Information Management Quote of the Week 30/08/14

“Ignorant power comes to the same thing as wicked power; it makes misery.”

Mary Anne Evans (George Eliot)

See also:

Follow the Yellow (Data) Brick Road, my pretty....

Chief Data Officers need to stay on the data path - even if we don't really know where it leads yet.

Just as in the "Wizard of Oz" when Dorothy dropped a house on the Wicked Witch of the East, the emerging role of the "Chief Data Officer" (CDO) has come in from nowhere and is suddenly being lauded as the solution to all our data problems. Plenty of Munchkins are looking to this new position to be a key mover and shaker in making a better data world - even if we're nowhere near settled in terms of what it's actually there to do, or how it's going to do it. 

Yet already, the flying monkeys are circling, with speculation that the CDO role isn't going to be wearing the ruby slippers for long. For example, as this ComputerWorld article explores.



This quote by Peter Aiken particularly got my attention: 

"Once that job is finished, and data management is ingrained in the organization, the need for a specific chief data officer may disappear."

Really? We haven't even started seeing whether this role is actually able to contribute, yet we're already writing it off? 

Now, I accept that it may have been taken out of context, so apologies to Peter if ComputerWorld has taken his comment in a direction that he didn't intent. But taken at face value, I think that's not just a polemic statement - I'd go so far as suggest that it's actually almost meaningless. 

From my perspective, there is no business that I've ever come across has reached a level of maturity such that information is ingrained as part of its behavioural DNA. It's a constant process of renewal, innovation and evolutionary change. Based on the fact that people are involved, it couldn't really be otherwise! 

If nothing else, Business Entropy comes in to play to ensure that there's never going to be an ordered state. Where data is concerned, it's the CDO's role to be accountable for facilitating information-enabled thinking and injecting energy into the whole ecosystem.

I challenge anyone to come up with one concrete example of a truly information-enabled organisation that doesn't need some form of catalytic input.

And while I'm at it, I might as well also take issue with David Mathison's contention as also quoted in the article, that there are three "CxO" roles in play (Data, Digital and Analytics). 

Lions and tigers and bears? (Oh, my!)

For me, part of the problem is that these things end up getting thought about separately, rather than being complementary facets of the same business problem - that of achieving evidence-based decision-making. (As I already made a case for when I published the example Data Governance Charter.) It doesn't really help to separate them out in this way, because sure as dammit, companies will actually try to hire three different people and chaos will ensue, rather than looking at implementing a unified vision. 

The ruby slippers need to be used in tandem with the witch's broom and the magic spyglass.





OK, so we're still working out exactly where this particular Yellow Brick Road leads (to the Emerald City, hopefully. Or at least an opal mine...). But one thing's for sure, I'd rather stay on the path than wander off into the haunted woods of data chaos, or fall asleep in the poppy fields of information apathy....

Keep to the path, and we might even make it back over the rainbow...

The "Three Business 'V's" of Big Data...

...and the last confession of a Turncoat.

Well, today I signed the paper that put me back into the ranks of the gainfully employed. There's a little bit of irony involved, as having taken Gartner to task a couple of weeks ago, I'm now going to go off and join them!

What? ADD as Turncoat Analyst? Gamekeeper turned poacher? Or is it the other way round?! I’d like to think of it more in terms of "If you can't beat them, join them." Or else, it's a unique opportunity to be a mole on the inside… However you call it, I'm not about to temper my outspoken approach, that's for sure! 

It’ll be interesting, however it pans out!

I'll admit that I'm really excited by the prospect of become a Research Director at Gartner. I'm hoping that this new role gives me a unique opportunity to further explore many of the issues that we experience within the Information Management and Analytics sector, and influence the way we think and act with data. It should also mean that I will be able to delve into a lot more detail than my blogging enables me to. Even my "discussion paper" series doesn't really provide the channel to go in-depth into issues and compile the supporting evidence that I would ideally wish.  

The down-side is that I will probably have to curtail my self-published content as I start putting most of my material out through the Gartner channels. (Though there's always the chance that a particular "too hot for TV" moment will need to come out under my own auspices!) I'll also continue to Tweet on a regular basis.

But before I turn my cloak altogether, I thought I'd follow up my previous comments on data quality in "Big Data" environments. More precisely, here's why I think our industry is currently missing a trick when it comes to "Big Data" (or as it should more properly be called, "data".)

Now, the first challenge is that there is still much disagreement about what constitutes “Big Data”. The original suggestion that it is “any data that can’t be processed by traditional methods” is hugely unhelpful, as would be any attempt to define any thing as being “not another thing”. (Would we be comfortable in defining a “dog” as “not a cat”?)

In the past few years, the technology sector has generally settled upon defining “Big Data” based on identifying certain characteristics of the data set, with those characteristics all beginning the letter ‘V’. Gartner analyst Doug Laney originally proposed three ‘V’ characteristics – Volume, Velocity, and Variety.

These three ‘V’s help to establish characteristics and bound the problem of what “Big Data” might look like from technical perspective. The new breed of data tools certainly enable the engineering of new and innovative methods of processing data that were previously out of reach to all but the most well-funded of organisations. There is no “so what?” factor that jumps out at us to make the problems of “Big Data” meaningful in a business context.

I therefore suggest that a shift in thinking is necessary, to examine the “Big Data” challenge not from a technical perspective, but from a business one. To maintain consistency with the original model, these additional business considerations for “Big Data” can also be expressed as ‘V’s – Variability, Veracity and Value:

• Variability: Within any given data set, is the structure of that data regular and dependable, or is subject to unpredictable change? If so, how can we understand the nature of the “unstructured” text data content (or sound, or video) and interpret it in a way that becomes meaningful for the required business analytic-ready output?
• Veracity: How do we know that the data is actually correct and fit for purpose? Can we test the data against a set of defined criteria that establish the degree of confidence and trustworthiness? What are the business rules that enable the data to be tested and profiled? If there are issues with the data, what actions can be taken to clean and correct the data before any analysis is carried out.
• Value: What is the business purpose or outcome that we are trying to meet? What questions are we seeking to answer, and what actions do we expect to take as a result? What benefits do we expect to achieve from collecting and analysing the data? Has the data been aligned with the desired outcome?

All three of these additional characteristics require a clear understanding of the business context, which then is used to frame the meaning and purpose of the data content. “Variability”, “Veracity” and “Value” all express different aspects of the fitness-for-purpose of the data sets in question, all of which need to be addressed in order to solve a business problem in business terms.

If expanding the "Big Data" lexicon to a "Six 'V's Model" becomes my first contribution as a Gartner Analyst, then it's probably not a bad place to start.

Information Management Quote of the Week 22/08/14

"He uses statistics as a drunken man uses a lamp post – for support rather than illumination."

(Andrew Lang)



See also:

"Neat but not gaudy"

When is "good enough" good enough in Information Management?

A recent article from Image and Data Magazine offered some ideas on achieving "best practice" Information Management. The advice offered in the article is well given and it raises some important points, through the lens of a Records Management/compliance-based agenda. 

What the article didn't ask was whether "best practice" is actually required in the first place, or whether "good enough" is good enough. What does "fit for purpose Information Management" really mean? 

And of course, that got me thinking...

First off, I'd offer that unless you understand what "best practice" really looks like, then you're not really in a position to assess how to deliver "fit for purpose!" Conscious competence is what we're generally aiming for, and the route to achieve that will depend upon the starting competency levels within the business community (as I discussed in my post from the very start of this year.)

Another challenge is that best-practice Information Management is a complex and inter-disciplined set of capabilities that many struggle to fully comprehend. It therefore becomes difficult to engage with the business community (and indeed IT teams), because they typically want immediate results, whereas IM requires a holistic change to develop a range on foundational competencies (see my Information Management Tube Map for a view on how these competencies inter-relate.).

Bottom-up, localised initiatives CAN deliver value, however. Indeed sometimes, they're the only way to achieve any progress at all. This will often depend on cultural and societal conditions within the organisation  and in some cases, it can actually be counter-productive to embark upon a major, centralised initiative. (See my discussion paper on Distributed Data Quality for more on this.)

Finally, I'm not convinced at all that a Records/compliance-based approach is going to pay dividends either. Quite frankly, it's boring and most business people aren't interested in discussing risks! Seeking out new value opportunities and delivering a real business outcome will stimulate interest and engage a business community. Once there is momentum, then you can choose which practices you want to develop, as well as establishing a level of robustness into the overall process.

Whenever he'd finish an odd-job around the house, my grandad regularly used to say, "Neat but not gaudy, as the monkey said when it painted the piano green."* When I was a child, I had not the faintest idea what the bloody hell he was on about. 

Thirty five years later, I understand that he was making perfect sense all along...

*Other sources offer a slightly different usage of the phrase, amounting to the same thing. And yes, I know that's an ape, not a monkey. So sue me!

Calling a spade a @£$!% spade!

A few posts back, I called out Forrester Research for propagating a technology-centric approach to Master Data Management (MDM). Today, it's a vendor's turn to be on the receiving end of my irascibility.

This video segment titled "Information Governance, buy now or pay later", was posted to BitPipe.com by storage vendor CommVault, which starts out by broadly describing a view of what "Information Governance" is. However, it then immediately goes on to describe the issues and rationale for better and more disciplined data storage.

Now it's bad enough that as an industry, we're guilty of having multiple terminologies for the same thing. This leads to confusion at best, obfuscation at worst. It's the Information Management equivalent of struggling about whether to to call something a spade or a shovel. And it's even worse that we see vendor companies actively inventing (ugly) new language to try and give the impression that what they're doing is fresh, innovative and exciting. (I'm accusing both product vendors and systems integrator/consulting services business here!) Whereas more often that not, they're just re-badging existing ideas to fit with the latest round of Bullshit Bingo hype. "Manually operated excavation device", anyone?

It's this sort of thing that's given us Decision Support System (DSS), Management Information System (MIS), Business Performance Measurement (BPM), Business Performance Management (BPM), Business Intelligence (BI), Operational Intelligence (OI), Online Analytical Processing (OLAP), Business Optimisation and Analytics (BOA), "Big Data" etc. All, effectively, describing the same damned thing.

But to my mind, what is going on in this CommVault commentary is something even more pernicious and potentially disingenuous. This is a classic example of "bait and switch" language use that is all too prevalent within our industry;  the new, exciting capability (in this case "Information Governance") becomes a bandwagon used to to re-badge products from a completely different domain (such as CommVault's backup & storage tools). 

CommVault have started out describing a spade/shovel, and then gone on to try and sell you a skip.

If the segment had been called "Data Storage Management, buy now or pay later," then I wouldn't have an issue (although data storage isn't at all exciting or vibrant, is it?!) And in any event, I can see why a product vendor like CommVault might want to do this - buyer beware, says I. 

To propagate such muddle-headed thinking just isn't good enough in my view. We should all look to the "Big-A" Analyst and Market Research firms like Garter, Forrester and IDC to provide better guidance which clarifies and helps navigate this mess, not add to it.  That requires more rigour, more care, more thought. Fast-and-loose with language doesn't do anyone any favours and vendors, integrators, IT departments and end-users all end up losing out.

So, until such time as we finally get our data/information stable cleaned up and all the sh*t shovelled into neat piles, it's up to those of us who have the temerity to define our terms before proceeding to try to hold the line, continue to raise awareness, and hold the industry at large to account. 

It's an uphill task, I have no doubt!