that really cost a lot money: free food and shuttle service to/from the office. But other than these, the programs Google offers are very low cost and could be implemented by almost anybody. An illustrative list is given on page 274. The free programs (free to both Google and employees) include:

ATMs
Bureaucracy Busters - Groups tasked with eliminating the small, bureaucratic annoyances that get in the way of work. Problems are expected to be small enough that they can be solved in less than 6 months with $30k or less.
gTalent Show
Holiday fairs
Mobile libraries
Random Lunch
TGIF - All hands meetings every Friday where executives share information on the company's direction and take open questions from associates

The following programs are free to Google, but employees have to pay for them. Most of these are offered by third-party groups that come on-campus to sell goods to Googlers. (Because of its size and the convenience to these merchants, Google is also able to negotiate bulk discounts on many of them.)

Bike Repair
Car wash and oil change
Dry cleaning
Haircuts and salons
Organic grocery dlievery

The following programs have negligible cost to Google and are offered free to employees:

Concierge
Culture Clubs
Employee Resource Groups
Equality in benefits
gCareer (return to work program)
Massage chairs
Nap pods
Onsite laundry machines
Take your child to work day
Take your parent to work day
Talks @Google

All of the programs are evaluated based on whether they improve efficiency for Google employees, improve community, or because they are just the right thing to do. You spend most of your life at work. Shouldn't you make your workplace more enjoyable and help people be more efficient so they don't have to spend non-work hours in errands that take time away from family? Google is always looking for ways to add more benefits for employees and tries to find ways to say "yes" to employee ideas rather than the (easier-to-do) practice of pointing out potential problems.

Chapter 12: Nudge... A Lot

In Google's experiments they found that dictating what people do rarely works. Instead, it encourages people to resist and fight back. Instead, it is better to get them going in the direction you want with gentle nudges. An example was given about trying to improve health by getting employees to eat less sugary snacks. Sugary pop and candy was in every break room. Taking the snacks out and replacing them with healthy fruit and vegetables caused employees to get angry. Posting signs about the number of calories being consumed had no effect. But putting the healthier options on the counters and hiding the sugary snacks in opaque bins under the counters caused many more associates to switch to the healthier options - without restricting peoples' ability to choose. Google went about this experiment with the same rigor as any others, counting the number and types of snacks consumed in break rooms for a few weeks before and a few weeks after each experiment to fine-tune a method that worked without upsetting employees. This was then rolled out across the company and resulted in millions fewer calories being consumed each week.

Google also nudges associates with emails and information that can help them just before making important decisions. One example involves an email reminder that goes out the day before a new hire joins a team. This email reminds the manager to:

Have a role-and-responsibilities discussion
Match your Noogler with a peer buddy
Help your Noogler build a social network
Set up onboarding check-ins once a month for your Noogler's first six months
Encourage open dialogues

Managers actually appreciate this checklist as they often don't spend a lot of time thinking about onboarding until right before the person's first day. Then they are worried about what to do with that person on their first day.

New-hires also get presented with their own checklist in orientation. This includes:

Ask questions, lots of questions!
Schedule regular 1-on-1's with your manager
Get to know your team
Actively solicit feedback - don't wait for it!
Accept the challenge (i.e., take risks and don't be afraid to fail... other Googlers will support you).

Of course, they have data to back up the effectiveness of this checklist. New-hires who get the checklist are 10% more productive in their first 6 months at Google than those who don't. When summed over all new hires, this "adds up to a 2% improvement in productivity for the whole workforce. It's like getting a free employee for every fifty you hire, or in our case a hundred free employees for every five thousand we hire. Not a bad return for a 15-minute presentation and an email."

Google also encourages employees to save more for retirement by sending them a reminder of the 401k program just before they get a raise. They also have the opportunity to increase their retirement contribution in advance and in a way that the extra money doesn't come out until they the first paycheck that reflects the raise. This made saving more money relatively painless. 78% of employees who received this offer took it. 80% stuck with it. And average savings increased from 3.5% of salary to 13.6% of salary. (Employees with low levels of contributions were specifically targeted by this program.)

These types of nudges - while each of them is small - add up to huge changes in the company and in peoples' lives.

Chapter 13: It's Not All Rainbows and Unicorns

Google experiments a lot on their employees, and these experiments do not always go well. Examples were given of times when these programs resulted in anger, resentment, and backlash from employees. One example included a program called "Meatless Monday's" where no meat was served in the lunchrooms. Some employees were furious. Some protested with barbecues in the parking lot (which Laszlo thought was a pretty great idea) while others wrote notes to executives with profanities or threw away plates and forks from the lunchroom in protest.

A program to give employees $5k if they bought electric cars (at a time when electric cars cost $5k more than regular) was discontinued when prices normalized. Some employees were upset. They thought this was a perk that they deserved and that management was evil for taking it away.

One of the longest email threads in Google's history involved a fight over a lunch menu that listed a "Free Tibet Goji-Chocolate Creme Pie." The pie was "free" (like all the lunch items), but some took this as a political statement on Tibet's status as a country. Backlash against the chef got him put on administrative leave for 3 days. Backlash against this response was even worse as people began to question Google's stance on free speech and a long, difficult discussion about corporate values and peoples' ability to say what they want or fear repercussions from the company ensued.

The author says there will certainly be tough times like this that will try your resolve. He advises that you take these conversations as evidence that you are doing the right thing - that you are encouraging active participation in the process of shaping the company by everyone in it. It won't be fun, but sticking to your values will get you threw it.

Some organizations will declare defeat, pointing to the smallest backsliding as evidence that people can't be trusted, that employees need rules and oversight to force them to serve the company. "We tried it this way," they'll declare, "and look where it got us. Employees got made, or wasted money, or wasted my time."
Other leaders will prove to be made of sterner stuff. Those of you who, in the face of fear and failure, persevere and hold true to your principles, who interpose yourselves between the forces and faces buffeting the organization, will mold the soul of the institution with your words and deeds. And these will be the organizations that people will want to be a part of.

Oftentimes, you will have to admit that an idea was a bad one and then reverse it. This was the case with Meatless Mondays and several other experiments where management had to admit they made a mistake. The author has the following advice for handling mistakes made during the course of experimentation:

Admit your mistake. Be transparent about it
Take counsel from all directions
Fix whatever broke
Find the moral in the mistake, and teach it

Chapter 14: What You Can Do Starting Tomorrow

The author summarizes the book with ten steps you can do right away to transform your team or workplace:

Give your work meaning
Trust your people
Hire only people who are better than you
Don't confuse development with management performance
Focus on the two tails
Be frugal and generous
Pay unfairly
Nudge
Manage the rising expectations
Enjoy! And then go back to #1 and start again.

My Take-Aways

Alright, so now I need to summarize all that into key take-aways for how we want to run Ops Research. I think the 5 keys I'd focus on are:

Define our mission a. A simple, ambitious, open-ended mission statement that gives work purpose and inspires everyone on the team to come to work every day.
Only hire the best a. Define a recruiting/hiring process that uses work sampling and structured interviews to find the best candidates b. Ensure that management is aligned on the importance of hiring only the best. If we all agree it is better to have an unfilled position than the wrong person in that position, then we should be good to go.
Empower associates a. Make sure there is a way for associates to provide feedback and suggestions for improvement b. A team survey (like Googlegeist) might be good for this. c. It could also serve as a way to evaluate the management team
Do a self-assessment on what skills it takes to succeed on the team a. What skills do people have? b. What skills do we need? c. Where are our potential instructors for specific skills?
Set up learning groups to teach key skills

These are listed in their order of importance. We could keep going with more and more ideas from the book, but I think #3 (the survey and team feedback) will be key in finding opportunities for where we want to go next. Where do team members see opportunities today? What issues do they want to address? If we can put in place a culture where asking these kinds of questions and making recommendations is encouraged, then the rest should take care of itself. (And we can always come back to the book later if we run out of ideas...)

Reading List