[update: to all those who got here by clicking a link entitled "why i will never feel threatened by programmers in india", that title has been retired as an unfortunate misfire on my part: much of the conversation was lead astray as though i meant to single out india in particular, and it landed for some as having racist connotations.
it is now titled "why i will never feel threatened by cheap overseas programming", which much more aptly captures the meat and message of this essay. and now without further ado...]
i got a call from a friend of a friend the other night. it was a fellow with whom i’d talked 11 months ago about a project he and his partner were looking to start. we established then that i wasn’t the guy for him, that i was likely too expensive for their big-dreams, small-means budget.
fast forward to present day: their project is still not launched, it’s still not right. they’ve paid for something between 600-700 hours of development with a firm in india, and they should have launched 6 weeks ago.
sure they’re only being charged $14/hour for that work, but i think the indian firm is, as the saying goes, making up for it in volume. and that’s to say nothing of the time our heroes have spent proctoring the whole process: in his words, they’ve got to be constantly super explicit in their instructions, or it comes back wrong–then then have to spend more hours getting it fixed (this is but one instance of why i am suspect of the hourly model in my industry).
“i’m wondering if you’re available–my partner says we just need an american programmer to get in their and clean up a few things to get us out the door, we figure it would take the right person 10 hours, instead of 50 or more with these guys.”outsourcing programming came in to vogue as a cost-effective strategy, what, 10 years ago?
for a number of years in my industry there’s been a sense of not-necessarily impending doom, but something akin to that due to the rise in outsourcing programming work abroad, most prominently to india. hoards of smart people, as the alarmist vision goes, willing and able to do highly talented technical work for a fraction of the going rate for us programmers. the labor force here in the states can’t possibly compete, and what savvy, responsible company could possibly pass up the opportunity of such cost-effective globalization?
the thing is: i have yet to see a project done overseas at that sort of hourly rate that has actually gone well. version 1 of monstermarketplace was first done overseas–took 4 months, billed for gobs of hours including a lot of testing and qa, and crashed predictably under the strain of campaign traffic. (by contrast, my complete rewrite of the front end was done in 4 weeks–at peak times it produced about 4% server cpu utilization, never crashed, and had correspondingly faster page loads.) this was my first inkling that the “outsourcing problem” wasn’t all it was cracked up to be.
my second close look at the outsource phenomenon was the code base for zowzee. as i discussed in the tale of building and launching spotlightdenver.com, rebuilding it made more sense than trying to extend the $12/hour piece of work (to this day i’ve been too polite to ask how many of those $12 hours it took–i do know that my work was vastly cheaper, faster, and smoother).
and now this. that makes three out of three instances in which outsourcing abroad turned not that great, if not outright regrettable1. this is a small sample, to be sure, but still serves to bolster my confidence as a us-based programmer.reclaiming our competitiveness
so how is it that we can compete with programming talent overseas, with figures like $12/hour stacked against $125/hour? my experience leads me to believe that us-based programmers can be the sound and “right” choice for development projects, and we don’t even need to resort to protectionist arguments for supporting our local economy & talent.
i think we can observe a few generalizations which, while they may not give due credit to the exceptional firms overseas, serve to counter the equally general notions that hiring offshore talent is an economical no-brainer.
consider the idea that the rock-bottom hourly rates creates a certain cognitive dissonance in decision makers. if i tell you that your job is going to take 500 hours, but don’t worry, it’s only $14/hour, you might think it a god send that such a rate is even possible–you might even think it not worthwhile it to consider us based options at $50+ per hour. sure, the quote from india might be a little inflated, but it couldn’t be that inflated… right? turns out it was in my 3 examples.
but even if multiplying hours times rate gives you a real bargain, there’s a very good chance the hours term will balloon. if they get something wrong, they can fix it with 20 more hours of work. but don’t worry, it’s only $14/hour. again, impaired judgment based on what a great deal you’re getting makes this palatable and thus probable: you wouldn’t tolerate it if you were paying a substantial hourly, and besides, you’re already in 500 hours, so what’s a few more? i think it’s safe to presume that overseas firms understand this reasoning and mindset very well, and at very least appear to exploit it accordingly.
our heroes’ need to be super explicit in order to get what they want reveals another major advantage that local programmers have: nobody wants to have to babysit their programmer with constant direction and correction. communicate the overall vision of what you’re trying to create for your customers, and any programmer worth their salt will bring their a-game to solve it from that shared understanding. the net result is a project that is completed faster, racks up far fewer billable hours, and saves you headaches and time.
i’ve noticed a number of other reasons to think twice about off-shoring:
these all come with the turf but seem seldom considered: the allure of cheap labor seems to trump these potential problems, or they are unimagined all together.
the good news is that each of these reasons constitute a means by which us programmers can justify their value as a worthwhile and economically sound alternative to outsourcing.
there’s fruit to all of this on more idyllic grounds, too: just as the threat of overseas outsourcing would tend to discourage mastery of programming here in the us (what’s the point when you’re told that someone else is going to eat your lunch?), the notion that the fight is more fair than typically envisioned may as well encourage it.
i believe when the myth of cheap, abundant programming abroad is shattered, there is a lot more reason for the next generation of talent to take on and excel at a craft that creates real value.
[update: there has been a lot of comments and conversation on this topic here and elsewhere. see the response to many of the consistently raised issues in the follow up post, cheap overseas programming revisited.]