Ben Buckman.net

I've been reading Tim Ferriss's book "The 4 Hour Work Week", about a "New Rich" system for cutting out the 80% of your time that produces 20% of the value (in a nutshell). It's sort of a practical counterpart to the inspiring-but-useless book "Rich Dad, Poor Dad".

One of the chapters is about "Outsourcing your life." He talks about outsourcing ordinary tasks to "virtual personal assistants" in India, outsourcing production of specialized products, outsourcing pretty much anything. (On the macro scale, I'm fairly convinced that in a global economy where labor races to the cheapest providers, American workers are fucked, but I'll leave the politics of this issue to other posts.)

Anyway, I read the chapter and thought about how I might be able to apply the concepts myself. Most of the actual work I do is highly specialized, and I subcontract parts of it (so far) only to people I already know and trust to do great work. But I did have a list of other stuff I needed done that I haven't had time to do, such as:

1. Research car insurance policies. (I've been burned several times by Liberty Mutual and need a new provider.)
2. Research some growth affecting our tomato and spinach plants.
3. Researching and/or scheduling various appointments.

Coincidentally, while thinking about this, I was reading fellow Drupaler Chris Shattuck's blog, and he had posted a list of "Stuff I own and like a lot," including 4 Hour Work Week. So I asked him about the outsourcing chapter. That prompted him to try an ongoing experiment in outsourcing his own life.

I've been reading his posts with interest, and trying the same on my own. I posted a job on eLance (a global freelancing/outsourcing connector) for a "Virtual Personal Assistant." In the job description I specifically mentioned the car insurance and plant research.

I got half a dozen responses, with hourly bids ranging from $5 (in India) to $10 somewhere in the US. I rejected the $10 bid as too high, and rejected the $5 bid because it was filled with generic gibberish and typos. I narrowed the rest down to 2, a woman at $7/hr that seemed good for the insurance research, and a student in MA on summer break with a bio background for $8.50/hr that seemed good for the plants.

I asked the $7 person for an estimate of how long it would take to get 5 competing policies and got a quick response (1-1.5 hours). So I hired ("selected") her for the insurance research: I took the crib sheet for my insurance policy, blacked out personal details (last names, ID numbers, address) - leaving just the car info, ages, and coverage amounts - and asked for all the details she could get and a summary of pros/cons, with some guidance like "if you're on hold more than 5 minutes, hang up, I don't want to work with companies that have bad customer service."

I'll select the other guy for the plant research in a little while and send him photos I took of the plants. If it takes him 1.5 hours as well, all this will cost:

• $15 for the premium job listing
• $10.5 for 5 insurance policy bids
• $12.75 for a prognosis of my plants' affliction

(Total $38.25)

Plus the time involved, a few hours probably to sign up on eLance, post the job, read the bids, send back details, and read the final reports... 4 hours maybe. (The biggest cost if I consider this lost work time.) Much of that is one-time only, though: if these people are good, for instance, it'll be easy to go back to them for later research. And while the concept of outsourcing is to save money, this exercise isn't so much for that as for the experiment. If I do get great results, it'll be worth the $38.25 and I'll write off the cost as learning time.

Of course a true "4 Hour Work Week" system means outsourcing pretty much everything, which this does not come close to doing. But with my line of work, for the foreseeable future, outsourcing/hiring out my work means I'll be spending most of my time managing other people, which is not necessarily less time-consuming, and isn't what I want to be doing right now. In the long run, since time isn't scalable, I'd like to switch to a product model, but that's a whole different ballgame.

I'll post updates here when I get some concrete results from the experiment. Stay tuned. (Also keep an eye on Chris Shattuck's blog for the results of his similar experiment.)