I dislike having a blog, particularly a professional one. It doesn’t seem a professionally wise move: you are bound to write something idiotic on them. However, while perhaps not wise, blogs are much too useful to ignore. So useful that I now consider it a professional responsibility to have one, and want everyone I work with to have one too. I find them to be the single most useful tool in the single most important decision that every company and every employee makes: who to hire and where to work.

The hiring process is broken. It’s very hard to hire the right person for the job. None of the normal ways of hiring indicate what you want to find out about an employee, which is not only about their skill set, but how they think. You want to know about who they are on a very deep level: what motivates them, what people are they in conversations with, what do they care about, how are they updating their skills, how does their personality interact with their job, how they react under stress, how they react to success, how innovative they are, how well they can communicate things they know well, and how they process information when they are not experts.

And, you need to observe these things over time. You need to be able to observe the person changing, adapting to new contexts.

The traditional methods we use to hire people tell us little about these things. In fact, they provide us with information that can obfuscate these things. They provide you with bits of data that we think mean something, but they don’t usually mean in reality what we tend to think they mean. For instance.

1. Success in school. Success in school only tells you if the person is a good student. We all know real-world successes that haven’t done well in school, and vice-versa. Being a good student often means a willingness to play well within a certain social structure: if your company has similarities to school (it’s big and bureaucratic with set measures of success) then this becomes more useful. Still, it doesn’t focus in on the skill set you’re trying to hire for (being a good student isn’t even a good way to predict whether someone will be able to teach others to be a good student: it’s a different skill set entirely.)

You’re also dealing with a lot of random factors, and judging people’s behavior from long ago. For instance, if you want to get into Stanford these days, you probably need to begin to behave a certain way, and learn certain things in elementary school. At that age, it’s more about your parents and your social context then your inherent personality. Sometimes these things end up shaping personality, but just as often they don’t.

2. Past successes at work. This tells you a bit more, and in certain situations might be able to tell you a lot. But, even if there are objective indicators that a person did well at a certain job, it doesn’t tell you about the context in which the work took place, which often is more organizationally than personally based. For example, some companies have a good sales pipeline and it would be hard not to do well, in other companies you have to start with nothing. Some companies have great management and it’s easy to do well, in poorly managed companies it can be almost impossible to succeed. This can even vary widely within an organization. Since employees can’t really tell a poorly-managed team from a well-run team at the outset (unless they read their employer’s blog), and no employee with good judgment is going to tell you how difficult it was to work with their former boss, this can be a very tricky area to interpret.

Also, a large part of success is motivation, and it’s hard to interpret the changing motivations in a person’s life. For instance, VCs tend to want to invest in successful entrepreneurs, and often that’s a good idea. However, of all the entrepreneurs I know, I can name as many examples of people that are failing the second time around because they no longer have the same motivation. Wanting to become wealthy and wanting to prove that you became wealthy because you’re smart, not lucky are two very different reasons to work. With one you’re likely to make the most profitable decision, and not care where the idea came from. In the other you’re going to be the one that originates the idea, and you’re going to tend towards implementing things that would make you sound “smart” when you tell the story.

Additionally, people change. Taking the case of a one-time successful entrepreneur trying again, I’ve observed the following changes in certain entrepreneurs. One I know isn’t as singularly focused on work now (he has lots more distractions: conferences, parties.) One I know isn’t focused on a single company. One I know is getting a lot more control the second time around, and seemed to make better decisions when his choices were more reigned in. One really did just get lucky: he’s lost when it comes to navigating difficult circumstances. One I know finally bought into his own hype and now believes he is a super genius, which seems to be getting in the way of him thinking of anything new.

3. Interviews. Some people just interview well. I do, and have certainly gotten jobs I was ill-suited for because of it. I’m just good at sales, no matter what I’m selling. I can sell anything to anyone. In a short time period, I can make anyone like me. Believe me, this has absolutely no relationship to who I actually am, what I will be like every day, what I will be like over time, and what I will be like when I’m no longer in “sales” mode. I could make a decision in an interview not to “sell” myself, because it might lead to a better outcome for us both. But that doesn’t seem like the right choice either. If you’re not selling yourself during the interview, then you risk looking like you can never sell, you seem like you don’t know how to give the socially appropriate response, and you seem to lack judgment. A perfect example is the question: “What are your weaknesses?” If you are honest, say you’re Churchill, you might reply, “I might be considered an alcoholic by modern standards. I start drinking champagne at 11am, although I never get drunk.” With that answer, there’s no chance in hell that you’ll get hired. Which doesn’t mean that Churchill wasn’t great at his job. But Churchill was a perfect diplomat in every situation, even when tipsy. To demonstrate this, you would answer that question with something more appropriate, such as, “I can be a perfectionist.”

4. References. I once had a discussion about this with a good friend of mine who is brilliant, honest, and powerful, and thinks that most of his employees do a bad job. He was complaining to me about having just given a good reference to his worst employee. I was a little surprised he’d do this: he’s usually so honest, to everyone. When he’s not honest, there has to be a really, really good reason not to be. And he convinced me of it.

He said that you should never give anyone a bad reference, no matter how terrible they are. Your reference isn’t likely to be very definitive as to whether they get the job or not. It’s just one data point out of many. Sometimes it’s just a formality. So, you never know where that person will end up, and you want them to like you. It’s a small world, and it’s highly likely that one day they’ll end up in some position of power where you need them. So you want to be nice.

I worked for someone who I also doubt would ever give a bad reference for similarly functional reasons. It’s in your interest to have people you know end up in positions of power, because it extends your network.

So the likelihood of a bad reference is slim.

If you know the person you’re asking the reference from very well, or are in certain industries, then he’ll probably tell you the truth. But even this might lead you astray. A friend of mine who knows me very well and is honest just gave a reference for me. But the problem is that we were very close four years ago. Since then, almost everything about me has changed. He meant well, but his reference was very inaccurate of how I am now, in my opinion. A year ago, his description would have been perfect. But it’s very inaccurate now.

Furthermore, the interpersonal interaction between two people is complicated: their working relationship might be difficult, yours might not. It’s easy to be convinced that someone is “difficult” when really they just have a more direct communication style. Or, that someone is a joy to work with, when in your office, that has a more direct communication style, they won’t be tough enough to survive.

Blogs get to the point more directly. A useful blog gives you a textual mainline into someone’s thoughts, over time, about topics both professional and personal, writing in a variety of contexts, writing about a variety of challenges, in different moods, with different motivations. Reading it, you can tell so much more. You can tell almost everything you need to know.

It’s not just employees who make it easier when they blog, it’s the people they’ll be working for. Blogging enables the employees to determine in a very direct way what the managers are like, and what working for them would be like. It’s an observation of their weaknesses and their strengths, and every manager has them. It’s a question of finding the right fit: employees who are strong in the areas you are weak, or simply don’t see your weaknesses as such. It’s about gathering a team that doesn’t have any weaknesses, or building a team that re-interprets a weakness as a strength.

For instance, you can now tell that I think things through copiously, that I’m very aware of the exception, that I’m always watching and judging, that I believe prediction is hard and that real-time experiments are the best way to draw conclusions. You can tell that I occasionally tend towards too-long emails. You can tell I need examples. You know I have “selling” mode, and if I’m not in that mode, then I can express myself in an overly-complicated, complex, and theoretical way. You can tell I only whip out selling mode when it’s absolutely necessary, although I delight in being skilled at it. You can tell that I am skeptical of existing institutions, and I’m always looking for new ways to do things. You can tell that I think honesty is important, and that I’m not afraid of flaws, I’m only afraid of not knowing what they are or what they mean. You can tell that I’m focused on actual indicators and metrics, and more importantly, about asking the right questions to get the right indicators and metrics.