I love them three-letter acronyms!
An “ROI” is a Return On Investment – it’s a basic financial calculation that places investments in a standard number so that they can be compared. If the number is positive, then it has a “good” return on investment, and if it’s negative, you probably shouldn’t make the investment. The basic form of the calculation looks like this:
ROI = [(Payback – Investment)/Investment)]*100
There are, of course, variations.
So what does this have to do with the Data Professional? Well, right now IT is being asked to consolidate, reduce costs, eliminate headcount, all kinds of moves to “save money”. But is anybody checking to see if any of that really works?
Don’t get me wrong – I’m all about saving money, and definitely about using every resource as much you can. Peg those CPU’s, spin those drives, cloud that app, virtualize those servers. But don’t do it for activity’s sake – learn how to do an ROI, and do some hard comparisons. In fact, ROI’s should be combined to form another calculation, called Compound Return. This lets you know about all of the savings combined.
Something you’ll notice in the ROI calculation – there’s no mention of time. There are a couple of ways to deal with this – you can factor the time as a part of Payback, or you can use the Investment part as a time calculation. How you do that depends on how you price out time.
So when you’re done estimating the ROI of one decision over another, don’t forget to come back and test your assumptions. This is the part I see done the least. Once the decisions are made, it can be “dangerous” to come back and see if it really saved any money – but it’s a step you need to take.