The Barefoot Philosophy Cornerstone #3: No Salaries

[This is part 4 of an 8 part series: The Barefoot Philosphy.  It is based on my experiences as the founder of a business—Barefoot Software—which I ran for 12 years.  Please start with the intro.]


Salary is one of my personal hang-ups.  I hate it so much that I’m not sure I can adequately describe it.  But of course I’m going to try.

The basic idea behind the concept of salary seems to be this:  We expect each employee to produce X hours of productivity (we’ll get to the sticky question of what “X” is in a bit).  Now, some weeks employees will produce exactly X hours.  But this is going to be the exception.  It’s far more likely that some employees, some weeks, will produce less than X.  In which case the company is getting screwed.  Whereas, other employees, or other weeks, will produce more than X.  In which case the employee is getting screwed.  Salary is, therefore, a system by which the vast majority of the time, someone is getting screwed.

If I were to come to you today, and you’d never heard of the concept of salary before, and I were to explain it to you, you’d look at me like I was insane.

And yet this is the way the vast majority of corporate workers are paid.  In fact, it’s become a status symbol in some weird way: hourly pay is for blue-collar workers.  White-collar employees get salaries.  It’s like you’re proud that you’re getting screwed.

It’s completely unfathomable.

The whole system seems to be predicated on the idea that the amounts that everyone is getting screwed will somehow balance out in the long run.  Perhaps ... for the company.  If it’s lucky.  But almost never for the employee, because it’s less likely that there’s variance over the weeks and more likely that there’s variance over the employees.  That is, some people are just prone to work more hours, and some people are just prone to work fewer.  So your underperformers are likely consistently screwing you, and your overperformers are likely consistently getting screwed.  And this is penalizing exactly the wrong batch of employees.  The last thing you want your most productive employees to do is wake up one day and wonder why they’re slaving away for you and getting paid the same amount as the people who slack off and only put in half as much time.

At Barefoot, we had a simple solution for this dilemma.  Everyone got paid hourly.  Everyone.  And we had a very simple policy: if you’re working, you’re getting paid.  Corollary: if you’re not working, you’re not getting paid.

Now, there were technical hurdles to overcome here as well.  We developed our very own “timer” program, and everyone had to remember to “start a timer” when they started working, and “pause the timer” when they stopped for the day.  Or when they got interrupted.  Or when they just moved on to a different task.  This was a difficult habit for us to form.  But we helped each other out.  Just as everyone knew what everyone else was making, everyone knew when you were running a timer.  (The timer program was run off a central server, so you could access anyone’s timer from anywhere.)  And it was everyone’s responsibility to police each other.  If you saw your coworker talking on the phone to his mom (or girlfriend, or bank, or whatever), you checked their timer to make sure they’d remembered to pause it.  If they hadn’t, you sent them an IM reminding them.  (Or just waved at them and pointed at your wrist, which was often sufficient.)  We did this to each other so often that we never got offended—after all, you weren’t accusing the other person of defrauding the company, which was never true.  It was always the case that they just forgot.  (If your employees are intentionally trying to rip you off, you’ve screwed up all the other cornerstones, and also your hiring process.)  So we were just being helpful.  The rule of thumb was: if you’re going to be away from the work for more than five minutes, pause the timer.  A quick trip to the bathroom was okay, or perhaps a dash to the kitchen to get a drink, but extended hallway conversations or smoke breaks meant you paused.  New people found this weird at first, but you rapidly get used to it and it becomes second nature.

Your timer, by the way, was not only what we used to pay you, but also what we used to bill the customer (assuming you were a billable employee, which most of us were).  So people were not only conscientious for their own sake, but also for the sake of the customers (more on why that would be true when we hit our last cornerstone).

Next, we have to consider what a reasonable value for “X” is.  This brings us to one of my favorite corporate myths: the myth of the 40-hour week.  I could probably write an entire blog post on this by itself.  In the interest of brevity, though, I’ll just hit the highlights here.  The fundamental problem that corporate managers seem to have when considering the 40-hour week is mixing up two fundamentally different measurements of time.  At Barefoot, we always kept them very distint, and we called them “work” and “availability.”

If you work from 9 to 5, as is typical in the corporate world, then you are at work for 40 hours.  This is radically different, however, from producing 40 hours of work (or productivity, or however you want to phrase it).  You can’t possibly do 40 hours of work in 40 hours of clock-time.  It’s not physically possible.  You have to go to the bathroom, at the very least.  And you have to eat every now and again.  And your family is going to call you, and your coworkers are going to pop by to talk about the game this weekend or that party they’re having or whatever.  And, even if you could approach 40 hours of work in 40 hours of time, you shouldn’t: it kills your productivity.  You need those regular breaks to keep your mind fresh.

At Barefoot, our completely unscientific research showed us that almost all of our employees produced between 30 and 35 hours of actual, productive “work” if they gave us 40 hours of “availability.”  We defined “availability” as any time that you’re available to do work, even if you’re not actually doing it.  You were generally available while you were in the office, of course, but you were also allowed to be unavailable in the office.  Or you could be available at home (obviously we encouraged working from home), although most of the time you were unavailable.  Your availability was set when you logged in for work, and unset when you logged out.  So, just like anyone could see your timer, they could tell if you were available or not.  (In our case, your availability was tied to our IM client.  Everyone was required to be online in IM when they were availble.  That made working from home, as well as full-time telecommuting, feasible.)

So, if we expect that people are only going to put in about 30 hours a week, then we need to pay them accordingly.  Which is exactly what we did: to pay a competitive wage, we took an annual salary for what the employee was worth (remember: merit-based pay), turned that into a weekly salary, then divided by 30 to produce an hourly rate.  So if you worked 30 hours for us, you’d make about what you made elsewhere.  If you worked more, you made more.  If you worked less, you earned less.  Nice and simple.

Of course, employees had to pay for their own health benefits.  If they wanted to take vacation days, they didn’t get paid, so they had to set aside money for that too.  So we just made sure that our hourly rate was high enough to let people cover those things.  Isn’t it better to let people handle those things themselves than try to enforce ever more baroque company policies?

This solves so many problems it ain’t even funny.  First and possibly most obviously, it completely eliminates the need to yell at people for not working.  As long as their timer isn’t running, who cares?  You want to sit at your computer and play Doom, or Star Siege Tribes for a few hours?^*  Fine.  Just clock out first.  You could even make it clear whether you were allowed to be disturbed during your game by setting your availability appropriately.

Just like merit-based pay, this is a self-regulating system.  People who slack off and never get enough work done also don’t earn enough to pay the rent: they either have to start working harder, or find another job that doesn’t mind their lack of productivity.  And the self-regulation is enhanced by the self-forming teams: no one wants the slacker on their team.  Well, unless you need someone to pitch in for like 10 hours a week.  And, hey, if you only want to work 10 hours a week, and you can afford to do that, and you can be useful to the company doing it, why shouldn’t we let you do that?

No salaries also completely eliminates the “death march.”  If we had some deadline and we needed people to put in extra hours, the first thing is, we never had to ask.  Partially because of the other cornerstones, but mainly because some people would want the extra cash in their pockets and they’d just volunteer.  If you weren’t one of those people—if you didn’t want the extra dough, or maybe you just couldn’t afford the extra time away from your family—that was cool too.  But more people would volunteer than not.  Because they knew that they might be putting in long hours, but at least they were getting paid for it.  And some of you more money-conscious entrepreneurs out there might be wondering: what about overtime?  Didn’t we have to pay time-and-a-half?  Well, first of all, if your normal work week is only 30 hours, that gives you a bit of headroom before you technically have to pay anyone overtime.  But, more importantly, the federal rule is, you don’t have to pay people time-and-a-half for overtime if they make more than six times the minimum wage.^**  And we paid everyone so well that that was true for almost everyone in the company (certainly it was true for all our billable employees).

This was probably the most radical departure from traditional business practice for us, but honestly I feel it was one of the most important.  It put employees in charge of their own financial destiny in a way that would have been impossible with salaries, and it completely obliterated a whole slew of sources of corporate friction.

Next week we’ll see what the corporate structure looked like.

Part: << 1 2 3 4 5 6 7 8 >>

* Remember, this was the 90s.

** At least that was the rule at the time; I assume it’s still true.