Wednesday, April 26, 2006

Early Amazon: Just do it

Later, Amazon thoroughly embraced the reckless passion of innovation. "Just do it" became a rallying cry.

In typical display of Bezos silliness, "just do it" was codified in the "Just do it" awards. Recipients were brought up in front of the entire company and given an old, used Nike shoe.

I got a couple of these -- including one for shopping cart recommendations -- but, after moving to Stanford and back to Seattle, the old, stinky, mismatched shoes have long been lost.

What was not lost was the sense of pride. I was proud to have gotten that crappy old shoe.

Of course, it was not the prize itself that mattered. It was the recognition. It was that someone had noticed and said thanks. That was what I wanted.

Recently, I have had some interesting debates about management goo with a couple different friends who work at two large, well-known Internet companies. One thing we were talking about was compensation strategies.

My opinion has swung 180 degrees on compensation in the last few years. I used to be a huge fan of pay for performance. Salaries, I thought, should be varied by how well you do and what you have done.

Now, I believe that pay and perks should be high but basically flat; exceptional work should be recognized in other ways.

What changed my mind?

While merit pay sounds like a great idea in theory, it seems it never works in any large organization. It appears to be impossible to do fairly -- politics and favoritism always enter the mix -- and, even if it could be done fairly, it never makes people happy.

A recent example of this is the news about poor morale at Microsoft because of their elaborate performance review and merit pay system. It is seen as unfair. It makes everyone unhappy. It just does not work.

Instead, compensation should be high but basically flat. Merit rewards should focus on non-monetary compensation. Maybe even an stinky old shoe.

That used shoe was worth far more than it might appear. It was a thank you. It was recognition. These are things valued by many, but offered far too rarely.


amckinnis said...

Greg -

Never lose the "Just do it" - "get the wheels on it, we'll fix while it's rolling" mentality. It's driven at the top and will easily flow through the company. But the more someone has to RISK, the harder it is to have the attitude.

Personally, I remember when I was single, no kids - I could (and did) it was easy to have an attitude of "damn the torpedos, full steam ahead". But now the RISK is much higher.

But nothing is life is not composed of RISK - so go ahead and jump and jump and jump.

Scott said...

Greg, check out Mary Poppendieck's article, Team Compensation (pdf). It's a good read on why bonuses/salary rewards can be the poison of a team.

Greg Linden said...

Thanks, Scott. That's an interesting article.

The book "The Human Equation" is also a great read if you're looking for more on this.

Greg Linden said...

Hi, Paul. I'm not sure that works either.

The problem is that it is nearly impossible in any large org to fairly evaluate whether a complex task was properly completed. Your approach probably would still make people unhappy.

What makes people happy? Often, it seems to be relative gains, not absolute gains. It seems to make people unhappy if they perceive themselves to be doing poorly compared to those around them.

With merit compensation, a majority of the company will be unhappy because they see themselves as doing poorly compared to those around them. In addition, they will see the way to improve as to compete against those around them, others at the same company. That can be quite destructive.

With high but flat compensation, everyone at the company is doing well, better than those outside the company. They feel lucky to be at the company, part of a select group, fortunate to be there. Even more important, rewards come only from helping the company succeed against other companies, not from competition against other people around you in your company. People work together to succeed.

It's a bit non-intuitive. What might work well at an individual level works poorly at an organizational level. You don't want people in your company to see the path to success as competing against those around them. The path to success should be to compete against other companies.

Anonymous said...

Your comments sound like a reasonable hypothesis, but I really question whether you have any data beyond a few anecdotes to back it up. I could just as convincingly argue (as many do) that knowing that whether you're #3 or #20 (in terms of skill and value of contribution to the team) you make the same actually depresses the work efforts of the average person. I'm not talking about the #1 or #2 person, who can win awards and public recognition. Most people will never get those things and, if they know that there's only a 1% difference in pay between working pretty hard and working just enough to keep a job, they'll trend down. What they're actually doing is moving their efforts elsewhere (like into their personal lives).

My hypothesis is that no company will ever have enough non-monetary opportunities for reward and recognition to really give a significant percentage of their employees a chance to feel "fulfilled". And, if they keep getting flat salaries, they'll feel like they can't keep up with expenses in the real world.

Sure, Microsoft's compensation methodology shows how you claim to be a meritocracy but still be purely political in doling out money. But that's just one data point. I've seen companies and managers actually dole out money fairly, based on performance too. Maybe you haven't.

In fact, books I've read like PeopleWare, and studies from Watson Wyatt, show the exact opposite of what you claim. Specifically, that those companies that create the most difference between how their best and average employees are rewarded with money far outperform their peer companies.

I suspect this is because the best and brightest contribute 3-10x as much as the next employees. And so retaining those people should be a #1 goal of any company.

It's not just money, of course, it's all the recognition, including the types of things ("Just do it" awards) you talk about. But to claim that money (which buys personal freedom, the ability for your spouse not to work, pay for your kid's college, pay for your house, etc.) should just be "out of the equation" is to ignore lots of research about the "heirarchy of needs" that people have.

For me it's not about earning more than my peers. I don't specifically know that they make. It's about keeping ahead of my expenses, saving for retirement, etc. And when I get more money, I can do those things easier, and it's a big load off my mind. If I know I'm at the "top" of the raise level for that year because of my hard work, it feels good. Sure, I also want other forms of recognition, but money's important.

Greg Linden said...

Anonymous, I may not have been clear what I mean by "high but flat" compensation.

By "high", I mean that you'd have a hard time finding a better salary anywhere else.

By "flat", I mean that there are not large differences in pay between you and people similar to you in the company.

I do not mean that people never see salary increases, have difficulty meeting expenses, or be able to find better compensation elsewhere. I do mean that they should not be competing against their co-workers to increase their salary.

I have read Peopleware. I don't recall it making arguments that support what you are saying. As I remember it, the book suggests methods for designing workplaces to improve productivity, discusses the need for people to be uninterrupted to get into the flow, and talks about methods to build team cultures. I might be wrong, and I would appreciate it if you could point me at specific passages that support what you are saying.

To be fair, you are correct that there is an active debate in management research about the benefits of merit pay. As you would expect in the social sciences, a definitive answer is muddled under a sea of conflicting studies. While I clearly sit in one camp on this debate, there are many who just as forcefully argue the other side.

I can say that I did very much enjoy the book "Strategic Human Resources". I found that it takes a balanced view of both sides and broadly surveys the evidence.

Anonymous said...

i've never seen this mythical place that has enough money to give everyone significantly above average industry salaries and also spends time significantly rewarding large swaths of the worthy individuals with other recognition and rewards.

What I have seen is that many companies, when they divide money evenly, merely depress the wages of the highest earners so that overall compensation outlay is kept the same as it might have been under the "unequal/bell curve" distribution. IOW, there's no "high" to go with the "flat".

It's not even just a question of merit vs. non-merit pay. I'm fine with the argument that all "SDE Level IIIs" should be paid around the same. But SDE Level IIIs should be certainly be paid a lot more than SDE Level Is. So your pay is based on demonstrating merit that makes you promotion worthy.

Unless you're arguing the opposite, that everyone within a class of jobs (all SDEs) should make the same, I fail to see how this "flat" notion is really "flat" at any macro level.

Too often companies strive to "reduce competition amongst employees" (which I agree is laudible) by merely limiting the amount to which anyone is incented to stand out.

From what I've seen (and plagiarizing a phrase), "Merit pay is the worst compensation method. Except for all the other compensation methods".

Anonymous said...

Another attempt to address the pay for performance issue is by creation of an employee owned company via an ESOP (employee stock ownership plan). This name is a bit of a misnomer, because it isn't ownership, but rather the employees are the beneficiaries of a trust (the ESOP itself is the trust). The idea is that each employee is allocated a percentage of the shares based on his or her salary compared with the total payroll. So, for example, if the total payroll is $10 million and someone earns $100,000, they would be allocated 1% of the shares for that year. Someone earning $50,000 would receive 0.5% of the shares. The company is then independently analyzed on an annual basis to determine its value, and whatever it's valued at is divided up amongst allocated shares. Typically, ESOPs are used to buy out existing owners, so shares are released as debt to the original owner is paid off, so there are shares to be allocated in the early years. (When shares run out, that's a different story.)

While ESOPs are designed to be an incentive, they have some fatal flaws:

1) The person in charge of the ESOP must act with the employees in mind. Thus, salaries must be at a market rate and cannot reflect a person's actual contribution.

2) Shares cannot be redeemed until the employee quits, which is a big incentive to quit once the employee reaches a sizeable sum in the ESOP plan rather than just staying with the company.

3) To address the problem of employees wanting to quit to cash out, most ESOPs have a provision that defers payment for five years after the employee quits, and allows payments to be stretched out over five years or a longer period of time if the account is worth more than some dollar value. This does not apply if the employee reaches age 65, which is an age discrimination policy if I ever saw one; the 65+ year old person is paid immediately upon termination of service, whereas someone who is 55 but might have contributed 10 times as much as that other person has to wait five years. To add additional risk, during the five years that the employee is waiting to receive payment, the dollar amount is not fixed in value, but rather is held in the stock. So, management could decide to issue 300% more stock and dilute the retired employees' ownership significantly before they receive any payment. Of course, the stock value could go up significantly but it's just a gamble at that point.

4) Along with these provisions, ESOPs are typically accompanied with systematic brainwashing to make employees think they are owners. They're not owners. Owners have responsibility and have risk; employees have no risk other than losing their salary, which is the same risk as any other job. But, with an ESOP, the idea that every employee "owns" part of the company is a good excuse to try to make people work hard like an owner (nights, weekends) but reward them like an employee (here's your 2% raise! owners are in it for the long haul, so your reward is the ESOP!).

So, with these flaws, an ESOP is an attempt at manipulating the employee into working harder by calling them "owners". The ESOP incentive is eliminated because any reward is too far in the future and far too risky to be an incentive and limits the ability for companies to reward key employees appropriately. With all of this in mind, the ESOP is a plan that's great for the company but completely ineffective in terms of a motivational strategy, and in fact makes the perception of unfairness referenced in that Mary Poppendieck article "Team Compensation" become the fundamental compensation principle. While in theory stock allocation in an ESOP is fair (you receive your percent as compared to others), it's actually not reflective of each person's contribution since salaries have to be set at a level that has other employees in mind (i.e. market rate) but the person can be contributing twice as much than another person receiving a very similar salary.