A Word (or three...) about Data Standards

I've had a number of recent conversations with my clients and business partners about the importance (or lack thereof) we, as a community of performance managers, are placing on data standards. After all, the ability to compare, analyze, and effectively mine for insights among peers depends on being able to have some type of common data lexicon to rely on for our conclusions.

Throughout the course of history in the PM discipline (which is still relatively young), we have seen some bright spots. For example, many industries have in fact established reporting standards that exist to this day. Safety concerns, for example led to the establishment (and proliferation I might add) of airline accident and incident reports. If one wanted to (and I wouldn't suggest doing this before your next plane trip), you could literally find dozens of online databases that profile this type of data and be reasonably assured that the data is comparable across carriers. That is because the NTSB and FAA (save for some inter-agency inconsistencies and recent infighting) require very clear standards regarding when, how, and to what level of detail incidents should be reported. We also see this prevalent in the Nuclear power sector, where again, the main driver is public safety.

But what about where the main driver is something other than public safety? In regulated industries, we see evidence of similar reporting standards having emerged in banking , healthcare, and local transmission and distribution utilities. In these cases,the driver was more the regulator (e.g.- FERC for the utility sector) who set up these standards to prevent monopolistic control from being exerted on the industry and to hold these otherwise shareholder driven companies to "reasonable" levels of performance. Most of these reporting standards worked well for a while. But as deregulation occurred and competitive markets began to control themselves from a performance standpoint (as capital markets often do), the need for these standards began to wane. Sure, we still see some of the reporting artifacts still in place today (for example, the utility sector still requires FERC reporting), but it is almost always viewed as a necessary distraction for the real operational executives within these organizations. The vehicles that were designed to produce a rsolid eporting standard, now produce some of the least reliable information around And that should be no surprise. When standards are set up to oversee or punish an organization for not achieving a result, you can bet the data will be "worked" and stretched to the maximum extent possible in order to achieve that result.

While I may be accused of being a bit pollyanna, I genuinely believe that the setting of data standards can in fact work well. But the underlying PURPOSE that drives data standardization must be changed. I am of the opinion that if these systems were set up to enable each company to achieve their fullest potential, from both the cost and effectiveness sides of the equation, companies would be a lot more diligent and honest in their adherence to standards.

In the power sector for example, there are some benchmarking and best practice sharing programs that do a great job at reporting consistency. In fact, if I were to bet on the result, I'd put my money on the data that was reported in these programs well before I would trust the more institutional standards like FERC, NERC, and the NRC, for example (all of which have many more years of history under their belt). Why? Because the benefit of complying in the former case is directly proportional to how much an organization LEARNS from the data that is shared. If you twist the data to reveal a better overall position on the scorecard, you're only hurting yourself.

I believe there is room for a new standard to emerge across all of the industries I mentioned and beyond. A standard that originates from a desire to be competitive and learn, versus one that is set up to regulate and punish. Such a system would need to revolutionize everything from accounting treatment to the work management processes themselves- where a dollar means a dollar (no red dollars and blue dollars) and an outage means an outage. Sure, it's complex, but as performance managers, we've seen these standards achieved way back when it was done for regulatory and safety purposes. Not for long, but it did work. It failed not because it couldn't be done, but because it had an underlying driver that was flawed. A good system became just another form to fill out.

So let's try this again, only this time lets make sure that the people reporting the data see the clear benefit of complying with a standard. Only in this way will we end up with a system that is sustainable in the long run.