Not in Argentina, at least. Most people there get two weeks off and these are generally taken, together, always during the same month (January to February, which is, weather-wise, their equivalent of August). This is the pattern by custom and by law there.

Lesson: If you are going to work with people in a foreign culture, inquire about their vacation culture. Do people usually take their vacations altogether, at once, or broken into bits? At the same time of the year, or whenever they want? Plan your development and work schedules around these.

Here are two fun examples. First, "excited." It's a very common word for any English speaker -- or at least we Americans, who seem to get excited about everything. But in Spanish, "excited" means the same thing it does in English--but with a very strong sexual connotation. I always say "enthusiastic" instead, or, "I'm really excited about this (in the American sense of the word)."

Secondly, "discuss." In Spanish, it means almost the same thing as in English -- but it implies "argue" or "fight" (unlike in English, where it implies only a normal conversation). I make a point of telling everyone, "let's talk about" an issue, rather than, "let's discuss it."

Okay, I just thought that the association of outsourcing with torture was funny!

Interesting: the argument against outsourcing is often, that it move jobs from our country to their country. But an interesting reversal has happened: their governments tend to be so corrupt, that the governments don't trust themselves with money, so they outsource managing the money to us!

The beauty of unintended consequences.

It makes sense to me to follow a few principles:

Principle #1: Keep responsibility as close to the person actually doing the work as possible. The practical implication of this, for me, is that as much responsibility should be with the software developers and graphic designers as possible. The challenge is that the work is multifaceted, with client face-time being fundamental as well.

Principle #2: Keep the lines of responsibility clear. Working internationally (cross-language; cross time-zone; cross-culture; unable to read non-verbal signals like body language) is tricky and it has to be clear who is doing what and whom they are reporting to, else confusion and ensue

Principle #3: Actively solicit interactive communication. It is not enough to say, "talk to me if you need help"; pro-activity is important because, human nature tells us, we don't volunteer as quickly as we should.

Principle #4: Context, context, context: Add in additional context for each request, even when not obviously necessary. This can easily become a time-drain (watch out!) but, when communicating with those far away, the reasoning behind requests may not be clear, and everyone understanding the objectives of each little thing makes the implementation much easier.

First, there is Lingo, a VoIP service. For $80/month, I can make an unlimited number of calls to Argentina, using my normal telephone. This is a brilliant User Interface: instead of having to install or learn new software or connect a headset and microphone to the computer, I use the UI I know very well, the standard telephone. Furthermore, the all-you-can-eat model is great for this purpose: if I used a cheaper in which I only pay a few cents per minute (say, Skype?), then I will be counting the minutes and minizing the interaction. But now, my incentive structure is so that, on a moment's notice, I call Argentina constantly and work out every issue over the phone. Very quick, very efficient, and reinforces the truth we know but easily forget, the hierarchy of interactive technologies in terms of the efficiency they promote in terms of meetings and dealing with important issues, with face-to-face always on top, phone below that, and then online technologies only when the other two aren't feasible.

The second technology is the Blackberry, which my friends have been referring to as a Crackberry. I'm addicted. The havoc it has caused for my social life aside--yes, that's essential to the analysis of it but we'll do that later--it has helped me solve a critical hurdle in my outsourcing business. One of the key challenges of outsourcing is the increased time everything takes because of the increased communication between all parties. Therefore, with much back-and-forth between the clients, the American team, and the Argentine team, there would be a time lag if, say, I was in meetings all day long and only at night saw the relevant emails to forward or ask a question to someone as a result. This would happen a lot. The Blackberry has solved this entire class of problems. Now, I easily and quickly forward anything that comes in to the appropriate person. It is not a coincidence that everyone on all sides has been happier these last few weeks. Except for my friends.

Like all human beings, professional service firms could be a great deal healthier if they eliminated their bad habits. This chapter focuses on a single bad habit that reduces profitability, adversely affects motivation and morale, reduces a firm's competitive capabilities, and in addition prevents senior professionals from spending more time with clients and investing in the future of the firm.

This bad habit is called systemic underdelegation.

Imagine that a questionnaire was sent to each and every professional in your firm, top-to-bottom, asking the following single question:

What percentage of your professional work time is spent doing thigns that a more junior person could do, if we got organize and trained the junior to handle it with quality?

(Obviously, do not include in this calculation that work which the client insists you perform, since the client must get what he asks for.)

Imagine that each person answered honestly, and the responses were tabulated to calculate the firmwide average.

My research shows that, for the typical professional service firm, the firmwide averge is frequently as high as 40 to 50 percent, and sometimes more. This is equivalent to saying that, of the firm's entire productive capacity, 40 to 50 percent is consumed with a higher priced person performing a lower-value task. Obviously, this is not a wonderful situation. [p. 41-42]

In Argentina: In addition to the government fees, you need to put 12,000 pesos (about us$4,000) in the bank just for this - in a country where that's something like the average man's annual income!

There are technological solutions (you can have them check in code regularly and review the code throughly). There are business process solutions (daily email reports, workflow checks on what everyone is doing).

The one that works the best, however, is trust. One of my employees reminded me of that just this morning: nothing compensates for really trusting deeply the people you work with, to do even the tiniest things well and, importantly, on-time.

The challenge of outsourcing is, of course, finding people you trust. And it's hard to do that.

DP uses a few processes to help us find people we trust. One of them is our insistence on "test projects" before we hire anyone, so we can get a sense of the quality of the programmer's work and his discipline as an employee. We have other methods, too, and they might be corporate secrets, or maybe I'll reveal them in future blog entries.

One of the basic challenges is, what happens when someone slips through, and you hire someone but start to have problems? Here are some strategies that are worth articulating at the very least (and I will examine later):

• The Jack Welch strategy is to make sure everyone always knows they're replaceable
• Always use pair programming and other sorts of redundancy, so that people can compensate for each other
• Never reward bad behavior: Be willing to fire people when necessary

More to come later...

Your defense of outsourcing was pretty compelling, in all seriousness, and that's what piqued my interest. I'm more of a bleeding heart, but with a global rather than national bent, and I like the idea of design/programming jobs being available in developing countries. And if it benefits you, too, then all the better. A friend of mine was drawing a comparison between the argument against outsourcing and the argument against hiring non-whites in the States before the civil rights movement. White people being like, "Hey! Those are our jobs! They're getting taken by minorities and immigrants who'll work for less! No fair!" That sense of entitlement irks me.

The second realization was this: If she didn't clean my apartment, then I wouldn't've done it (in anything more than quick, trivial sorts of ways). That is to say, without her, my apartment would not have gotten the real, thorough cleaning it deserves.

This is an important realization because this blog is dedicated to discussing how efficient--or inefficient--doing something outsourced is compared to doing it in-house. But Graciela added a key fact to the mixture: that outsoucing doesn't just make it more efficient, it makes it more likely that it will happen at all.

Business analogs to this are common enough: I want to improve the User Interface for ProductX. I could find time (in six weeks!) to do it myself, do it, review it, show it around, and then maybe eventually I'll implement it. But, when I make the decision to do it, instead of scheduling time for myself to do it, if I hire someone today to do it, then it is much more likely to happen at all.

Outsourcing, then, doesn't just make things more efficient; it makes things happen that otherwise wouldn't.

This difference, then, is not just the difference between two con man movies, but a summary of the difference between the American and Argentine cultures. In Argentina, it seems, everyone is always trying to screw everyone else -- and when you find someone who isn't, then you are pleasantly surprised. But in America, it seems, no one is trying to screw anyone -- and when you find someone who is, you are disappointed.

The percentages of people conning others may end up being the same with both countries--I don't know the numbers--but these difference in expectations give very different feelings to interacting with people and having relationships of all sorts in both countries.

This insight applies to outsoucing there as well: I would be lying if, working in Argentina, you don't get the feeling that everyone is trying to take advantage of everyone else. At the very least, even if you don't see it much, everyone says it so much, and with such insistence, that you can't help but feel it. (Similar to the pattern of talking about depression making you more depressed!). It is scary, and it is a risk.

How is this risk minimized? One of my methods is to be very explicit about the professionality of the enterprise. That is to say, in a land of many crooks, if you make your honestly very explicit, then there are two results. One is that you will be repected and the other like-minded honest people will want to come on board with you. (This is the bet I'm making.) The other possible outcome is that lots of people will try to take advantage of me: and, being very conscious of this, I fight this by both constantly being on the lookout and using my standard safeguards, and by always advertising our strength and willingness to stand up for what we believe in: no one wants to rob the strong who he knows will fight back, instead they target the weak.

First, my plane ticket to/from Buenos Aires in May. The couple of weeks spent working there, getting to know everyone in person, was invaluable. This reinforced our community and dedication, allowed everyone to see that the other sides of the company are "for real," and we had various long meetings--planning projects, discussing the company--that would have been impossible over the phone. Hiring people near you is more expensive because, well, they are near you.

But I can't fly to Buenos Aires weekly. So this is why the second most useful investment I've made in the company is buying a Lingo box and subscription. For US$80/month, I can call about 35 countries, including Argentina, for zero marginal cost--ie, for free for each call. I've been speaking to Argentina for 2.5 hours every day since I began the service a few weeks ago, not to mention the extra calls to my friends in Canadia, Mexico, Chile, and elsewhere I've been making.

On the cost: yes, there might be a cheaper service, possibly even Skype. But what I like about Lingo is the ease-of-use--I don't need to connect microphones to my computer nor dial long special codes before each call--rather I use my standard telephone, just plugged into their device. I also like it for the same reason I buy an unlimited metrocard every month, even if I might save a bit of money buying 10-packs: I know I watch every penny, so if I am paying for each call, then, before each call I will ask myself if I really need to make it, I'll be watching the clock every minute, and so forth. But knowing that I've already committed to the sunk cost of the monthly fee, I will comfortably take advantage of it and talk without worrying about the cost--and this is precisely the incentive structure I need to ensure that I speak to everyone in Argentina lots. Not to mention, for just a few dollars more, I can get a local Buenos Aires phone number and have it routed to me!

Talking on the phone with everyone in Argentina these last few weeks has been wonderful. It is the next best option to being there in person. The downside is that I've become the psychiatrist to more than one person in Argentina; the upside is that speaking on the phone is about 5 times more efficient than instant messaging (our usual standard form of communication)--so I end up saving a significant and measurable amount of time. We've also been avoiding misunderstandings and other problems that previously were too easy to happen when you chat via the computer. Everyone in the company feels the improvement--except for one graphic designer, who doesn't enjoy speaking on the phone that much. Oh well.

Overall, Lingo: two thumbs up!

One thing that it makes sense not to outsource is your core competency.

In order to be successful, you need strong and direct control over your core competency. A company that invests money, that outsources the investment decisions, isn't left with much else. Because they aren't investing their own resources in making the best investment decisions themselves, they are likely to be worse off.

Outsourcing, by definition, gives you less direct control over what is happening than insourcing. Since you know what you want, this results in extra time needing to be spent on the project. And that's fine--but, if it's your core competency that is being delayed, then that's a problem for the company.

Maybe, then a general rule of thumb is that, the further anything is from the heart of what you're doing, the more sense it makes to outsource it to the specialists whose core competency it is to do that thing.