The IBM Data Governance Council sent out a press release this week predicting that within the next four years, data will become an asset that is reported on the balance sheet of corporations, and that data governance will become a statutory requirement.

This trend could bring a new emphasis on data quality and potentially increase corporate use of social networking as a means of improving that quality.

The IBM Data Governance Council is a group of 50 companies, including pharmaceuticals, telecommunications, and, of course, financial services firms. (The last category is key: When it comes to data governance councils, you have to have representation from financial firms, and this council does, from companies like American Express, Bank of America, Citibank, Deutsche Bank, and others.)

What struck me were two predictions made by the council, namely:

1) Data value will become an asset that the CFO reports on the balance sheet.
2) The CIO will be responsible for data quality, which will become a new IT key performance indicator (KPI).

I was struck by this because about a year and a half ago, Michael McCreary, then the head of Pfizer's Legal IT department, told me the CIO was going to increasingly focus on information asset and liability management. He convinced me that I should start writing about this, which I did, both individually and with Michael as a co-author.

When I sent him the IBM council's press release, he wrote back: "Looks like we're on the right track!" Perhaps Michael's right, but now I'm trying to envision how the council's predictions will play out for IT. Initially, we were just using the balance sheet as a metaphor to describe information as both an asset and liability. The IBM council is going much further, predicting that financial reporting and accounting practices will change to actually accommodate this vision.

Let's assume this will happen and that somehow companies will be able to reasonably quantify data value and consequent risk and make data a balance sheet line item. What does this mean for CIOs? Here is how I see the situation developing:

• IT organizations will become increasingly expert in information quality.
• Testing will become a factor in making this happen. Does the product, in this case data, meet the expected requirements and specifications set forth?
• Feedback mechanisms will be required. Companies will need to track when things are not right and initiate a process to correct them and an understanding of why things went wrong. It will be vital to have a process to eliminate these problems so they don't reoccur.

As a simple example, consider customer data (e.g., name, address, and contact files). What percent of customer data is correct? Statistically, do you know what sources of data are more likely to be incorrect? Do you have processes to improve the lower-performing data sources?

How about a more complex example, specifically, the accuracy of projections? How correct are sales forecasts, financial forecasts, project completion dates, assessments of external factors? How dependent is a business on these factors? Can data sources be traced, improved, rationalized, and integrated into a holistic decision-making system for which quality can be measured?

If you feel organizations already have a handle on this and there's really nothing new here, I have two words for you: 1) subprime; 2) meltown. If the quality of decision making becomes a KPI, that will ultimately feed the balance sheet. But there's a lot of work to be done.

So what's the final answer? I'm honestly not sure, but I can give you some buzzwords to chew on that might point the way. These would include data governance, true information lifecycle management (ILM), automated data classification, automated risk assessment, new system designs, and automated policy management. And I can think of some measurements that might become KPIs too, such as accuracy, consistency, completeness, cycle time, and continuous improvement metrics. But frankly, these leave me with more questions than answers.

Cue social networking
Here's a thought: Can social networking technologies play a role in boosting data quality? Is there an application for collaborative decision making whereby investment and risk management decisions, which are often made in ivory towers, become more transparent and complemented by the wisdom of the crowds?

This is social networking applied not for meeting people and taking movie quizzes, but for sharing knowledge in near real time and creating a continuous feedback loop where business users directly participate in improving data quality.

Here's a simple example that doesn't require a functioning semantic Web: A user of a particular data-set rates and perhaps even reviews the quality of that data. This information is made available in a blog, wiki, or other other social networking platform to business users so that individuals can more confidently understand the strengths and weaknesses of its use, or even suggest ways to improve the quality of the data. As data quality evolves, so will the rating of that information and its business value.

Social networking could also be used to avert potential business disasters. For example, certain investments could be flagged as having an unusually high risk profile. Often voicing such concerns in an open dialogue can completely change an organization's behavior the way a whistleblower can bring corporate wrongdoing to the forefront.

Ultimately, though, data quality control and valuation is one of those problems that perhaps machines can't completely solve. Humans will have to be involved in the final decision process in some way. While humans won't get it right all the time, groups of business users in a social network clearly can play a role in providing early warning signals to potential threats as well as finding new ways to improve data quality and enhance business value.

Crazy? What do you think?

David Vellante is a co-founder of ITCentrix Inc. , Barometrix, and The Wikibon Project.