TCO or "Total Cost of Ownership" is a notion that one can calculate (with some accuracy) the complete cost of owning something, including all the weird side effects of acquiring and owning that thing.
For example, I can by a new 3.2GHz notebook for $2,000 and it comes with Windows XP. But odds are that I'll spend 20 hours in the first year dealing with device drivers, spyware, and viruses. If I value my time at $50/hour, then the total cost of owning that notebook for the first year is actually $3,000.
Then I'd take that number and compare it to the Powerbook I was thinking of getting. It'll cost me $2,600 and have a "slower" CPU, but I'll only spend 2 hours screwing around with it in the first year figuring out why my scanner doesn't work when I plug it in. That puts the total cost of the Powerbook at $2,700 in the first year.
Anyway, you get the idea. It's why we think about maintenance costs and gas mileage when looking at car's sticker price. We want to know which car is cheaper in the long run.
(Those numbers are completely made up. If they were real, the "cost" of spyware and viruses would probably be higher.)
IT organizations often use TCO numbers as way to justify inflicting poor technology choices upon those they exist to serve (or abuse, as the case may be).
"But this saves the company money."
Often times, organizations try to take TCO driven decision making to the extreme and mandate a single standard for this or that. My previous employer was, unsurprisingly, good at that too. In fact, at my old job we often referred to "The Cost of Being Different" (TCBD?). It was used to win arguments and sometimes short-circuit groups who began to stray from the heard and look at software that was not on the "approved" list, regardless of their reasons for doing so.
In theory, this all works well form a high-level organizational point of view. But if you ever venture down the ranks and ask to folks who must live with the results of TCO and/or "standardize at all costs" decision making, the tone of the discussion changes quite a bit.
We need a way to quantify the negative effect that these decisions often have on the day to day folks (who'd rather be left alone to get their jobs done). The pain they endure. The countless hours spent fighting with a technological choice that was clearly not optimal. The effort required to work around product glitches or to bring in a replacement "under the radar" and keep it there.
I think this should be the TPU or Total Pain of Using.
I have no idea how to setup the scale and ultimately convert it into a dollar figure (because that's all the CTO and board of directors seem to give a shit about), but it'd be worthwhile to compare the relative TPU of "similar" products, I think.
Hmm. I guess TPU could also mean "total pain units." That'd be just as accurate in many cases.
Posted by jzawodn at August 16, 2004 01:52 PM
You'd have to look at it as an extension of ergonomics . Maybe someday it will split into it's own specialty, "Zawodnometrics?"
Anyway, it should be possible to measure some important things:
- how much time is spent fighting with the wrong technology vs. time required using the right technology for a given job.
- amount of time burned as employees evade a given task made particularly painful by poor technology choices.
- number of ancillary coworkers' hours consumed for the employee primarily responsible for the task to be completed due to poor technology choices and resulting distribution of burden.
Others should be able to help fill in the blanks.
In my experience, TCO is ONLY used in the making of the purchase in so much as it can be used to steer the decision in the direction of the predetermined choice of the supervising manager. Once the purchase is in place, none of the rest matters. Most small companies (250 of fewer employees) don't take into account the cost of employee time. If a salaried employee has to work 20 hours per week overtime to complete 30 hours per week of work, they really don't care if the other 10 hours could have been put to profitable use.
Go figure.
funny you should mention TCO Jeremy. Immunity Security posted a slightly tongue-in-cheek look at Windows Total Cost Of 0wnership (thats a zero at the start - geddit?)
You can grab the PDF report from
here
TCO (as a term) isn't used much anymore thanks to the legal departments of big companies.
To compute true TCO you'd need to know the cost of electricity + replacement batteries + (optionally) extended warranty + a myriad of costs that are difficult to compute (and vary by geography).
Yahoo now uses the term COO = Cost Of Ownership.
If you need more info on this, speak with someone from Michael Callahan's office.
Yah, but what about the time you spend porting software to an obscure platform that about 7 people use, when you could find easy (and good) alternatives that are already prebuilt for the more popular platform.
What about the (hard-to-quantify) costs of having to live with less-desirable software because your first choice is simply not available on your platform?
There are a lot more "costs" than at first meet the eye.
Does "straying from the heard" mean going to a quiet room? :-)
I think you meant "herd".
As far as laptop TCO goes, I find that I fiddle with Debian on my laptop with remarkable infrequency. Except for running apt-get upgrade occasionally which has only once let me down. (Badly, I'll grant you, but still it was only once.)
