Categories
Politics and Economics Ramblings

Status Reports

A few weeks ago Rands posted a couple of articles (1 and 2) about status reports, those things that people hate writing, hate reading, and rarely tell you anything useful even when you do. In particular, he tries to come up with some ideas about how they can be improved in order to actually contribute to the running of a company instead of just slowing everyone down with paperwork.

A few weeks ago Rands posted a couple of articles (1 and 2) about status reports, those things that people hate writing, hate reading, and rarely tell you anything useful even when you do. In particular, he tries to come up with some ideas about how they can be improved in order to actually contribute to the running of a company instead of just slowing everyone down with paperwork.

I’ve been wanting to comment on these articles for a while, but I’ve had trouble crystallizing my thoughts. I’m still not sure if I can, but if I don’t get this out before the New Year, it’ll just sit in my head forever. It’s not a refutation of anything in particular that Rands said, just a bunch of ramblings that his articles sparked off in me.

To start with, there are two issues:

  1. Why are status reports necessary?
  2. Why are they such a problem?

The answer to these are linked:

  1. Managers need true and accurate information to run the company
  2. Providing true and accurate information–whether it’s good news or bad–is rarely in an underling’s best interests.

In any human organization, whether that’s a family, a company or a country, there is a certain amount of friction generated by self interest and lies, however white and small. Like in mechanical systems, this human friction can be minimized; but just as the second law of Thermodynamics forbids perpetual motion machines, basic human nature means that the whole truth will never make it from one end to the other intact.

A typical modern company is made up of three elements, in varying proportions:

  1. Systems
  2. Processes
  3. People

Those are the ingredients for a beast that eats raw materials and shits finished product (metal into cars, requirements into code, whatever). Status reports are a process. Wikis and blogs are systems. There is only so much in the behaviour and output of a company you can change by tinkering with its processes and systems.

In traditional industry this “so much” can be large, because you’re delivering tangible output from assembly lines (systems) and logistics (processes). Still, there is a limiting factor imposed by the people who have to operate the factory: it’s the workers who implement the directives from management.

In the “knowledge” industry (software houses, financial institutions, etc.) people play a much greater part. Which is a problem, because people are so much more complicated than systems and processes. They get depressed, they affect morale in their departments, they raise awkward questions in meetings, and they need paid every damn month.

Executives hate this, which is why they cream themselves over workflow and knowledge management systems that promise to get the workings of the company out of the heads of their staff, and into easily tweaked databases. These knowledge systems can then be shipped offshore to wherever the labour costs are lowest this month, the original staff can be made redundant, and the executives can jerk off about shareholder value in their annual reports and reward themselves with some healthy stock options.

But surely reducing a company’s reliance upon its people, and increasing its systems load can’t be the only option for affecting overall performance? I don’t think that Rands’s quest for a more systems-based approach to status reports is going to lead to massive redundancies, but I do find it symptomatic of this particular school of thought.

So what’s the alternative? I can only think of one:

  1. Hire people you can trust
  2. Give them a measurable stake in the success of the company

The (big) problem with this approach is that it isn’t scalable. From my experience, it works fine with a company up to about 30 people, but after that it breaks down. First of all, the company starts to get too big for the founders to handle all the recruitment themselves, and secondly, unless the company’s revenues scale with the number of employees (hint: they don’t), the “measurable stake” dwindles to the point where it’s nothing more than a 5% Christmas bonus.

So here we’re back to where Rands started: how do you improve communication in a larger organization, when you’ve had to hire people you don’t even know, let alone trust, and where the only stake an employee has in its success is the continued arrival of his salary every month?

Here is also where Rands ended: have people tell the truth.

Teams represent[ed] by more compelling Status Reports are going to be rewarded by getting their agenda fulfilled. People will talk about these teams and wonder about their success. Soon, we’ll be talking about the products created by these teams and trying to figure out what is the secret of their success… which is simple… they’re just writing down the truth.

Except…they won’t. Human friction, selfishness, and little white lies to cover your ass will get in the way. Better social software (whatever) will result in more innovative ways for staff to hide what it is they’re really doing all day. The content-free status report will be replaced by the content-free daily blog entry. It may be a slight improvement, but only a slight one. If you want to tweak the people of a company, you actually have to tweak the people, not just the systems they work with.

Summing up: damned if I know. But if I ever start my own company, I’m going to try and keep it small and successful, rather than aiming for enormous growth and a fat IPO.