| Home | FAQs | Book Contents | Updates & News | Downloads |
I was discussing the benefits of improving data handling with a particular executive of a Middle Eastern oil company. Unlike the majority of his contemporaries he didn't want to know anything about how improving his organisation would open up new opportunities; his sole concern was to know how it could reduce current costs. To me this focus seemed completely nonsensical. Currently oil is selling at, let's say $100 per barrel, and recovery costs in the Middle East are, let's say $10 per barrel. Saving 10% of the extraction costs (somehow without reducing either the current production or the eventual recovery rates) only increases profits by $1 per barrel, and the fact that their overall production volumes are so high makes achieving those levels of savings really very hard. In contrast if investing increases the recovery costs to $11 per barrel (and remember that for the same reason this is actually a ridiculously large amount of additional overall spending) the profit stays the same if production (or perhaps more accurately the recovery rate) increases by more than 2%. In the current situation investment would seem to be the obvious best strategy. The maths means that the required increase is proportional to the inverse square of the ratio between extraction cost and selling price. When that is in the range 10-20 (which it is at the moment) then increasing recovery is attractive, when it's in the range 2-3 (which it was for much of the 1990s in much of the world) then cost savings are the order of the day.
Naturally this also depends on how easy it is to make savings without affecting performance. With infrastructure the ways to contrast network equipment from one source with that from another are fairly widely understood. OK, so there may be minor differences, like which government has the back-door access codes, but fundamentally network kit is a commodity. Like all commodities you specify what you want and select the supplier that can meet your requirement at the lowest effective cost. In that case a cost focus might work (sometimes).
If data handling were that type of a mature field, that is if everyone agreed about how it should be done and skills to deliver it were widely and easily available, then doing it as cheaply as possible would be exactly the right thing to do. In the real world that is clearly not where we are, every company has anecdotes about how bad data has exposed them to serious risks, and stories about when unexpected key data saved projects or revealing new opportunities. The fact is that there is not (yet) general agreement about what "data management" is, such as what ranges of technical data and activities are covered. More importantly there is no easy way to distinguish between oil industry data managers who deeply understand (or to use a specialised term "grok") the subject and those which are just doing it until a "real job" comes along. Now this isn't new, I've always said that one of the big differences between "data management" (DM) and "information technology" (IT) in the oil industry is the fact that in the current financial climate the first is an opportunity to invest in improving the current (frankly poor) situation while in the second the best approach is to focus on continuing to deliver the current level of service cheaper. Surely, by this point, everyone should know that?
 Article 50 |
 Articles |
 RSS Feed |
 Updates |
 Intro |
 Book Contents |
 All Figures |
 Refs |
 Downloads |
 Links |
 Purchase |
 Contact Us |
 Article 52 |