In the last couple of weeks I saw lots of discussions about working long hours and pulling all nighters. For example an interview from Marissa Mayer. Sadly most articles glorified working long hours. Or even worse: declaring that this shows passion and if you don't do it, you do not belong into the startup world.
My view on this is a bit different. And I received many questions and talked with a lot of people after I tweeted the following:
Dear CTOs: if one of your developers has to work long hours, don’t brag about it. It’s a sign that you are failing at your job. Big time.— Bodo Tasche (@bitboxer) August 7, 2016
I think it's best to explain what I mean with a small example.
A good friend of mine studied architecture a couple of years ago. Right from the start she had to do assignments that got rated by her teachers and influenced her score. And sadly after a few weeks she was stuck in a 16h work day (including weekends). So what happend? Group pressure. She got rated in comparison to all other students. And if you do 2 more hours a day your work will get a better rating. And then the others worked 3 more hours. This is how the world works, right?
After two years I asked her again about her workload. She said it was okay now. So did she just get used to the workload? Yes and no. After half a year of assignments she realized that working 16h a day, 7 days, had actually a negative influence on her grades. Reducing it to a healthy schedule did not only help her life in general, but also got her better results.
Sadly most people in the startup scene won't come to that realization so easily. I hear this a lot:
But I think this is mostly because people don't create one startup every week. You just don't have enough test data yourself to compare the results of doing 100h for several months to a more relaxed schedule.
This relaxed schedule doesn't mean that you don't have enough "passion" or "drive" to be successful. Like the quote in this tweet suggests:
Bad idea? Bad execution? Just rub a little death march on it and everything will be right as rain! pic.twitter.com/pP9qEX8f1N— wolf (wild fort) (@ztellman) August 10, 2016
I think it is the opposite. A business is a marathon. Not a 50 meter sprint. You have to be careful what you spent your energy on. The pace you walk in has to be sustainable for you and your health. If you are totally overworked you might not see the real solution to the problem at hand. Getting home, doing some sports, walking in the woods helps to get the brain into a relaxed state. I stopped counting the times where I couldn't find the solution and then went home just to realized on the next morning that the fix was done in 15 minutes. Forcing myself to do an all nighter would have done more harm than good in those situations.
You can do all the >100h work weeks you want and fail. And yes, you could do 40h work weeks and fail, too. But the real reason mostly is not the crazy amount of time you worked on your product. In 90% of the time it's something different. Maybe the timing was wrong or the usability had issues. Fixing those issues usually involves talking to users, stepping back a few meters and looking on the idea from a different angle. Trying to brute force success with long hours can be successful, but rarely is the key ingredient.
In my company we don't do overtime because of all of this above. If I see a commit by a person outside of her regular schedule, I talk to her and try to see what the problem is and how I can help to reduce the stress. This should be one of the main roles of the CTO: getting everyone comfortable enough to be as productive as possible. Overworked humans create errors. Errors that could potentially be costly. You don't want that, right? Right?
There is also another good reason not to force your developers into these crazy all nighters. In Europe they are illegal. For good reasons. If you want to read the details, see here. It sums up with: max 8h per day on average in a 17 week period. If you work in a company within the EU that forces you to do more, talk to a lawyer.