Or, SAO Thinks About Buffer, Part Two. Here’s Part One.
In chatting with folks about Buffer’s approach to salary (making not just the process of assigning a salary, but the final number for each employee) transparent both internally and externally, there have been some common arguments against it. I’d like to talk about them a little, and discuss my thinking about each.
You’ll recall from Part One that I think that the transparent salary system is a great one, though I would impress upon the skeptical reader that my primary philosophical approval is for the transparency of the process – the actual visible list of salaries at the end is, in my view, a pleasant side effect of a much bigger, and more important, piece. The process being transparent is really the lynch pin here, I reckon.
Here are some arguments I’ve heard against salary transparency, and my thoughts on them:
“I think salaries are a private and personal thing.”
I’d suggest that this is a cultural artifact that holds us back. Seeing how salaries are assigned takes something that is currently invisible, and makes it visible. The final numbers are much less important than the fact that the process that results in those numbers is visible and accountable – making the actual salaries visible is simply a check on that process, a verification that it works as intended and displayed. Invisible salaries (and salary assignation processes) open the door for unjust practices that have become endemic, and are likely often simply the result of unknowing implicit biases – women being paid less, minorities being paid less.
I think transparency around salary processes and final salaries may place some tension on our traditional ideas of what should be private, that is certainly true. But, I think that making them public and visible is much better for us, as workers and as a society that desires equity, in the long run.
“I trust our HR department to take care of that.”
That’s great! I also am lucky to work in a place where I truly believe that our HR department has the best interests of the employees in their hearts, and I trust them completely. That being said, I don’t have to trust anyone else I work with – because their work is visible and available and under review from the rest of the company. This black-box nature of salary assignation is not only bad for non-HR employees, it’s bad for folks in HR – it means that they can’t have open and frank conversations about issues that might concern them, it means they’re denied the usual diversity of perspective and insight from their comrades with particularly tricky issues.
As well, it’s worth noting that the Buffer system is entirely self-contained – the questions of salary exist entirely within a box of particular questions and qualifications. During interviews and salary discussions, it becomes much easier and less stressful for the HR staff – no more fuzzy edges or uncomfortable conversations. It’s all in the spreadsheet.
“This is a non-issue. If you’re happy with your own salary, then stay where you are. If you aren’t happy with your salary, then find a new job.”
This mistakes the final result for the process – the question isn’t about the particular salaries of employees, or my salary specifically, but rather visible assurance that everyone’s being compensated fairly. That’s it – whether or not I’m happy with what I’m paid has no impact here. In a more transparent system, I’d at least be able to ask questions about why I’m paid what I’m paid, and how to make moves in the right direction (“So, how can I move from Rookie to Journeyman?” etc).
“I have absolutely no interest in seeing salaries.”
I don’t think anyone’s suggesting that it would be mandatory reading – I’d be curious to know how many folks actually do review the pay sheet, internally, at Buffer.
“Public employers have had transparent salaries for a long time, and they’re famous for being inefficient and having stagnant promotion patterns.”
True as this may be, I think we can acknowledge that there is an essential and important distinction between government employees and folks working at cutting edge tech companies. New companies doing business in new ways bring all sorts of interesting iterations on longstanding traditions that can often bear excellent organizational returns – I’d argue that transparent salary assignation processes is a great example of this. Just because transparent salaries are a property of a class of organizations we do not want to emulate does not mean that it won’t really shine when we try it in a new class of organization. So, let’s do it!
What do you think? Do you have an argument against fully transparent salaries and salary assignation processes?
Simon Ouderkirk 
Interesting Simon. This concept would also show how some people can go broke on plenty and others thrive on little.
If you haven’t already you should read Drive as it has a bearing on this topic. Motivation is not tied to compensation; my worry is that making salaries public might result in more emphasis being placed on it as a motivating factor, and actually reduce our enjoyment in work etc.
If salaries are made transparent so should the way they are calculated, including what other people in similar roles in our location earn, the cost of living the the individuals location etc.
Ben, I 100% agree – in fact I think just final salaries without a transparent process would be a really poor execution of this whole idea – you can see Buffer’s formula here.
And Drive was great! I recommend it as well, for anyone interested in motivation at work.
Interesting. 🙂
I don’t have yet a clear position on the topic, so it’s useful to read different takes on it. 🙂
Can I throw in a couple of questions too?
One side effect of salaries: people would start to rank themselves against it, so tying their self-worth to the number (Social Comparison Theory, if I recall correctly the theory behind it). This happens naturally. How do we avoid that?
The second is: making a process transparent makes it also something that can be gamed. So, in this sense, people might start to “tune in” (even unconsciously) for the metrics that trigger salary increases instead of the actual results. It’s a normal behaviour in social dynamics, that of course is the initial step for explicit cheaters. How would you approach that?
Let’s do it! First: I agree, people will compare themselves to their coworkers (and friends, and family) – that’s a natural part of being part of this social life, right? This problem exists now, with hidden salary assignation processes, so I think we agree that it isn’t something that will be created through the move to transparency. I would suggest that having the process of salary selection be transparent would actually push against this tension in a positive way – you would know who makes more than you, and you’d know why, and you could then ask good questions about why – as opposed to now, where it is all guesswork and anxiety. I would prefer data and quantified questions over ungrounded worry and general anxiety, and I think you would, too 🙂
Second question, I would suggest that the unconscious (and conscious!) ability to game a transparent system can be seen as a strength of a more transparent system, as it allows you to direct the gaming – right? If I want to value growth above all else, then I will build that into the salary assignation process. Then, if my employees change their behavior to gain more pay in pursuing the goals I have for the company, everyone benefits! If an employee gaming a transparent salary process results in suboptimal results for the company, I would suggest the problem is not in the employee behavior, but in the salary process 🙂
Nice answer on the first one. You make an excellent point, mostly I’m curious to see if I will be more or less stressed then. 🙂
On the second, I think it’s a bit too abstract, in the sense that “build that into the process” is still not a metric. Once you fix a metric, any metric, the disconnection between the metric and the objective appears. For example: growth might translate to a metric like % of new registrations. Great. Gaming the system in this context means that I will then pursue “% of new registrations” regardless if that means growth or not. That’s how the disconnect metric/objective appears, and will appear not just if it’s the wrong person – a cheater – but also because it’s natural: focusing on a metric makes people intrinsically over-focus on the metric, thus more often than not forgetting the goal. Again, nothing strange here, no judgement: it happens. 🙂
Great post, Simon! Interesting topic and thought provoking feedback.
Thanks Paul!
I think the transparent salary bit is also important for the trial phase at Buffer. IIRC, they require their trials to work full-time throughout the 45-day window before getting an offer. Quitting your job is a tough decision to swallow for a shot at a new job if you’re not sure you’ll be fairly compensated.
I also really dislike the “Compensation based on skill-level, education, and previous experience” cop-out. While we’re lucky to work for a company that doesn’t do this, it always felt like companies were just scheming to pay the least amount for an employee possible.
I personally have no problem being one of the lower paid people on the payroll if I know the metrics behind how my salary was determined. I imagine that the Experience breakout (Master, Intermediate, etc) is more defined internally somehow. That way, if I’m the lowest paid Developer, I know exactly the steps to take to become an Intermediate developer and up my salary.
Interesting! 🙂
This is only one of many points in your post, and is not addressed in any of the other comments, but this is the single most important argument for salary transparency. Until it’s fixed, it completely trumps any other consideration.
+t+d