This post in particular echoes with me, as I am coming of age as a fledgling manager at a company struggling to establish functional org charts. As you aptly described, I’ve found a large part of my managerial role involves spending most of my working time in meetings. I have a broader definition for meetings than others I work with, as I consider equally (if not more) important the offhand comments made in the hallway, the discussions and brainstorming at the common lunch table, and the formal discussions with accompanying minutes that take place around a table in a closed conference room.

Outside work, I love to talk. At work, I favor quiet listening and observation. I seem to filter conversations for bits of strain, curiosity, praise, need, ideas, goals, how-to and “wouldn’t it be great if…”s. I have my particular way of mentally organizing, cataloging and maintaining the information I hear, allowing me to remember the cumulative history of filtered bits from most of the conversations I hear.

My excitement comes when I hear in a new conversation a bit that matches with something I’ve heard before; something in my memory. At this point, I actually speak. My favorite contribution to the managerial gluepot is to link people who are searching for an answer with others who may have the answer; or more frequently, are searching for the same answer. Sometimes, the simple process of uniting two people who are in search of the same unknown is the only catalyst required to start the chain reaction leading to new discovery.

Then the management changed, and was replaced by a paranoid, insular group of new people. They didn’t try to get to know the people who worked for them, or the processes that had been in place before they were put in charge. Instead, they formed a little clique who issued dictates that were obeyed, or else. Bad in every way, and you could literally watch the team fall apart before your eyes, as people couldn’t abandon the sinking ship fast enough.

On this project, the management simply doesn’t care about team coherency. They vaguely watch the schedule, don’t tell people what’s happening, and don’t foster any sort of team atmosphere. What’s weird about this is that the motivated people get to do good work, because they’re not being interfered with, but when things go wrong, they go wrong for a very long period of time without being corrected. There’s almost no communication, organization, or sense that we’re all even really working on the same thing.

It’s quite weird.

I mean, in the end what this really just means is, “I only really like working with good bosses,” which seems pretty reasoanble a conclusion, but it’s somewhat odd that I’ve worked with three completely, completely different groups in less than a year. The first, good group, illustrates the flaws in the others so clearly that it’s sadly thrown into stark relief how hard it is to find good leadership for a company.

You see… I’ve pretty much been put in the position of “the glue.” The walking-around-and-talking-to-people part is great. But just as glue works nicely to pull things together, it gets the crap beat out of it when things want to fall apart.

As “the glue,” the point of the fictional org chart vs the reality is a good one that I hadn’t quite formed into coherent thoughts. There are definitely people within the company structure that wield power simply because they make things happen, and there are those who are given power, but are unable to wield it due to the fact that no one follows or trusts them.

There is a definite need for structure; if there is no org structure on paper, information starts to fly every which way, and people start to go off and do their own thing. The paper chart gives them a guideline, and the individuals within the structure then start to make shortcuts on their own…

I’ll go through and read more as it sounds like it’ll help me wrap my head around my job.

The office politics are always a problem. If somebody is hired as a lead, but is much better suited to being a developer in the trenches, then do you reduce their salary to match the other developers?

But I agree with you. I’ve always thought that it’s a manager’s responsibility to put people in the position in which they are most likely to succeed, where their strengths are maximized and their weaknesses minimized. Sometimes you are resource limited and don’t have a choice about who gets what tasks.

The Google guys had a great quote about this: “You can try to control people, or you can try to have a system that represents reality. I find that knowing what’s really happening is more important than trying to control people.”

I think one part where I slightly disagree with you is that the fiction of the org chart is still a useful artifact as a starting point. If you just start off everybody with no structure, then chaos ensues. An org chart imposes order. But it shouldn’t be static. Imagine if you rapid prototyped the org chart as well. You throw one out as a starting point, you see what works in it, you refine the org chart, etc. This ignores how many people don’t feel comfortable in a fluid environment, and want to just know their roles. But it’s an interesting idea.

One place where the org chart is rapidly revised to fit reality is when the organization is in trouble. I’ve had two friends who ascended from engineer to VP in no time at all, because they were at startups which were floundering, and they showed they could get things done. In more stable organizations, that would never have happened – the upper management would have been content with the fictional representation. I’m not sure if that means we should have our companies always be run as if they were in trouble.

Gah, the longer comment box just makes me want to write forever. I’ll stop now.

But I didn’t put it nearly as succinctly as this. But here’s where my head gets wobbly:

You say you want to be the social glue. That’s good – I think you’d be good social glue.

I think, though, what I got out of the “reality of the org chart vs. the fiction of the org chart,” is simply that the essential structure of a system simply emerges. That, at least, has been my view of the companies I worked in in the past.

So, rather than assigning roles, and having people fill those roles, how do you make the fiction fit the reality, instead of the other way around? Blah. I’m not making this clear. The text box I’m entering the comment in is too small, incidentally, and I’m almost having a hard time “cohering” my thoughts, because I can’t see what they were as I’m writing. Nonsequitur, I know.

Gah!

Ok, so, what I’m trying to get at is, is there a way to simply have an org chart be flexible enough to reflect reality, instead of dictate what reality should be?

That is, let’s say you have a project, wiht a couple designers, and someone who’s been assigned to be the lead. And let’s say that everyone knows that lead isn’t the right person for the job, for whatever reason. So, the org chart says Person X is the lead designer. But then let’s say *in reality* the bulk of the decisions fell on the non-lead roles, and people would approach other designers to answer questions that should have been directed to the lead, but weren’t because the people know the reality better than the org chart.

Is there a way to simply dynamically redraw the chart to reflect reality, and move the now-ex-lead into a role that better optimizes their utility?

I know office politics are always a really huge factor in things like this, but just throw out the conventional wisdom for a while – what sorts of operational gains could you make by understanding the *reality,* as judged by the actual actions of the actors, rather than the intention of the group imposing the structure?

Is such a thing desirable? Possible? Because clearly, the structure already occurs, and we just ignore it because it’s not polite to not abide by the org chart. So, certain people are radically undervalued, while others radically overvalued, creating tension, strain, blah blah blah. If the actors were safe from large negative repercussion, what kinds of benefits could you get from a system that was actually *honest*?