Archive for February, 2011
In his excellent book A Whole New Mind, Daniel Pink writes about how the future is bright for right brain folks–i.e., “creatives.” Pink argues that left brain folks are great at analyzing information but not so great at synthesizing it. He further writes about how right brain folks are particularly adept at systems thinking:
is the process of understanding how things influence one another within a whole. In nature, systems thinking examples include ecosystems in which various elements such as air, water, movement, plants, and animals work together to survive or perish. In organizations, systems consist of people, structures, and processes that work together to make an organization healthy or unhealthy.
Examples of systems thinkers include artists, designers, and poets. They see the whole, not just the individual parts.
This got me thinking: Do we need more “poets” in the data management space?
Systems Thinking and Data Quality
We’ve all seen complicated system architectures before. In most cases, the level of complexity can only be described as superfluous. If I had nickel for every convoluted IT setup I’ve seen in a large company, I’d have an awful lot of nickels. The usual big company IT configuration is overly complex because each person, department, division, or group got to have its own way over the course of a number of years. Rarely did someone with the requisite authority step in to ensure that an organization’s data, systems, applications, and overall use of technology met the needs of the entire organization.
Even within a department, I have seen tremendous dysfunction. (Perhaps I am more right-brained than I give myself credit for.) For example, I can think of few departments that could see the big picture. Each person had his or her own way of doing things, making reporting or any type of meaningful aggregation essentially meaningless.
The ability to analyze is critical in solving problems and thinking about potential downstream effects of a change of some sort. However, consider the following scenario. Let’s say that your organization is only comprised of analysts and middle managers. CXOs are loath to get their hands dirty. That is, no one is minding the store. Is it any wonder that data quality suffers?
Further, the lack of systems thinking often leads to mass employee disaffection and frustration. After all, if you have best-of-breed solutions, then why can’t an organization answer ostensibly simple questions without having to move heaven and earth? While the answers vary, often the answer to this query stems from the lack of a
What say you?
It’s been a lucrative decade for consultants in information management with new work being as easy to win as saying the word “compliance”. Executives are more than willing to sign-up new consulting engagements based the need to meet their compliance and regulatory requirements. The trouble is, this type of information management engagement breeds a defensive rather than a confident enterprise.
A defensive organization believes that data needs to be locked-down, that risks need to be taken out and the analysis resulting from any dataset should be predictable. Of course, any regular reader of this blog would know that we view data contained in large enterprises as complex and displaying all of the attributes of chaos mathematics which means any attempt to remove surprises from data is a fruitless endeavour.
A confident organization, on the other hand, recognizes that data is complex and chaotic but seeks to gain benefit from that complexity. Rather than be afraid of randomness, they use the techniques of MIKE2.0 to identify the risks and then focus on monitoring and measuring. In general, I observe a strong correlation between the confident enterprise and the adoption of Web 2.0 techniques and principles. The confident organization believes that there is more value in collaboration and is willing to sponsor individual innovation.
A good example of why this is so important can be seen in social networking sites such as Facebook. With the rapid growth in their use by a new generation of consumers, service providers ranging from telecommunication and financial services right through to government, need to come to grips with both the technology and the cultural drivers behind them. Consumers are becoming more confident in sharing quite detailed information about themselves in a way that they expect others to pick-up. Increasingly it will make no sense for providers to ask individuals to provide data about their relationships, locale or other details when those are already available in the public web.
In fact, one of the reasons why Facebook is so powerful is its ability to interface into custom applications. Imagine the impact if you wanted to sell these consumers a new financial or telecommunications product and you made it possible to apply online from within Facebook! More importantly, you can give the individual a sense of control by allowing them to privately share critical information with you and then maintain it in a form with which they are comfortable – perhaps for a multitude of providers.
Obviously there are challenges in this type of initiative, but good use of data measurement, reconciliation and parsing approaches allow it to be done. The question is whether your enterprise has even considered whether it’s worth doing? You can bet it won’t be long before your competitors do!
Ever wonder why the technology field is buzzing with the term “big data?” Many sources would tell you that the capabilities of data management and analytic technology can produce greater decision support abilities, business intelligence and higher revenues. Which is for the most part true. But for vendors, big data is boiling down to big bucks.
With the growth of web and transactional technology in recent years, companies are stockpiling data at an exponential rate without a vague clue of how to use it. Vendors are seeing an opportunity to create programs and platforms to analyze and report on this data, and companies are buying in. The catch? It’s not cheap. It’s so expensive in fact, that many companies are left wondering if the business intelligence received in return is worth the hefty price tag.
I’m wondering if a simple cost-saving solution may be found within the business itself. If management was more thorough in deciding what information needs stored and what does not, it would, in many cases, alleviate the need for a huge data warehouse and the expensive analytic technology that goes along with trying to unravel it.
Just a thought.
Nathan Jones is a Manager in Deloitte’s Information Management & Integration practice in London, specialising in information quality and enterprise data management.
He has focused on operational business intelligence, data management, data quality and data warehousing for nearly 10 years, and worked with clients from blue chip energy and finance firms through SME’s to large and small public sector organisations.
Connect with Nathan
In an episode of the popular American TV show ER, an alcoholic patient with political connections is moved to the top of a liver transplant list. One of the staff’s doctor’s voices his outrage after catching the patient sneaking in a few drinks prior to the operation. The doctor doesn’t believe that the patient is worthy of the transplant, as he’ll just destroy another liver better served or someone else–i.e., a non-alcoholic.
I often think about that episode and related questions:
- Is the hospital enabling the alcoholic’ dependency?
- Would “tough love” send a message that the individual has to change his behavior–or face death?
- Is there an incentive for people to cease engaging in destructive behavior if they know that they themselves will have to bear the consequences?
The same holds true for the world of data management. In Why New Systems Fail, I write about how many times consultants have to save the day for their clients–or at least try. When things break bad, organizations often call in people like me to save the day. And sometimes we do. (As a result, some of us consultants have a Superman Complex, although I like to think that I usually keep my ego in check.)
So, we show up and try to fix things. I can’t help but wonder: Are we consultants enabling our clients’ difficulties? Are we really solving their problems?
Note that I am not advocating refusing to do the work that clients are paying us consultants to do. I am simply going to make the argument that, by constantly bailing out employees and organizations with lax data management practices, consultants may in fact be enabling the very problems that organizations are hiring us to solve.
The Usual Suspects
In my experience, most data management and quality problems stem from:
- poorly trained or lazy employees
- poor documentation for business processes
- inadequate internal controls
- redundant or overly complex internal systems
- a culture that tolerates errors
- poor performance management
- weak senior leadership
In other words, rarely do discrete and external events such as a software bug or rogue end user cause major problems. The seven culprits identified above do not fall into the “easily fixable” category, at least in the long term. As such, the consultant(s) that triage the situation do very little to prevent the same problem(s) from recurring.
In fact, by heeding our clients’ calls, we consultants might even be increasing the chances of recurrence. Why? Because we show that we can often work our magic and return things to normal states without the organization or its employees making any fundamental changes to their behavior, processes, or systems. While consultants aren’t cheap, depending on the engagement, a $200,000 USD cost may in fact be less expensive than changing the organization’s culture or replacing ineffectual CXO’s–much like getting a new liver is “easier” than entering Alcoholics Anonymous.
Look, it’s always tempting to believe that the solution is an external entity, be it a consultant, acquisition, or piece of technology. Why do you think that, particularly in the United States, sales of weight loss products are growing by so much? It’s just simpler to try and find salvation in a pill than watching what you eat and exercising a few times per week.
The burnt hand teaches best. Sometimes, organizations that experience major data management problems ought to try and solve them on their own.
What say you? Are there merits of not fixing organizational data management problems?
Information is a key asset for every organization, yet due to the rise of technology, web 2.0 and a general over abundance of raw data, many businesses are not equipped to make sense of it all.
How can managers overcome an age of “information overload” and begin to concentrate on the data that is most meaningful for the business? Based on your experience, do you have any tips to share?
Ken O’Connor is an independent IT Consultant, specialising in Enterprise Data Management and Quality Assurance. He has almost 30 years experience across the full development lifecycle.
For the past 15 years, Ken’s work has centered on Enterprise Data Management. As a Senior Consultant in IBM’s Euro Centre of Competence, he helped develop End-to-End processes to handle Euro Changeover for clients in Ireland and across Europe.
Ken has helped maximise the reuse of intellectual capital by IBM consultants across Europe through the effective use of knowledge management tools. He has published papers on IBM’s Intellectual Capital Database on topics including Data Management, Testing and the use of Formal Software Development Methods. As an independent consultant, O’Connor has worked with multi-national clients on a variety of data quality and MDM issues.
Connect with Ken.
A dentist friend of mine (call him Larry here) recently emailed me via LinkedIn, asking me to endorse his work. It struck me as odd, but not because he wasn’t recommendation-worthy. Larry’s quite good at what he does. Larry had taken care of me before and I had subsequently recommended him. My initial reaction to his request was befuddlement.
I assumed that he had deleted my recommendation by mistake.
I was wrong.
It turns out Larry had created a new email address for himself in Gmail. Rather than editing his LinkedIn account settings to reflect this change, he decided to create a new account and identity, something that Larry had done as well on Facebook a few days before. Larry didn’t understand the consequences of creating a duplicate record for himself, none if which is good.
When I received Larry’s new LinkedIn endorsement request and figured out what he was doing, I responded by telling him that he needs to stop immediately. Change your email, I wrote. Preserve your unique identity.
Now, Larry’s not a data management professional. Again, he’s a dentist. But this story illustrates a major and often overlooked point about data management: Some people just don’t understand the consequences of their actions. In Larry’s case, adding additional LinkedIn and Facebook profiles (read: duplicate records) effectively splits his world into multiple parts. To put it mildly, this makes things confusing for others–and him, for that matter–to effectively manage his contacts, recommendations, and groups. Larry1 and Larry2 will compete for attention, unless the real Larry deletes the Larry2 profile. The longer he waits to do this, the more work he will cause himself.
Thankfully, I stopped Larry before the Larry2 profile really took hold. I nipped it in the bud. No real damage was done.
Now, let’s take things up a few levels. Consider the following scenarios:
- Think about the sales rep who enters new customer information (e.g., a new record) every time that he speaks with the same prospect?
- What about the HR clerk who carelessly adds additional job codes because she’s too lazy to find the right one–and the application hasn’t been configured to prevent her from doing so?
- An employee in purchasing creates new entries on the item master, negating internal efforts to clean it up.
Unfortunately, I have seen all of these things happen in my years as a consultant and, unlike with Larry, in each case the fix has not been nearly as simple. Many people do not understand the downstream effects of engaging in poor data management practices. While applications and systems can be locked down to prevent outright abuse and certain innocuous mistakes, no system can be completely safeguarded against human error. People still matter, big time.
In the Rush song, “Chain Lightning”, there’s a particularly profound lyric: Ignorance breeds imitation. The implications of those three words for data management cannot be understated. Intelligent organizations need to make sure that their employees are sufficiently trained not only in what they need to do, but in what they need to avoid doing.
Employees who know the right way–and the consequences of the wrong ways–can act as an additional internal check against data management abuse and errors stemming from ignorance.
What say you?
The Benefits of a Semantic Enterprise
MIKE2.0′s Semantic Enterprise Solution Offering provides a layer for the enterprise to establish coherence, consistency, and interoperability across its information assets. Applicable information assets may range fully from structured to unstructured (text and document) sources. The methodology of this Offering is:
- inherently incremental
- layered onto existing capabilities and resources
- flexible to accommodate expansions in scope, new learning, and changes the continuum
As an implementation proceeds and extends across the enterprise, there are exciting prospects to shift the locus of knowledge management and tools from vendors and the IT function to practicing knowledge workers. Real prospects exist from this Offering to overcome decades of frustration in breaking down information silos within the organization. In the process, the organization can enable more effective information sharing and interoperability. This approach is generally based upon the languages and techniques of the semantic web, as distilled and refined by the pragmatic lens of business requirements.
Read more about it when you have a moment.
This Week’s Food for Thought:
Data Management Tips for the Growing Organization
Data management and quality issues are like termites: once you have them, it’s tough to completely eliminate them. What’s more, they’re going to get worse. Much worse. The best approach in combatting them is to be vigilant from the beginning.
For example, employees who aren’t exactly diligent about data entry need to be straightened out immediately. Sure, people make mistakes. No one needs to be crucified for making a legitimate mistake or not fully understanding something. I’m not talking about that.
I’m talking about people who oblivious to the consequences of their actions. Remember, decisions based on inaccurate or incomplete information are less likely to be the right decisions. No one needs reasons to distrust the data.
Information Systems and Business Performance
The lack of understanding of the relationship between process and platform is a key contributor to the extreme spending on information management systems and technology in organizations today. Many managers are quick to adopt a technology to solve their organizational needs without clearly understanding the underlying business process beforehand. It is often seen that these systems are installed and cannot accomodate the current workflow. For example, they can create new processes, but not faciliate existing ones
Process, Content and Collaboration: Key Ingredients for Enterprise 2.0
Taking a look at today’s business, most companies still have a lot as far as adapting to and leveraging Web 2.0 possibilities is concerned. Sometimes it’s because of fear to open up, sometimes it’s non-compliancy reasons but in most case it’s still unfamiliarity with what benefits it could bring a business or what it is about anyway.
If we zoom into what Enterprise 2.0 or applying Web 2.0 within a corporate environment really implies, there are three factors that are most significant and describe the core of Enterprise 2.0.
Jean-Pascal is the founder of 3org Conseil, an information management and governance consultancy. He has over 15 years experience in the technology services industry and knowledge sharing in the governance, collaboration, culture change and information risk management fields. Perrein’s Information Management and Information technology experience enables him to take part in high level business consulting in different market sectors such as Telecommunications, Distribution (luxury and textile), banking, utilities, and tourism.
Connect with Jean-Pascal.
TODAY: Fri, April 28, 2017February2011