2) Admittedly, much depends on the talent market to which one has access, but sometimes all the tooling and infrastructure in the world can’t make a dent in the fundamental requirement that the junior associates be extensively conversant with the subject matter as well as a variety of technical skills.
I wouldn’t know; I have only been able to afford salaries entry-level folk would take. When I say entry level, I mean entry level. Like, “previously cleaned spyware at Geek Squad” entry level.
The upside of this bargain from the perspective of someone in IP telephony is that at least one doesn’t have to pay for a raft of highly marketable IT skills (e.g. web development) that aren’t needed. That’s about the only upside, though.
3) I was referring specifically to the multibillion dollar PWC and Accenture scale, and made examples of them for that very reason.
Did Joel say something like that? It sounds like something he’d say, I guess. I actually know a few fellow humanities majors who went to work for Deloitte—-people as far from a background in financial auditing and compliance as Heaven is from Earth. Yet, clearly, that’s what they do, because a group of very smart folks figured out how to dumb down, automate, package and bill that.
4) This article was intended for one-man bands more or less, not for people who have scaled up their consulting enough to fund one or more developers. Although still small and not without challenges by any means, that’s a considerably more luxurious place to be.
Also: my whole point is that "previously cleaned spyware at Geek Squad" is kind of a silly hiring premise.
And yeah, hiring that way is stupid, don’t do that. This article was largely a list of things not to do. :-) An omitted premise is that I don’t (any longer) live in a market with a non-negligible concentration of IT talent. Not that it matters much; I spent the last decade in Atlanta and couldn’t have conceivably paid anything remotely resembling a decent IT salary even by those standards.
So, just some general shotgun-blast responses:
* Whether you're trying to scale out or just keep a steady income for yourself, if you're not specializing, you're going to have a bad time. If you're trying to sell basic IT services on a piecemeal basis, consulting is going to be a drag (but: if you can reliably deliver IT services for 20-60 person organizations, you're I think actually in a pretty good place right now, since a lot of our clients are auditioning firms for just that, and there aren't many good ones!).
* There's a whole spectrum of aptitude and experience to hire in between "entry level" and "fully billable" and part of the basic challenge of running a consultancy is figuring out how to ramp up talent and where on that spectrum to hire people. The basic job of a specialized consultancy is finding sharp people and rapidly conferring on them the tools to get billable work done, and then structuring projects so that everyone cal be billable. The analysis of "you can hire entry level people but then it takes months for them to be billable" is pretty superficial; every serious consultancy has strategies for dealing with that problem (some better than others, but all better than "hire and then wait a year and hope").
* The bit about how scaling consultancies is a matter of creating a big binder of methodology so that an English major can execute your projects is lifted straight from Joel Spolsky, and Spolsky is just as wrong as this person is, at least at the scale consultancies started by HN'ers (for all I know it's actually true at Accenture scale, but it's definitely not true at for-hire DBA or security or IT or distributed systems or iOS development scale).
* The idea that consulting work will sap your ability to build product is a bit of a head-scratcher, or, at least, the explanation this article gives is. The "reliable" way to do a product/consultancy is to use consulting headcount to fund development headcount. Developers aren't billable, full stop. When you've reached the point where you can fund 3-4 full time developers, you've run out of "consulting made this hard" excuses. The simpler reality is: building and launching products is hard, and, by a wide margin, most startups (consulting or otherwise) that attempt it fail.