Archives for June 2015

Data is great, but whenever you see data being applied to areas that you can’t easily measure then it feels like it is a recipe for trouble as the room to distort is abundant.

There are many fields like this but have you noticed how this happens in spades in HR (or “People Ops cos we are talking about people not RESOURCES”)?

At most companies with a sizable employee base you find The Levels. I am filled with dread as I open up the excel spreadsheet that defines each of the degrees of various professions. It is extremely rare that there aren’t far too many stratifications.

Where else do we see the notion of levels? Martial arts are the prototypical examples. As a practitioner you climb up the ranks with hard work and by proving yourself mastering and demonstrating your proficiency. In many of these practices it can be somewhat clear what to master, but even so it isn’t an exact science.

Helen may be able to beat John in 2 out of 3 bouts, but Frank tends to beat Helen even though he can’t handle John.

Also, martial arts and the levels are not just about combat. Whoever defines the level requirements ends up setting the incentive structure and thus defining the type of martial artist they are moulding.

“I have a black belt!”

This phrase is right up there with: “I ran a marathon”. It is a sentence that tells you something about a person. With the marathon runner you know they put a lot of effort into getting their fitness level to push their body over 26 miles. Sure, they may have done the race that didn’t have any hills and with good weather, but even the best case is an accomplishment.

With martial arts it isn’t quite as simple. I doubt that someone had an easy path to black belt, but the milestone differs by a wide margin. I enjoyed Karate as a kid, and have also seen martial arts through the eyes of a parent. A black belt for me was at least a 10 year odyssey, but at some of the dojos they give out belts like candy. I get it: “if we keep giving out belts kids will like it and keep coming (and paying)”.

Should you be able to “test out” and jump to a particular belt if you studied all of the requirements? Or, is the journey part and parcel of climbing the levels? There are many intangibles such as building the relationships with fellow students and instructors.

Can you quantify the value that a job role or a person brings to a business? It is very very hard indeed. There is so much context and external forces (Frank, an engineer, builds a tool that delivers some value, but his profession puts pressure on the salary required to keep him!)

The reality is that compensation is not fair for everyone at your company. The process was never setup correctly to begin with and it has changed constantly over time. Some would gain from that, and some would lose. In the era of acqui-hires, a distortion effect happens where peers doing the same work (but who came in the front door) don’t get the same rewards. While it can be a good lighter fluid to bring in talented people this way, you shouldn’t ignore the negative side effects. You have to ask yourself: if they wouldn’t join the mission as an employee, are they right for us? Are we taking care of our core employees? Are we building a great base to smooth out the distortion?

As a society we tend to have shorter and shorter stints at a job. Executives are also getting paid (proportionately) more and more. Add this to the nature of public markets, and are we not building the incentives for an incredibly short term culture? We may trick ourselves into thinking “it isn’t short term thinking, I am just trying to get shit done!” but I have made many mistakes in judgement by taking an easy short term approach. One reason why “still having the founder” can seem to matter so much to great companies is that they are there to think long term. They aren’t thinking about how “if we can just make it a couple more years we can cash in on our huge executive salary!”

Back to the insane number of job levels. These quickly show employees how unfair compensation is. If there are 13 engineering levels (I have seen this before!) with a large number of engineers, then it is pretty simple to find someone around you who is above you in level but not competency (especially in your biased opinion).

If you are struggling to define the differentiations between levels then you have too many. Matt Briggs has a fantastic article on the role of a senior developer which clearly articulates some differences:

A good junior developer can be given a known task, and be expected to execute it quickly, and well.

A good intermediate developer needs less supervision. They can be trusted to raise issues of code design, and play a valuable role in design discussions. They are also the “workhorses” of the dev team. However, further mentoring and higher level supervision is still vital.

A senior developer is intimately familiar with their own failure. They have written code both under, and over designed, and have seen both fail. They are reflective about the things that they do, evaluating their successes and failures when approaching problems with intellectual honesty. A senior developer has fallen out of love of the complexity which dominates the intermediate, and is obsessed with simplicity.

If you haven’t done this thinking and laid it out for all of the engineers then you aren’t done.

This exercise also allows you to separate various needs, such as:

• How do you handle compensation?
• How do you make it clear who owns particular decisions?
• How are the management and individual track different? (please don’t conflate them!)
• How are we helping our engineers get better at their craft?

It will help you with the hiring process too. The obvious question is: “How should our process work to know if someone is at a particular level?”

You may surprise yourself with the conclusions: “You know what, it has been really hard to get this right. Instead what if we slot people after we have hired them and they have spent time with us?” This is a trade off too, as it means you can’t be as clear and you risk pissing people off if they think they should be more senior than the slotting process found them. There is a constant battle between clarity vs. doing the right thing and the reality that we don’t have enough information up front and this sets people up based on too much luck. This realization may also have you setup the right levers so you can move people around as you find out you are wrong.

At a larger company the leveling can far too often be relative to the group vs. the organization. This happens with performance reviews in spades. If you have ever had to deal with the pain of stack ranking then you know how frustrating it can be. Let’s say you are on a small team with high talent density. You know that a weaker team elsewhere with a poor leader is rating their folks high, but your team with strong expectations blows them all away, yet rates lower.

I have gotten a real appreciation for organizations and setting up the right incentives and tools. In a startup you can wing it so much more, but that is much easier than solving at a bit of scale (truly a small number of people).

I think it will be increasingly important to define the parameters and values, and arm people with the right tools to offer a flexible career path at their job. Expose the trade offs to the entire team and let them be part of the process. Focus on areas such as role clarify vs. too much process in the wrong places. Somethings will not change much over time as they are tied to the company values. Other solutions are solving temporal problems and should be clearly marked as such so the team can understand the why not just the what.