Does a stitch in time always save enough?
January 20, 2011 10:22 AM   Subscribe

What metric can I use to decide whether and when to do will-save-time-in-the-future type tasks?

Does anyone know of a simple tool or formula I can use to help me with this kind of stitch-of-time business? My examples below are IT tasks, but I'm really asking in general.

For instance: I work a fair bit in Flash CS5, but I've never learnt how to use its visual debugger. I think I could learn how to do this in about 90 minutes of watching tutorial videos and experimenting, and I think if I did so I'd have earned those 90 minutes of work back in about a month through saved debug time. So this seems like something I should probably do.

On the other hand, I could learn how to type with a dvorak keyboard. That would take longer to do, and a lot longer to pay itself back (I imagine I'd still be losing time for months. But over a lifetime, I think this would pay for itself too.

My instinct is that I should do the debugger thing ASAP, and the dvorak thing never, but is this instinct correct? And how on earth can I know? I feel like the idea I'm looking for is something to do with some sort of opportunity cost but I can't but my finger on it.
posted by piato to Work & Money (8 answers total) 3 users marked this as a favorite
RoI time. Ideally, the RoI itself would be high and the time would be short. If the RoI is low and the time is long, you don't want to bother. I'm sure you could devise some equation (if someone hasn't already) that figures out if something is worth it. I think logarithmic returns would also fit in here, but not sure.
posted by Brian Puccio at 10:39 AM on January 20, 2011

Seconding ROI (Return on Investment). There's plenty of math over at the Wikipedia link.

It's the same kind of math used to figure out if it makes sense to re-finance your mortgage when interest rates drop.

I used this general method to figure out if it made sense to invest in the tools to change my car's oil myself. (It did.)
posted by Wild_Eep at 10:49 AM on January 20, 2011

Response by poster: Thanks for the help!

Just to clarify - I'm not really looking to work out which time-investment will save me more time over any given time period. I'm trying to work out whether I should do these things (which I think all will save me long term time), or whether I should get on with some work! The tricky factor here, I guess, is that I have have no idea how to calculate to RoI of 'doing some work' - it seems like it should be '1' (one hour spent working is one hour I don't have to work later) but in fact if I did nothing but future-time-saving projects in a month, I'd get fired.
posted by piato at 11:33 AM on January 20, 2011

It seems to be that you would need to do some sort of an ROI on your time doing work as well, then you can compare which activity will produce better returns. If you "invest" one hour of your time into work, your return is what? Some number of dollars per hour. Learning DVORAK will save you some minutes per year (at the cost of some hours per year). If you can work a little more efficiently, you could get more work into each hour, and increase your return on an hour's work (unless you are paid hourly, I guess).

(Very simply, because that's my level) So if your original number was $20 an hour, but working more efficiently with DVORAK you can decrease the number of hours it takes to get a job done, now you are getting $20.05 per hour. Over a year that comes to ($20*40hours*52 weeks = $41,600, compared to $20.05... = $41,704) a time savings of $104, or 5 hours. Will it take you more than 5 hours to learn DVORAK? If yes, don't bother. If no, might be worth it.

Just spitballing here, but I think some variation of ROI is what's needed.
posted by arcticwoman at 12:38 PM on January 20, 2011

There is a school of thought that if there is something that will save you time in the long run, you should always do that thing. Why wouldn't you?

However, you must be sure that it will save you time. What is your typing speed now? Have you learned to properly touch-type on a qwerty keyboard? If not, start there. Don't leap straight to Dvorak.

For one thing, you will often find yourself having to switch back and forth between Dvorak and qwerty, which is terrible. For another thing, unless you work as a transcriptionist of some sort, it's unlikely that your raw typing speed is holding you back, productivity-wise.

In the case of the Flash debugger, there is no down side. It will take you only 90 minutes. You'll get that back in just a month. It's a skill that impacts your ability to do your job well. It is clear that it is a skill you will use. The "use case scenario" is there. The time involved is minimal. Frankly, I don't see why you haven't done it yet.

I suspect - based on just a gut feeling - that you're spending so much time trying to decide what to do and why, that you end up not having time to do anything. Trying to make a decision is an insidious form of procrastination. It masquerades as An Important Thing, but don't be fooled! It's a time thief just like all the rest.

In short: when in doubt, just do the thing.
posted by ErikaB at 12:58 PM on January 20, 2011

However, you must be sure that it will save you time. What is your typing speed now? Have you learned to properly touch-type on a qwerty keyboard? If not, start there. Don't leap straight to Dvorak.
What's the RoI on improving your QWERTY skills only to switch to Dvorak 3 months in?

(Though if you know Dvoark, you really need to know it, it's impossible to rearrange the keys on some laptop keyboards, so you'd have to be able to use it completely blind 100% of the time. Or switch whenever you're not using your primary, Dvorak keyboard.)
There is a school of thought that if there is something that will save you time in the long run, you should always do that thing. Why wouldn't you?
Because the main question is will it actually save at least the time/effort/money I'm putting in and if so, should I do it now.
I'm trying to work out whether I should do these things (which I think all will save me long term time), or whether I should get on with some work!
Don't miss deadlines due to self improvement. If you've got the time (and money) to spare and wouldn't rather spend it watching the tube/surfing the tubes/something else (with or without tubes), then opt for self improvement. Everyone makes that choice every day.
posted by Brian Puccio at 1:33 PM on January 20, 2011

I'm trying to work out whether I should do these things (which I think all will save me long term time), or whether I should get on with some work!

I am not an IT expert, but I think you're groping with two separate issues: (1) how much time should you invest every month (or every day, or every year) in learning new skills and upgrading your human capital; and (2) how do you know which sorts of skills are worth investing time to learn? The answers to this vary quite a bit depending on your work situation (are you a freelancer or salaried? if you're salaried, do you bill every hour or just need to show up for your 40 hours per week?), but there's some broad outlines that probably apply to everyone.

1. How much time should I spend learning new skills? I firmly believe that the answer to this should be more than "zero hours" no matter your job or position, but not so many that your clients or boss gets pissed off at your slow response to billable or assigned work. If you're not sure, you could always start with a commitment to spend one hour per week learning new skills; if that seems like too much (you can't figure out what to study, or that puts you way behind) then scale it back and if it seems like it's working really well you could experiment with scaling it up. This can get a bit tricky if you're salaried and expected to bill for every hour of the day but even if this is the case I'd recommend setting an hour of your lunch time aside once a week to teach yourself new things.

2. How do I know which new skills are worth learning? This one is harder because it's much more job-specific. To start, you can always look at colleagues or higher-ups (if you're employed by a company) or at other successful freelancers in the field (if you're self-employed) to get a sense of what skills differentiate someone who is really good at what they do from you. You can browse publications or blogs in your field to get a sense of what skills are going to be particularly important in the next 5 years. If you have a trusted friend you work with, you could ask them what the most useful thing they use at their job is.

Regardless of how you come up with ideas for what skills to learn, you should be able to start estimating which ones will save you time*** by taking a very clear-eyed look at exactly how much time you're spending on a task that might be helped by a new skill. Typing in Dvorak is an instructive example here. Being completely honest with yourself, how much time--really--do you think you waste because you type too slow? Unless you're a brilliant computer scientist who thinks up algorithms faster than his fingers can type--or you are a serious hunt-and-peck typist--my guess is the answer is "not much time wasted." The biggest drag on your speed is probably the process of understanding the problem, thinking through how to address it, looking up the functions or syntax for something you don't remember how to do offhand, and testing and debugging your code. None of these things will be faster if you improve your WPM by 5%. On the other hand, learning your visual debugger might cut your debugging time in half (TOTALLY MAKING IT UP HERE, I HAVE NO IDEA WHAT A VISUAL DEBUGGER IS) and if you spend 10 percent of your time right now debugging, that's a potential gain of 5% efficiency. If you work 160 hours per month, you've just saved yourself 8 hours--well worth the 90 minutes of learning to use the debugger.

***I suspect from the way you phrased your question that you're a fellow wage-slave rather than a freelancer. One thing I'd caution you against is focusing too much on learning skills that improve your speed in performing your current tasks, rather than learning the skills that will get you promoted. Your employer will no doubt enjoy squeezing more productivity out of you each month if you get faster, and there may be some limited raises down that road, but true advancement comes from learning to do shit that you didn't know before. If you're spending 90 minutes learning to use a visual debugger, I hope you're spending at least a few minutes learning new programming languages or the soft skills necessary to budget and lead projects or some new, useful software that none of your fellow wage-slaves knows how to use.
posted by iminurmefi at 2:51 PM on January 20, 2011 [1 favorite]

If you are working under a deadline of any kind, these timesaving calculations are more like procrastination than any kind of timesaving investment. When in this type of situation, write down the timesaving idea in your "to do someday" list and get back to grinding out the task.

Then, later, schedule some time to work on the problem. Maybe the first step would be to give yourself a little extra time the next time you do the task to document each step you take. So the NEXT time you are brainstorming The Solution, you will have real data to use.

My own experience is that most of these things are a sort of fantasy ("oh, the things I could accomplish instead of doing this task"), when in reality, I would have more than enough time if I just worked more efficiently. I've also noticed that the people that create various timesaver things (scripts, machines) usually aren't the ones who take long lunches or finish work early.

Unless you are on speed, you probably don't think fast enough to really benefit from learning how to type on a non standard keyboard that no device has. Just think about when you are typing- is it the movement of your fingers that is really slowing you down?

(But, I edit as I type. I think of a sentence and then type it. I suppose some people type their thought-stream and then go back and turn it into English. Maybe they would benefit from it.)

On the other hand, the debugger likely is a good investment of time. Adobe (or whoever) took the time to create the tool, it stands to reason it has value. But again, do a pro and con for it: how much time do you spend debugging? What kinds of bugs are they- maybe the 90 minutes would be better spent making a cheat-sheet of syntax, and then you wouldn't have any bugs to begin with. For example.

My own philosophy with this is that every "thing" we own (even software things) steals time from us in the form of maintenance. I have a script that logs onto a VPN for me. It has the appearance of saving time, but it is another thing that I have to update when I change my password. That eats into the time saved.
posted by gjc at 3:59 PM on January 20, 2011

« Older It's our responsibility to authoritatively...   |   Abracadabra! I give you something that you'll... Newer »
This thread is closed to new comments.