[I’m on holiday this week, So I won’t be able to reply immediately, but I will once I’m back. Probably doesn’t matter, because this is about accounting principles, and that’s going to put a lot of people off reading this post, far less commenting on it. I expect the sound of crickets.]
I got posting on the subject of spindown last week, and made the assertion that it was a Jolly Good Storage ThingTM (patents pending), if and only if the cost of being able to take advantage of it was less than the cost of doing nothing.
Spindown falls into that category of unknown; how much does it cost me to take advantage of this technology? It might seem pretty obvious that it can save you money, but measuring the effectiveness of something that looks at first blush to be cost efficient isn’t that easy. (There’s a difference between efficient and effective in my book; being efficient means doing things well, effective is doing the right things well.) Spindown is just one example, as I’m going to try and develop a costing model for storage (it applies to other aspects of IT too, btw) that can answer the question; is the next cool technology cost effective?
In the late 1980s I was privileged to work at a company where we developed and promoted an IT based ABC (Activity Based Costing) methodology. ABC was then in its infancy, having been first formally defined in around 1987, but the things I learned are as applicable today as they were 20 years ago. Yes, I’m that old.
So, to business.
Defining the Terms
First, let’s define some terms that are in common use. This isn't going to be a lecture on the subject (I think…), just a light overview with some serious simplifications along the way.
Activities, or doing stuff, costs money. Building cars, baking bagels, assembling office furniture, making phone calls; all have a cost associated with them. There are a few cost types we’re interested in, and I’ll give examples of each, but I won’t start with storage because the principles are best illustrated using other technologies. I’ve chosen printed output as an example.
- Fixed costs; example, the initial purchase cost of a printer. They tend to be long term costs that don’t vary, regardless of how much we output in terms of product (well, up to the capacity of the printer, but that’s dealt with in a subsequent post).
- Variable costs; costs that vary in proportion to some measure of output. Paper costs and toner cartridges fall into this category. No print, no variable cost.
Two further classifications of cost are useful. They are;
- Direct costs; costs that I can directly assign to some product or service. For printing, an example might be a printer that I have on my desk or in my department. All the costs can be associated with what I do, and be directly assigned to my product or service.
- Indirect costs; what’s left. for instance, a shared printer that provides many departments or functions with their output.
Costs can usually be assigned to one of both categories. Buying a printer for departmental use can be considered a direct fixed cost. Centrally purchasing paper can be considered an indirect variable cost. And so on. I’ll demonstrate where and how storage costs get allocated in the next series of posts.
(There are some costs we can’t classify as neatly, what are sometimes described as business sustaining costs. Where they occur, it’s normal to spread them out over the products or services in relationship to their costs. Like, for instance, the chief executives’s salary. For the most part, I’m going to ignore them. Both the cost and the chief exec :-)
One further piece of terminology we need is
- Cost driver; this is the activity that causes the costs; in the printer case, I might be generating documents associated with an insurance claim. The claim process is the cost driver.
Why Classify?
Take the current recession we’re in. We want to drive down cost for every unit we produce. It’s doing things, or activity, that costs money, so the classification process helps us understand how profitable a product or service is based on various volumes of output. An example, this time from the automotive industry;
- This month I sell 1000 cars at $10K each. They cost me $9K to produce. Net profit is $10,000K-$9000K or $1000K. Happiness.
What if there’s a slump in orders? Can I work out what the effect would be? Without some form of costing, that could be difficult.
- Next month, I sell 500 cars at $10K each, They cost me $11K each to produce. Net profit is $5000K-$5500K or -$500K. Actually, a loss of $500K. Sadness all round.
A diagram is in order. (The unit cost line has a different scale from the total cost line).
Some costs (plant, machinery, labour) are all fixed. The variable part is the cost of the parts that make up the car. So my unit costs vary by volume. But if I don’t classify my costs, then I can’t work out what my profit or loss would be at, say, 2000 cars or 200.
That’s it for this post. Next time, cost drivers in the storage world, how we can apply these categories. With some simple models, I hope to demonstrate that we can put ABC to use, and test some of the technology-that-saves-money claims.
.
My granny always used to say; “One man’s meat is another man’s poison”. And Marc Farley’s 
Over on his blog, Steve Foskett has created a list of his 
