5 ms·

Ask HN: How many (actual) hours do you work a week?

7 points by sam36 9y ago

I've been mostly freelance/1099 for the last 2 years doing full stack dev work. I'm happy with everything but I am struggling to get more than 30 hours a week of billable coding time. My employers are starting to complain about needing more work from me (ie, put in more hours). Problem is I simply don't have any more time to give. I typically get up around 9-10am, take care of a few morning rituals and might get an hour of billable time in before lunch. Typically I'm back after lunch around 2pm, and there I remain in my office until between 11pm-1am (break for dinner of course). So for being in my office for 10+ hours, at the end of the day I'm lucky if I got 6 hours of billable time in. I can only reference my old actual office jobs for comparison. I know at the last place I worked, after all the meetings and facebook browsing time, I really don't think anyone put in more than 20 hours of actual coding time. But there everyone was salary and as long as you were in your seat, you were considered "working" I don't think I slack off. I don't browse social media. I do get caught up sometimes in watching youtube videos on programming topics and conferences or maybe taking care of my many personal development machines. But sometimes the programming tasks I have are not simply trivial, requiring many hours or days of thought, in these times, the youtubing does go up as well as other side tasks. Not really sure how to bill for that, if it took two days of thinking and screwing around the house before I had my "ah ha" moment for a complex issue.. yea I don't know. I'm just really not sure how I can possibly type for 8 hours straight without ever stopping and at the same time write code that works. I do pay for access to a co-working space, but I normally don't go as I assume the 30+ min. trip could be better used. Either way, going there hasn't magically gotten me 8+ billable hours a day. Not really sure how to improve at this point. Thoughts? Edit: Thank you so much for the quick responses. For a little bit more back story, I actually work for a small dev shop that has several contracts. They send me work that is usually already bid on as far as hours and price. Sometimes I have a say in that, and sometimes not. They briefly tried to make me salary, not sure why, I wanted to stay hourly. But then started complaining that they were paying me for more hours than I was working. Meaning taking the number of hours allocated for each ticket/feature that I had completed added up to less hours than what they were paying me for, so then they stuck me back to hourly. Basically since the tickets are hourly, it is very hard to complete 8 one hour tickets in one day (that is an extreme case but you get the point.) There are bigger tasks that are allocated 80+ hours, but still not easy to get those 8 hours of coding.

kylecordes 9y ago
Thinking back to years ago when I did individual freelance work, I would say you are actually doing quite well to get in 30 billable hours. Realistically you probably need to spend 10 to 20 hours a week on all the ancillary tasks (especially doing work that will get you more, better, future clients). So if you want to have a long-term sustainable result, it's probably best to plan on at most 30 billable hours per week. (Of course, people who can disregard or don't have families and other obligations, could plausibly work an enormous amount of hours. But that doesn't seem like a great strategy, there are so many good opportunities to produce a lot of value and receive reasonable pay for reasonable total work.)
smt88 9y ago
I don't know. I've never tracked my hours, despite working contract for most of my career. Based on managing contractors for years now, I think you might be the most honest contractor I've ever heard of. Most of them bill for more hours than they're actually coding. I don't think this is malicious -- it's simply replicating the salaried structure, where goofing off and Facebooking are still part of a work day. If you feel like you're pretty efficient, I think fixed-fee projects are good for both parties. Hourly tracking is pretty miserable. My clients are usually excited that I offer only fixed-fee engagements. If you do continue time-based billing, you should try to switch to weekly or monthly instead of hourly. This allows you to build in a little variability. Somewhat unrelated: someone posted on HN that s/he only codes 4 hours a day at the most, because most people's brains only have a certain amount of cycles to devote to something truly thought-intensive. Many, many people chimed in with their agreement. That's an anecdote, sure, but there's lots of research to suggest that much more than 30 hours of productivity per week is a fantasy.
- sam36 9y ago
  > Most of them bill for more hours than they're actually coding. I pretty much figured that. Problem is how we manage our internal work flow is based on tickets with an amount of hours allocated to each one. Some have 4 or 8, others have 40+. Many times I am given 3 or 4 four hour tickets to tide me over for a couple of days (while some bigger tickets are negotiated), but trying to get those 4 four hour tickets done takes all week (or sometimes more). I've stopped really caring if I overrun the times, but the main issue is these smaller tickets branch out across many different apps, all with their own quirks and there is a 3+ hour ramp up time just to get familiar with what the ticket even wants me to do. Then you end up with the dreaded "Why did it take 8 hours just to add a button to a page?" scenario.

byoung2 9y ago
Reminds me if this video: http://www.dailymotion.com/video/x2gp98t http://www.dailymotion.com/video/x2gp98t
smt88 9y ago
It sounds like the problem is that your org's internal workflow depends on accurately estimating the time things will take. As any experienced coder will tell you (and you've just told me), time estimates for tasks are nonsense. You should never run a team by relying on time estimates because they will always be wrong and people will be unhappy (both clients and employees).
- jklein11 9y ago
  Yeah at the very least you should have gotten a rate hike when they changed the way your hours are calculated. You are now taking on the risk of the extra complexity of the project, which was previously on them. They should have to pay for that.
jklein11 9y ago
It is unfair for them to not allow you to bill for the 3+ hour ramp up time. A couple possible ways I could think to resolve this: 1) You could start pushing back immediately on the issues that you know are unrealistically allocated. Point out that the change is across different systems, or any other amount of complexity. 2) You could start tracking your time on each individual issue. This includes ramp up time, tracking your time for the project, emailing back and forth, or any other overhead that goes into the ticket. At the end of the sprint you could go to the manager with it and say here is how the actual time spent differs from what you told me it would be. 3)Start looking for a client that has a more sane workflow.
- sam36 9y ago
  > It is unfair for them to not allow you to bill for the 3+ hour ramp up time. This is just one case of many (where I loose hours/money), but to elaborate; say I get a ticket for 2 hours, "Just add a delete button for an order item on a form". Sounds sane, and I even agree that it should only take 2 hours. But of course once you get into the code you find some snags. After two hours of digging through the code you realize why the delete button was never added in the first place (was not a simple task). But either way, I just spent 2 hours reading code, and still have another 2-3 hours of coding left to do. Might take a break at that point and collect my thoughts, get something to eat, might get side tracked then and not get back to actually working for another hour or 2. Then code up the feature. So that's 7 hours+ right there. Then I bill 2 or 3 hours for that, but it took all day. Boss is normally understanding of going a little bit over (though not really for going 150% over). But really the end customer that pays for the work is worse to deal with. Half the time if they think something (even if it is a "fix") is going to take longer than a couple of hours, they just won't bother with authorizing the work. Then we end up with a big wart in the code base that we have to work around constantly (which in many cases, refusing to pay for an 8 hour fix has caused 100's of hours of other issues.. but another issue) So I guess I feel guilty about working 7 hours on a 2 hour ticket when I know the end customer will just throw a fit. I guess I should just start billing for those hours in this case. But it feels like it is my fault. If I could just remember the internals of the code base, I wouldn't have had to spend two hours just reading code. Seems reasonable.. but maybe not.