I’m writing this post on a dare. I was challenged to write about some of the questions we thought were settled in 2005 but that in 2012 are again hot topics as we head into EMC World 2012. This one is about Information Lifecycle Management, or ILM.
Everything old is new again. That’s because no matter how much IT changes with tools, density, capabilities and whatever is the flavor of the month “strategy”, the basic plumbing problems simply don’t change very much. To wit: growth in unstructured data is anywhere from 30%-70% per year for the typical enterprise and has been for what seems like forever. Since spending on IT has been down the past few years I now have clients sitting on petabytes of the stuff, befuddled like a hoarder after the city comes by with a condemnation order. The closets are full; it’s time to purge.
It’s always been time to purge (or at least archive). Good progress has been made on the e-mail front in large part because archiving solutions and underlying platforms like Exchange are married up and have matured nicely over time. The e-mail archiving space is littered with dozens of on premise, hosted and hybrid solutions. Similarly, data stored on file shares or in a CMS can be addressed with one of the scores of effective tools on the market. The question to me is “why do clients seem to have e-mail archiving figured out but eschew archiving or purge in file shares and other unstructured repositories?”
Before I go on, plenty of solid IT shops employ good strategies and tools for managing their unstructured estate. My exposure to the largest data hordes is biased, of course, and is self-selected for organizations with serious data management challenges and few tools to tackle them. Nobody who has it under control calls me.
For calls I do receive, the skeleton in the closet is that ILM has never really been addressed from a practical standpoint. Call any vendor of a data management/content management solution and they are all the magic bullet. Therein lies the root of the matter. Magic bullets make us think we can slay dragons we have no business engaging.
Practical ILM is elusive because those tackling it tend to focus solving every problem in the unstructured space. Sally Dovitz runs our Compliance Practice in EMC Consulting and asked me to share two things she wishes IT shops would stop doing:
- Stop trying to classify all your data. You’ll never finish.
- Think twice about that 500 page record schedule, nobody uses it.
From her perspective it is enough that we handle the protected data differently. Our policies should cause the least harm and accommodate the truth that people are crafty. If someone really wants to squirrel away some data because your policies are draconian, they will find a way. Overboard record schedules create false positives which crush your savings or the workability of your data management strategy. For Sally, it comes down to setting up repositories and taxonomies that make sense to everyone; those are the policies that are sensible and successful. That means talking to the users or stewards of the files to understand how they leverage those documents in their jobs. In most shops, “records” or officially regulated data are a small component of the overall picture, so why spend the bulk of your time and budget managing the minority slice only to alienate the majority of your consumers?
ILM is back on the business dinner agenda. I’m already teed up for 2 conversations at EMC World this week on that very topic. I’ll be telling my clients to read the above and keep it simple. The follow-up question will be “that’s a nice strategy, but how do I clean up the mess I already have?”
I’ll share some tactical steps tomorrow. I wouldn’t want to let spoilers go before dinner. Back to EMC World 2012!