Big Data, Good Information and A Way For You To Use It

In this article I want to explore 3 connected ideas. The first is about big data, the phenomenon that now makes available enormous, staggering, volumes of information almost instantaneously. The second is a condition that says information already known to us can limit how big data can be used because there are other types of knowing to expand thinking. And the third idea is the sum of the first two: that the intersection of big data and a different way of seeing information means a model must be designed to utilize large amounts of validated information in a reasonable way.

So the era of big data is here. Imagine Niagara Falls and the millions of gallons of water that shoot over the precipice virtually every minute. That’s the scale of information we envision when thinking of the amount of data we can reach out and grab—or in some cases is pushed to us—everyday.  It would be impossible to know it all. But one benefit of massive volume is, when looked at through a certain lens, we have an opportunity to connect seemingly unrelated bits of data and discover trends, make predictions, even pre-position products and services long before we click, point or touch. It’s the compilation of colossal amounts of data that presents a challenge. How do we pluck just the right information we need from this torrent of bits?  This is the difficulty with information management in the era of big data; it’s like trying to take a sip from a fire hose. Our need is not to get information it’s how to get just the right content to help us work with more accurate and insightful facts and, smarter and faster?

Clearly then we see how working out a process to employ big data and make quality business decisions is difficult. Furthermore consider this context, our second condition:

“There are known knowns” began an answer to a question at a US Department of Defense News Briefing made by Donald Rumsfeld while serving as United States Secretary of Defense in February 2002. Actually, here’s the whole tortured phrase… “there are known knowns; there are things we know that we know. There are known unknowns; that is to say, there are things that we now know we don’t know. But there are also unknown unknowns – there are things we do not know we don’t know.” Though it may seem convoluted it is a “brilliant distillation of a quite complex matter,” said Mark Steyn, a Canadian columnist and echoed by many others, even legions of his detractors.

While good information on its own is valuable its utility when combined with other data to discover other, perhaps new data and still newer meanings is really profound. Sometimes the information is known and we need to fasten it in context, other times we don’t know there is… and what is…  trustworthy information but have to discover it; and more abstract yet, as unknowns the potential of useful but opaque information demands we peer into the future and ask ‘what if’ and proceed to manufacture information on (hopefully intelligent and intuitively perceptive) speculation.

If you’re in the business of solving problems—and who isn’t really—you’ll need an information life cycle model to regard big data and the ‘knowns issue’ to manage a collection of information for maximum use. And beware; too much data without vetting and affirmation, means you might miss the really important stuff, an effect that keeps security services awake at night. And therein lies the third concern of massive information management.

By summary then, we face three elements in our quest to make big data work for organizations:

    1. Gathering information factoring the effect gained when combinations of content reveal even newer more, newer, meaningful data
    2. Respecting knowns and unknowns as  fact and as potential ‘black swans’ (an unpredictable or unforeseen event, typically one with extreme consequences) that can and will skew results if not discovered prior or during information capture or application
    3. Culling the really useful information or data—those bits directly related to the problem at hand—from the gargantuan amount of information flying about making it accessible, contextual and changeable.

Here’s a model than might help us slow down a bit, turn down the faucet and cull out know information and potentially new content when big data offers additional tonnage of content.

InfoMgmt

The flow chart illustrates how information would be categorically organized; a model for the standardization of an information life cycle in big data world.

Ultimately culling useful information from an almost limitless stream comes down to energy, resourcefulness and commitment. When building a learning course for example, your subject matter experts deliver very specific information as they must do. However, is there other data in text, as visuals, in video that might provide a different way to see the information? Clarifying content by shifting the context just a little bit can often shine a light into corners formerly unseen. Whether one has the time or inclination to make the effort to go shopping for more information is dependent on time and budget, yes, however, when looking to make learning better and richer, drinking from the stream is often a task worth enduring. Creating metaphors mined from a combination of newly discovered information can improve the user experience—and enjoyment—like spinning a kaleidoscope and seeing new patterns. Using a model such as the one proposed might make such an effort more reasonable.

Advertisements

WITH REGARDS TO mLEARNING: A CASE IN POINT OR UP IN THE AIR

I promised a colleague a week ago I’d share an experience I had producing a mobile learning project for a major airline. So to him, I apologize for this installment being a bit late…think of it as slow 3G, OK?

A caveat. So many of us are involved in love affairs with the latest technologies sometimes we forget to brake our enthusiasm and learn after too much money and effort that the latest isn’t always the greatest. Case in point is the following true story—a real business event that occurred far back enough that if executed today would be much more powerful because of current technology. Of course, the outcome would not be substantially different because the desired results would still be the same.

A few years back as a VP for a large computer solutions company in New York, one of our account executives specializing in elearning opportunities had managed to seduce a notoriously reticent airline into considering change from their mostly instructor led training to online education. Their target cohort was employees who worked, in their jargon, ‘above the wing.’ This title refers to anyone whose employment had nothing to do with aircraft, maintenance, baggage handling, etc. Instead—and most important to our story—these were the folks customers dealt with at the airport, at check in and on the concourse; ticket agents, gate agents and personnel, support staff and customer service agents assigned at two of their New York City airport locations.

This was, and actually still is, a young airline; established and branded for quality service and a unique series of amenities on their aircraft. Moreover, their preferred hires were/are youthful, adaptive, and enthusiastic. These trainees had the right stuff but lacked knowledge of procedures, policies and in some case behavioral insights into how to deal with all sorts of customers. They were also paid in ‘prestige’ dollars—not too far above minimum wage.

Training on the Fly
After hiring, a cohort of at least thirty trainees would be exposed to a minimum of 45 days in classrooms at their HQ not far from one of the airports. The training and the trainers, based on our observations, were good and often better, effective at conveying all the obvious information and many of the nuances of operating in a regulated environment with the general public. After passing a series of qualifying tests trainees got their uniforms and were sent out to the concourses and ticketing stations for field experience.

So far, so good.
After a week to ten days working in the real environment, meeting all sorts of challenges for which the airline thought they had been prepared, the drop-out rate—that is the number of resignations, topped 35% and sometimes upward of 40%. Extrapolating the training costs, the quant’s figured each loss was worth about $30,000. Each. Yikes. At a minimum, each training class represented a loss of around $300,000. What to do?

To their credit, the trainers devised an exit instrument asking each drop out specifically why they were leaving. In addition, before starting the next series of hires, managers spent much more time at the airport observing the activities of each trainee.

The results were brutal.
Though trainers thought they were preparing new hires to be self-sufficient and make good decisions, they discovered something unusual. While, almost to a person, trainees knew policies and procedures, they were paralyzed when situations veered away from the typical. For instance; while they could modify ticketing and even handle families needing special requirements, passengers who needed to make late changes to their itineraries and other point of attack problems, when a real crisis arose—for which neither they NOR THEIR SEASONED COLLEAGUES had been formally trained, they panicked. In those situations where a resolution came about it was because someone had learned through trial and error, ways to handle the challenge. Realizing no one can be trained to handle every type of emergency; nevertheless, without a substantial set of guidelines the organization was placing too much responsibility on inexperienced…mostly new trainees. Faced with too many nail-biting situations…and realizing neither the romance of air travel nor the respect they had anticipated with the uniform hardly balanced out the anxiety, abuse and low wages, there were substantial resignations.

This was the situation uncovered by my colleague. He also realized that the airline had no real solution—not one that was economically viable. Trainers recognized, to their credit, training had to change in a significant way. But how?

Here is what we proposed
Those parts of the training that worked well, like procedures, regulatory issues, basic airline operations and the roles and tasks for each position should remain in place. However, the time needed to accomplish competence, especially with a new manual and meaningful assessments we would design together, could be reduced if we migrated much of the rote material and built it online. This component would be replete with simulations and scenarios that would build more lifelike experiences into training EARLIER in the process. This would accomplish specific goals; transfer information for use in nominal situations and then prepare trainees for some of the real life challenges they would face on the job. In addition, invite those less committed to bolt before too many training dollars were exhausted. The online experiences would be reinforced with classroom role-plays that were frighteningly realistic. I know…I wrote them.

However, this was still not enough. In challenging situations, not so atypical of life on the concourse, no one could be expected to rise from panic with Zen-like tranquility, and resolve every issue. No, we figured, above the wing personnel needed the kind of manual pilots had when systems were not, shall we say, cooperating.

So we devised as part of the new manual and online learning, a smart help feature with plain language key word searches wherever possible. Using their Palm Pilots (I told you the technology was ‘old), which held the manuals and the full course in memory, trainees and experienced personnel could get immediate answers when called for by keying in simple phrases. In addition, we configured it to learn—so that when a new situation arose it could be posted and all above the wing personnel throughout the system could review it. What we found was that in more than 90% of the cases, some clever or talented employee had a viable answer. This would be added, after tagging, into the course for learning and smart help feature as well.

Here’s a real example
A young mother approaches the ticket counter at John F. Kennedy International Airport in New York as the late afternoon rush of business travelers started to crowd up the lines. She is pushing a stroller with an infant and has, by the hand, her 3-year-old daughter. She beseeches the ticket agent to move her seats so she can have easier access to the toilets. While the agent is reconfiguring the seating chart, he hears the mother say, “Honey, I told you not to eat that whole hot dog…now your tummy hurts…I know. Well as soon as we get done here, we’ll go get something to drink and go to the bathroom, OK?’ The child says, ‘But mommy I don’t feel so good now…” She then projectile vomits on her mother’s legs, the front of the ticket counter, and for good measure on the shoes of the businessman in the adjacent line.

Let’s just freeze this scene for a moment. Here are some things you should know, and that the gate agent had learned. Firstly, vomit is considered a hazardous substance. Hazmat regulations apply in all cases. Secondly, as we know all too well, the odor is a contagion that can set off ‘sympathetic’ reactions. Finally, the material was on not only the mother, but also airline property, the carpet and another customer.

What does this mean? Most importantly, the agent, even if he knew how to clean it up, and he had the implements to do so, was prevented by law from doing so. He knew that but what could he do? He whipped out his Palm and keyed in one word, “Vomit.” And unfurled before him was the entire procedure for handling such a case beginning with calling for security and a Hazmat team. Furthermore, it instructed him to close down the line, come out from behind the counter, and move everyone away from the affected area. Upon completing each operation, he checked it off the list. Thus, a record was generated of all actions taken. Also, it automatically sent the location of the event and alerted supervisory personnel. Clearly, this accomplished a huge gain in ameliorating a terrible situation, supported the agent preventing HIS panic, averted a larger catastrophe, and projected competence and professionalism manifested before the general public.

To sum up, by changing the training to an integrated education and knowledge management model resulting in just-in-time access to information through mobile learning, the airline not only began to resolve its somewhat informal emergency procedures, but was able, as new trainees were hired, to prepare them with a much more complete repertoire of real-life events to study. Finally, with real time access to help in emergencies, trainees had confidence in the procedures for those real life panic situations. When finally exposed to the concourse, trainees were better prepared for all situations. The retention rate held—only 12% dropped out.

Coda
A change in training management and budget considerations subsequently stalled the growth of the mLearning component. Interestingly, during the ensuing winter, freak snowstorms created havoc and this airline, with few procedures in place to manage a complex reshuffling of both equipment and personnel, could not sustain an angry public and government scrutiny. The CEO was let go, its once highly polished image and reputation for excellence in service tarnished (and— some say has never returned) and passengers loads shrank significantly for a long time. And, sitting in Fort Lauderdale International Airport waiting to see if my plane was one of the few, and last to leave, I watched in horror as the ticket and gate agents were forced to call airport security and then the police to keep order as passengers were panicking and personnel could not contain the chaos. Had our system been in place, a key word search for “grounded aircraft: storm”, would have directed a senior gate agent  to cordon off 3 lines for each of the New York bound flights, and begin to organize anxious passengers and further plan and communicate next steps.

I must admit I found it not a little bit satisfying. Oh, and even better, I did get the last flight out.

KNOWLEDGE MANAGEMENT IS A SIMPLE PROCESS ONCE THE RHETORIC IS REMOVED

The benefits of knowledge management (KM) are a monster value-add to any organization. Nevertheless, the more I learn how companies capture and leverage their intellectual property, the more disheartened I become. How could such a straightforward process for transferring information and learning become bogged down in dense MBA rhetoric taking what is essentially a simple idea and obfuscating it in layers of process and jargon? Some might think large enterprises require significant resources to carry forward a KM initiative. I’m not one of them.

A Flash History of Knowledge Management
I can remember lecturing graduate educators that schooling started when one man stood beneath a tree and told stories to his Grecian disciples who sought to learn. Then each would become a teacher and spread knowledge throughout the empire. Why is the oral tradition so different in the 21st century.  Certainly technology has made it even easier to move ideas with an immediacy not easily imagined in the not too distant past. Is, KM , as some claim become the fiefdom of experts with metricians and quants creating a lexicon and modality to which only they hold the password? If knowledge transfer has grown into a complex system, it’s only because the nature of organizations to pile on layers of management seeking justify the effort and deflect external examination has become endemic.

I know there are nuances and specifics necessary in many systems and it’s no different in KM. Nevertheless, KM is about smart enterprises discovering and sharing winning strategies and techniques to improve performances of many kinds. Creating a method to discover useful information that is ultimately accepted as knowledge, then storing it for easy retrieval and communicating how to access it, is far from the challenge some would have us believe.\So, if you indulge me a bit, I can deconstruct this business practice and translate it into plain language, offer basic guidelines for creating an effective and direct KM system and then release it to perform. Consider this KM for Dummies—No offense intended.

What is Knowledge Management
The ‘knowledge’ that we have internalized by experience or education is our ‘tacit’ knowledge. When we externalize it by communicating with others, our knowledge is made explicit. Explicit knowledge is what counts in knowledge management.

Where does knowledge come from if not tacit? Knowledge is a product of innovation and exploration. Usually this comes about when a problem looks intractable. Yet, within each organization—and perhaps inside all of us—there is a spark of genius that, meeting a challenge for which learning and experience has prepared us, yields a viable solution. These insights, concepts and experiences, when polished and vetted, tested and found to resolve the problem has enough value to be circulated. How this new found knowledge is expressed by an individual, discovered by the organization; how it is brought into a system where its application will have far-reaching effects is the management part.

Strategies & Practices
Knowledge management is a formal range of strategies and practices used by an organization to identify, create, verify, distribute, and enable adoption of insights and experiences. Once aggregated into a body of useful knowledge  the purpose of KM is to focus on organizational objectives such as improved performance, competitive advantage, innovation, the sharing of lessons learned, integration, and continuous improvement of the organization. Naturally, this demands a company commitment since these activities and will need creation, invention, and management.

It works like this:
There is a problem…

  1. An individual or group has a solution…but it requires testing and such a mechanism must be emplaced
  2. The solution is tested and found to be of further value and an important addition to an enterprise
  3. The solution is made available for distribution or dissemination…a location and method of retrieval requires development
  4. For those who encounter the problem, a solution has been developed and is now available…presuming the access system offers multiple ways to locate the information

Let’s not forget the climate for innovation and the dissemination of knowledge will only thrive when there is a culture of collaboration. Though an enterprise might have a slick mission statement and therefore a common mission, goal, or objective, sometimes we have to be taught how to effectively collaborate. But sharing valuable information throughout an organization so knowledge may be leveraged and intellectual property maximized is an achievement that often distinguishes winning organizations. Here’s the part where leadership steps up and says we believe in knowledge management and expect this initiative to yield an improvement of business.

Two Interlocking Parts
Managing knowledge effectively starts by identifying critical information that makes a ‘big difference’. Capturing and synthesizing new learning and ideas, and applying knowledge to make the best decisions, requires great communication and collaboration. If KM is about anything it’s using learning strategies and processes; methods, tools and techniques.

To benefit from a rich archive of proprietary intellectual property, designing the means of transferring, sharing, and ultimately disseminating knowledge is step one. Step two, retaining knowledge for the future and providing easy access closes the loop. The best part of a well-choreographed KM system is a problem solved once never needs to be solved again no matter where in the enterprise it subsequently appears.

This, in toto, is ‘knowledge management’. Done well, KM also provides so much fertile information it often generates a critical mass of ideas that, exposed to a large body of users, leads to further innovations—new ways of doing things company wide.

Here’s a working rubric that is but one way to define and assign responsibilities for making knowledge management work. The emphasis is on simple, direct communications with the goal to make the process less difficult than it need be. One caveat; there is often a need to bring in a consultant as a change agent to not only develop the systems needed to make KM work smoothly but to side-step internal politics that often obscure a clean shot at a great implementation.

Example:
After the introduction of a new financial product in the U.S., it became evident advisors could neither understand how it could benefit their prospects or client base, nor create an adequate story from which to describe its value. Therefore, it was undersold. However, in Brazil, a small team of advisors was having great success. Management wondered, “Is a cultural bias that would predispose local investors to accept this type of product, or, did the advisors have a key plan, language, and technique that invited interest, opportunity, and sales that could be imported back to the states.

Managers traveled eight thousand miles to the geo to discover what was driving this success.  They discovered that a three person sales team had devised language and selling scripts that communicated benefits with clarity. They were careful to make the discussion of value and risk not to dissimilar to others with which clients were familiar. Drawing comparisons for their prospects and clients allowed an easy introduction to the new product, meeting with fewer objections and faster acceptance.

The managers captured the processes and language, codified and tagged the elements and upon returning to headquarters archived them in the database. Only then was a company-wide communiqué released superseding the original collateral in favor of a selling guide that leveraged lessons learned for selling this product. Additional information and access to the ‘inventors’ was also part of the knowledge management strategy. The result was an enterprise wide uptick in the sales of this product.

Conclusion
What it boils down to is first seek internal successes before reinventing the wheel. If a problem or challenge occurs, it most likely has a solution created by someone who has already met that situation and created a response. Accessing the answer and using it, even if modified, as is sometimes the case, saves hundreds of hours, enormous effort, and financial resources. Let’s keep knowledge management simple; for the straighter a line to the answer the more confident users become and the utility of the system ingrained within the corporate way of conducting business.