I’m trying to learn from him. He has a great sense of timeline whereby he gets people to take good breaks while making sure things get done on time.

Your clients and your bosses will take whatever you’re willing to give to them. Nobody will tell you you’re working too many hours.

I think that I may be more sensitive to lack of sleep than a lot of people, as even minor things like that can leave me feeling groggy the next day.

My problem is when you’ve just got all of these ideas in your head and you can’t sleep. You spend so much time telling your brain to shut the hell up. These are the times that I have to decide… do I a) lie in bed and hope I fall asleep, b) get up and start working since I’m awake anyway, or c) go lie on the couch and watch tivo’d Extreme home makeover and hope the mindless entertainment puts you to sleep. last night I chose c. Sometimes I choose a and sometimes b.

I decided something had to change. I sat down and made out a schedule and started following it. Having everything mapped out with the amount of time I wanted to spend on each project helped take the pressure off and I actually got more done than I would have had I tried to cram it all into one night. I’m able to get up at a reasonable hour (if 5.30 can be deemed reasonable) and call it quits in the early afternoon.

So far it’s working, but like they say, “One day at a time.”

First, excellent article. I love that you’re debunking this completely irrational myth. So, my rebuttal is as follows:

1. The difference between a creative all-nighter and a programming all-nighter: Creative all-nighter has the potential to turn to design gold… your mind opens as the night goes on and you can find yourself unencumbered by distractions and traditional thought processes when you’re all alone. A programming or any other all-nighter often dissolves and unravels as the night goes on… you actually become far less productive trying to complete otherwise mundane tasks (writing CSS, an essay, etc) than if you were working during the day, fresh and awake.

2. My little brother once did a research study on sleep and discovered this little tidbit: it takes, on average, 2-3 days of full sleep to recover from 1 night of 0-4 hours of sleep. Yup, that’s right – for the all nighter you just pulled there is a quantifiable loss in productivity. This is established fact and should be carved into the wall of project managers: to maintain a consistently productive team, each team member needs to balance his work sessions with a healthy night of sleep AND non-work -related personal time. What science tell us is that cracking the whip is often effective in short term spurts, but results in inconsistent and low quality work over the long haul.

3. The lesson: If you’re a creative (like me), all-nighters can be effective if you use them right and understand the cost. More often than not though, the all nighters that I pulled this year were huge creative pushes that truly did compromise my rational abilities for the next couple of days. The one successful all-nighter that I pulled this year(I’m thinking of one project in particular where all 3 of us worked until 4am) actually did pay off and it was well planned with the understanding that we would decompress for a week afterwards. I liken this to video games where you can use a “hyper mode” for 10 seconds with the implication that your character is completely vulnerable for 30 seconds thereafter.

4. Common sense: All-nighters are not only usually unproductive, but are often completely avoidable if you have a good project manager and a good grasp of the actual time it’ll take to complete a project.

The theory of the all-nighter shouldn’t be grounded in “the last minute push to finish a project”. It should be based on a justifiable need for 8-10 hours of pure silence… which, when you think about it, can be accomplished in a variety of other methods (unplug your phone, ignore your email, tell your wife to catch a movie and hit the mall).

I’ve adjusted (and I’ll blog on this as soon as school’s done) my work habits to revolve around a new habit: the work-sleep-family 3 pronged zen approach. More on this soon enough

Thanks for an enlightening article Shane!

1 – “This is so cool I forgot about sleeping” all-nighter. I had this when I was working on the Rails Rumble 2007 programming contest. It is where you are driven to work because it is fun.

2 – “This has to get done” all-nighter. This is the common all-nighter we all know about. I have found that I actully do get more work done in that time but it is of lesser quality and is no where near as maintainable as my standard work. In a few cases I even had to throw away all the work I did.

I think the main thing to keep in mind is to work at a sustainable pace. Not so fast you burnout but not too slow that the project gets backed up requiring an all-nighter or feature cut.

The C2 wiki has a lot of classic information about this:

http://c2.com/cgi/wiki?FortyHourWeek
http://c2.com/cgi/wiki?EightHourBurn
http://c2.com/cgi/wiki?SustainablePace

People stumble because they’re not engaged, and often they’re not engaged because they’re in the wrong job (or role.)

Tim Ferris’ book “The four hour work week” has some nuggets of wisdom. One of them is when he recommends that we all set timers and each time they go off, we ask ourselves: “Are we doing real work or just screwing around and fooling ourselves that we’re working?”

I have done this, and while not perfect, it’s helped me course correct.