« Announcing Edison 2010-12: Privately-shared experiments, Facebook Connect, and more! | Main | "Love your experiments as you would an ugly child." »

Discuss: The Quantified Worker

[cross-posted from Quantified Self]

... big computers!

While much of our work here is focused on individual development, there are plenty of circumstances in our professional lives where we can apply the ideas of experimentation. Let me set the stage with some background and ideas, and then I'd love to hear from you on how you widen self-tracking to apply to your occupation.

First, experimentation at work is not new. Frederick Taylor's Scientific management popularized applying metrics to factory worker performance in the late 1800s. Later came W. Edwards Deming, who influenced the Japanese Lean manufacturing movement in the 50s, which integrated experimentation, measurement, and continuous improvement. A more contemporary thinker is Thomas Davenport and his ideas on How to Design Smart Business Experiments (an excerpt of a paid article).

A natural starting application for self-experimentation at work is at the individual level, such as by trying out new tools or methods, either in self-management (e.g., productivity systems like Getting Things Done) or by exploring work-specific ideas and techniques. Here at the Quantified self we've had nice discussions on time management, including:

Of course self-tracking tools abound, like the oft-mentioned RescueTime (see Kevin's "Productivity" Dashboard Monitor for a bit more). Putting the Hawthorne Effect and placebo effects aside, the value in all cases is getting insight into ways you could improve how you work and then implementing them. (You can find ideas for the latter in my post Add, subtract, multiply, divide: Productivity lessons from basic math.)

Beyond the personal level, experimenting at work is baked into some domains, such as industrial design (e.g., the work of IDEO's Tom Kelley), innovation (e.g., Google's approach), marketing (e.g., SEO or Test and learn), entrepreneurship (e.g., Lean Startup), and software development (e.g., Agile methodologies). I wonder what we could learn from studying them?

Given all that, I'm curious: How have you applied experimenting and data tracking to your job? How has it worked out? What tips can you offer for making experimentation effective?

Reader Comments (2)

I havent really experimented per se at work. But rather constantly tried different ideas until stuff started to work. I think of experimenting as a more formal sort of activity where you have ground rules going in, such as "I have to try this thing for 30 days" Or I have to be as neutral as possible with respect to outcomes.."

Anyhow, so much of what I do is on the fly that I really dont have formal rules to go by. I often re think what I did and then make obvious improvements. then stuff starts to work and I stay with it, although I have no baseline to compare it to, to see if it is really working better than other possible methods.

One practical problem that I keep coming back to is in, I think it's GTD where Allen says that if you can finish some immediate problem in 2 minutes or less than you should address that now. Many of us do stuff in this manner, picking up dishes or papers or stuff as we move through a house or office. But: what if you cannot know how much time a given task will take, then how do you approach it?

I was just discussing this today with an attorney friend of mine and he was talking about the oil refinery invention he had to file the patent on . He wasnt sure how much to charge but took something close to industry average that he had googled. It turned out that the invention was just super difficult to envision and the only diagrams the inventor sent him were drawings of parts but DISASSEMBLED. then the inventor described how to assemble the parts, which is extremely difficult to envision how they fit. Let alone try to understand how they operate...

As my friend said, he stared at the drawings for about 2 weeks, starting to write and stopping again when he had gone down the wrong alley. I told him the problem is that unlike inventors we as attorneys often dont have strong visualization skills like these guys, and you have to just sit there and try to figure it out and you have no way to know when your brain will finally click in and you get it...

My favorite was the pecan breaker, it had all sorts of cams and levers and all of these parts moved in variable distances in order to adjust for certain conditions. they interacted with one another. I started at those drawings every day for about a week until stuff finally started to make sense.

But there's no way to tell before you start just how long this process will take. And hence the problem with quoting a legal fee for all that.

Any ways to deal with this?
December 14, 2010 | Unregistered Commenterjp in maryland
Great questions, JP!

> constantly tried different ideas until stuff started to work

That's a fine definition, in Experiment-Driven Life terms. I think we can still find value in this approach without necessarily having to apply the more rigorous definition you talk about.

> I often re think what I did and then make obvious improvements...

I'm hip - sounds like the Think, Try, Learn cycle to me.

> although I have no baseline to compare it to

Then it becomes anecdotal. But again, the goal is to improve, and if your gut tells you it's better - and you don't want more evidence than that - then go for it; I won't criticize.

> what if you cannot know how much time a given task will take, then how do you approach it?

I wrote about this in my "Where the !@#% did my day go?" ebook (http://www.matthewcornell.org/consulting-products/). The key is to hone your estimating skills by starting to track and compare estimated time to actual time. I use the daily plan as a platform for developing this talent, since you're already writing down the tasks you do anyway. (This idea is popular in Extreme Programming, BTW, but not so much in personal productivity.) But as your legal story points out, when something is *really* new then you have no baseline for comparison, and you have to do your best, even if it's unpredictable. In fact, that's an excellent characteristic to define "experiment." I.e., the test is: "Do you have anything to compare this to?" I like it.

I don't really have a solution in mind, other than the general time management (and software) guideline of estimating, then doubling it. I.e., be conservative.

Great comment, JP. You and I might be the only people around the blog right now :-) Thanks for sharing.
December 14, 2010 | Registered CommenterMatthew Cornell

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
All HTML will be escaped. Hyperlinks will be created for URLs automatically.