#1: You'll Get There Sooner if You Have Help. Since this isn't just an IT problem, you'll need stakeholders from IT, Legal, Compliance, Records and business units to identify the most likely targets for defensible disposal, and to establish standards and mature information lifecycle processes. With the right people at the table, you can use the guidelines established by the CGOC and the Information Governance Reference Model to help you determine your best plan to shed that unwanted data weight. The goal is to tie information value and duties to the data assets you maintain. Make sure you also have executive sponsorship. The reduction in both cost and risk will usually get your general counsel, CIO, or CFO behind you.
#2: Start With Small Gains. When there's a lot of data weight to lose, the process can seem overwhelming. Instead of spending many months or even years in planning and coming up with the perfect plan for the entire enterprise, target one or two departments. You'll learn a lot about the kind of data to target to inform the rest of your process. Also, the success you see with a small win or two will create momentum to help power bigger initiatives.
#3: Invest in Equipment. With clues you've gained from the targeted efforts you'll have some metrics on the amount of data you can defensibly dispose of. By extrapolating this to the rest of the organization based on volume, you'll get a rough idea of the amount of savings in data storage as well as the downstream impact on regulatory and legal compliance costs. That will help you define the projected return on investment to create the business case to invest in technology to automate the rest of the process. It doesn't require enterprise wide, seven or eight figure investments either.
Look for technology that will help you analyze and target existing data stores for disposal candidates. New generation solutions can index and parse through large quantities of data to speed the process while also providing documentation for defensibility. These new engines can classify data by volume, age, type, content and context while also extracting entity information and providing usage information. This sheds light on so called dark data, data that has been neglected or forgotten in archives or other locations and adds to data bloat. The analysis will also help you sweep out dusty data, the data you may know about or suspect but is obscured by time or is hard to determine disposition. Both dark and dusty data add to organizational risk and should be dealt with to bring your data back to health.
#4: Focus On Your Biggest Problem Areas First. All the steps above will show you some of the biggest areas for data improvement. Take active steps to remove those easily identifiable, non-value data classes right away. Even if you only find half of the typical 70% that has no business, regulatory, or legal value, that's still 35% of your data and savings in the millions of dollars.
Sign up for CIO Asia eNewsletters.