Monthly Archives: June 2012

My Latest Collection of Hotbot Collection of Articles on BIG Data, Data Mining, Predictive Modeling, and Visualization

 These get updated regularly.

My Latest Collection of Hotbot Collection of Articles on BIG Data, Data Mining, Predictive Modeling, and Visualization is

An accompanying news aggregation for Analytics Jobs, Analytics Training, and Analytics Contracts is

Most Important and Powerful Factor that Predicts Your Market Opportunities

Do not become food for your competition: 

Ganges Shark Picture

So what are you supposed to do?

This picture of ‘Strategic Plan’ is just to indicate that your strategic plan should incorporate this most important predictive factor, a factor among many that contribute to the dynamic behavior of your consumers so that you are accordingly retooling your organization or redesigning the organization steadily towards incorporation of that most important factor

We know demographics is a sure fire way of predicting the future that will bring enormous changes in a society.  We know geo-politico-economic tectonic changes due to war, new find of mineral/oil, or significant changes in political philosophy will also bring a definite change and changes that would be enormous, some will bring destruction and some more opportunities for common people. 

In the following I want to bring one important sure-fire factor that every company should be watchful and factor into their strategic marketing vision for the next 5 years type highly confidential internal document.  This will include mergers and acquisitions, specific technology investments and upgrades, and special technology teams and smart engineers to build the organization.

This factor is stable and its effects are highly predictable yet more faster in bringing change in market place compared to demographics.

This destroyed already lot of mom and pop stores or gave new opportunities for those that adopted this in US and provided new opportunities for faster and better (read competitiveness) alignment of globalization concepts that were used by mom and pop stores around the globe.

This is not just for mom and pop stores but also for every organization small or big.

This factor is “lifestyle changes and associated market dynamics or market dynamics with associated life style changes.”.  While there are many technological innovations that will change how the world operates influencing the daily lifestyle activities of consumers in the next 20 years, we will concentrate only on the convergence of audio/video/phone/interactions/news/education/health/security/money transactions and everything in between that will become more and more digitized, where the common denominator will become 0s and 1s; The picture on the right captures its essence.

Take the case of print.


NewYork Times retooled itself so well it is now becoming a world wide daily news paper, all in digital form! – I finally became a subscriber of this great news paper. However, three years back, has pointed out the dire situations of the following 10 that are on the edge.

1. The Philadelphia Daily News

2. The Minneapolis Star Tribune
3. The Miami Herald
4. The Detroit News
5. The Boston Globe
6. The San Francisco Chronicle
7. The Chicago Sun-Times
8. The New York Daily News
9. The Fort Worth Star-Telegram
10. The Cleveland Plain Dealer


The cable and telephone industry is in a dire situation, and the great successors are going to be those who retool for the coming challenges of consumer lifestyle changes. The world is moving in the direction of everything cellular (oh, let me say no wires, if not cellular) including the audio video delivery and audio video interactions, a traditional area of cable companies with its thick coaxial cables with its inclusion of telephone services.  But it is not enough.

– Verizon is lot more retooled compared to many companies on the telephone side, while comcast is much better tooled and invested in diverse portfolio compared to any other cable company.

Watch out cablers or Telcos (whichever is dominating in your portfolio)

  1. Armstrong Group of Companies
  2. AT&T
  3. Atlantic Broadband
  4. Blue Ridge Communications
  5. Bresnan Communications
  6. Bright House Networks
  7. Broadstripe
  8. Buckeye Cable System
  9. Cableone
  10. Cablevision Systems
  11. Cebridge Connection
  12. Charter Communications (bankrupt)
  13. Comcast Cable Communications
  14. Cox Communications
  15. Galaxy Communications
  16. General Communications
  17. GCI
  18. Insight Communications
  19. Knology Holdings
  20. Mediacom LLC
  21. Midcontinent Media Inc.
  22. RCN Corp.
  23. Road Runner
  24. Service Electric Cable Television
  25. Tele-Media Corporation
  26. Time Warner Cable
  27. U.S. Cable Corp.
  28. WideOpenWest
Circa 1910 – Eastman Kodak – Rochester

We all know the status of Eastman Kodak, once a reigning supreme film maker and imaging products and services, since they didn’t get it.

Miniaturization, Broadband, and Smartphone/device – The Mobile Technology

In technology, my childhood hero, Sony is in distress now, because they are not able to be as relevant as Samsung and other smart phone companies, which carries everything from camera to HD video recorder to HD AV player.  For a different reason, but due to the same effect of smart phone,  market competition, and broadband proliferation, Bestbuy is getting beaten up.

To put it politely in this competitive world, it is disheartening to see that they do not get it.  Is it difficult to understand these problems five years before and accordingly change to be relevant to consumers.  So why doesn’t it happen.

Mega lifestyle changes is not easy to predict completely with complete certainty, but none the less predictable.  So if you are a corporation, the first and foremost thing you need to have is courage to change, to change to be relevant to the consumers whom you are targeting and serving and prelude to that is the insight as to what will change.  The power and variety that will happen will tame the picture on the right, which nicely provides indications of the trend.

The impact of broadband is still in the early stages, if we have to believe in the things that are coming.  Every one of that is the globalization of the quality and quantity of diverse and yet standardized products and services that will be made available to all parts of the world and US stands to gain in this as a leader and yet it is a phenomenal opportunities for people who are not yet exposed to the global market.

Imagine the following, which are not anything new and quite possibly most of you have heard of this.  But then, are you factoring the convergence of broadband, mobile, AVI into your five year strategy plan? Perhaps, these are not in 20 years, but 80% of the revenue transformation will happen in the 20% of the market in the next 50 years.  Some interesting projections:

– Not just newspapers will become international, but also the movie theaters
– the first line security is not your local police, but some one monitoring 10,000 miles away
– Your, not yet born children will be educated by the best people from around the world and schools will become interaction centers for daily life
– More and more jobs will be done from home (and that may not translate that we will have lot more personal time for many of us as the global competition for labor will become intense; this I call it the effect of virtual neighborhood
– There will only be digital money; no need to carry any printed money – wow that would be great; the sooner the better, though this will not assure zero fraud
– The following are 10 things that will change your environment and your lifestyle according to futurist: Richard Worzel, some are more directly and immediately related to my main objective of this post.  To keep it 10 I will replace some of them in brackets or identifying the attribute with more immediate happenings.
  * Everyday robots (automatically driven cars, for example)
  * Dramatic increases in productivity (suddenly a small villager in India or Philippines or China is able to contribute to the world economy like never before)
  * More ascension of women (some countries are leaders here, but with out this the mega lifestyle change will not happen in a shorter time span)
  * Healthcare revolution (personalized medicine, complete genome evaluation for new borns or sick people, continuous health monitoring with embedded measurement monitors, and complete 360 degrees health records management)
  * Transhumanism (start of more bionic human)
  * Critical economic uncertainties (raising foundations for more multilateral economic participation, contribution, and control of economic uncertainties)
  * Growing political and social turmoil (Expanding individual freedom; simple number of people free from suppressive political shackles)
  * Acceleration of climate change (Recycling and
  * The energy revolution
  * The purpose of life (Becoming more responsible citizens – Going green and more respectful of environment, animals, and plants)

Ganges Shark PictureOk, this has been a  long article to bring your attention, to the importance of lifestyle changes and associated market dynamics.  The key point, in terms of your action is the following.   If your ‘Strategic Plan – 2012-2016’ does not factor the lifestyle changes and market dynamics then not only your food is eaten away but also your organization is likely to be eaten away by competitive sharks.

PS:  the references to pictures and animated gifs are available as part of the image information.  Even though animated shark is to dramatize the communication, I love them.  The sharks (and hence competition) make our life very interesting and fun.

From Data Monster & Insight Monster

When to Use What Graph? Trainer’s notes on Information Visuals – Part 1

One chart is worth 1000 images! – Michael Fink: Production Manager, Google Chart Tools

The challenge:
There are so many attractive looking graphing types available in any standard software there is a tendency to forget the main point of a graph for a given data set and the hypothesis that an analyst has in his mind, that is, to get effective, full, and detailed representation of data in a picture with out distorting the true summary of the data.

The true chart is so artfully captured, it will yield any minute differentiation of the data elements (details) if needed for the purpose of chart, and yet provide the summary (structure) of overall data structure of the data. It is true that it takes a good training in reading details vs. details. When one is in the early stage of learning data analysis, selecting the right graph looks daunting. 

What is a good graph?
At the core of a graph, truly represented details and truly represented structure are the two basic requirements of a good graph.  While advanced graphical objects may have any number of layers attached to it making it more and more richer, these two layers are fundamental for communicating the power of a graph via the claim ‘a picture is worth thousand words’.

I use the word ‘true’ more often because from the point of view of cliche ‘lies, damn lies, and statistics’ one can distort the visual details and pass through the eyes of a commoner (interpret it as populace reading, not a unsophisticated) while it will not pass through the hawkish eyes of an analyst.

The effective representation of data is all about the true details of data, and true summary is all about the hypothesis that is being illuminated; sometimes, it could be the hidden agenda, be aware and that is indeed is the reason why people easily being slipped into with out their own volition.  Some graphs give more weight to the structure and some gives more weight to the details.  Some times, we can combine these too together to represent the details to the maximum and yet the summary also in the same graph, using two layers of representation. 

With that basic requirement let us see the basic list of the following graphical types and what hypothesis each one represents.

Bar graph is the most basic of all the graphs, with a structure for comparison and pie graph is the most basic of all graphs for probability distribution. Scatter plot is the most basic of all the graphics with only details.

Bar Graph:

In its most simplest form, it is just a simple pairwise comparison between two classes(groups/categories) of a variable.  The comparison can be unordered measurement of a variable like height comparison between male vs. female, or ordered as in annual income of college educated vs. not college educated, or time related as in weight loss before the experiment vs. after the experiment.  So remember the following two things for this: (1) a measurement (height or income or weight loss), (2) comparison classes (male/female or college educate/not college educated or before the experiment/after the experiment).

Question:  Is there something called best representation of differences, amount of differences getting the attention as the core message of the bar graph?  There are many visual ques one can bring to represent the details here. Give one such visual detail.

You can see how it extends to multple comparison when we have IQ comparison of 10 year old children in Northest, Southeast, North, Northwest, Pacific, Midwest, Southwest regions, an unordered collection of classes.  All types of comparisons are hidden in one single graph – every possible comparisions one can imagine.  A visual statement, a statement based on visual details of the differences for any comparison set, can always be made using this graph.  However, note that for such statement to have statistical validity, one has to go through more advanced statistical computations which is a different topic for now.

These multiple comparisions can be made whether those classes (groups/categories/segments) are unordered as mentioned above, or ordered as in income comparison among high school not-completed, high school completed, college educated, and graduates and above or as in weight loss in one month, one month to 3 months, 4 months to 6 months, 6 months to 12 months, and above 12 months with the placebo group followed for the same durations.

Line graph:  What can we say regarding changes between nominal or ordered classes of a variable? where does the largest change, smallest change, middle of the road change happen? How much are those changes?

A line graph is a close extension of bar graph.  When there are too many of those classes (categories of the variable especially where ordered or time related variable is studied) then bar graph becomes too croweded.

The line graph quickly captures the following salient structural questions.  Is it systematically increasing, or decreasing or shows up with a hump or is it changing drastically in some section of the graph or at the latest points. 

Question: What kind of visual ques one can bring in in the graph that gets the attention to the key points of the structural questions?  This of one simple visual que.

Pie graph: 

Area graph:

X-y plot:

Top Analytics Blogs

The purpose of this note started with the idea of  identifying top analytics blogs for my students.  As I started writing, I found that the list is useful for practitioners too with a caution that it is a list that I found valuable information or fun information for my practice.  If a reader points out to additional aggregated site on ranked blogs or just a great blog I am more than happier to take a look at them and add them as necessary.

Analytics blogs are important sites because they provide very interesting views, aggregation of information from multiple sites, understanding diverse points of views, and latest views/news on data and data intelligence.  These are updated constantly and are available in real time as the information and views are put together by the professionals and aggregated from around the world.  The analytics community owe these professionals and their passion for providing such interesting and useful information, usually free.  Some times they may just ask you to become a member for the benefit of reading their sites.  It provides a venue for authors to broadcast their writings and consumers pick and choose whatever they want, whenever they need it. 

While trying to put together the list of useful blogs for practitioners as well as fast growing student population, I find two characteristics as being useful in identifying the top analytics blogs; methods and tools vs. domain specific application.

The vast diversity of analytics blogs are identified by domain of application called verticals and the other type is about methods of application which could be used in any vertical.  For example, an analytics blog which is focused on web analytics is a methods blog and it is applicable to any number of application domain/industry such as insurance, education, banking, credit cards, investment services, retail, CPG, energy, sports, auto industries.

And, then there are different methods blogs, such as mobile analytics, direct marketing analytics, risk analytics, big data analytics, sports analytics, text mining, voice and video analytics, social media analytics, …

Some times, an analytics blog is a highly specialized analytics blog for a specific industry or specific sub-set of an industry.  For example, sports analytics is a very different group and a sports analytics blog for baseball is very different compared to a sports analytics blog for football, vs. a sports analytics  blog for golf, because of terminology differences, the passion of the followers and the nuances of the game, and the different measurements that defines the game.  While it is still in the early stages, the education analytics using big data is very different compared to the urban/city analytics using big data. Then there are prominent software company hosted blogs and prominent consulting companies hosted blogs.  Some of the list below is partially tilted with these sidings, which I could not avoid. This is a start and it will get trimmed or fine tuned as we go along.

My purpose is not to impute any ranking of importance to these blogs.  The purpose is to put together a meta-note that aggregates best blogs, some of that may be referring to sites that have already been aggregating blogs as top 10 or top 100 list by some other process.

In the list below, there is no reason or ranking as to why one group of blogs is mentioned first vs. second, and so on.  This is a just a collection, again with out any imputation of ranking from my side.

The top 4 blogs that I value for students’ exposure are

KDnuggets collection of blog 
– – visualization guru
– – web analytics guru

To create your own views, you may see the details below.

Top Ten Web Analytics Blogs: July 2007 – Though this is 5 years old, it has some scientific method of rankning blogs along with a list of web analytics blogs chosen by that method for readers to familiarize the concepts and to follow analytics details.

KDnuggets collection of blogs –, one of the pioneering data mining community blog list provides an amalgamation of blogs that are prominent across verticals and methods.  It is a honor to be in their list, by the virtue of their contribution from the beginning when data mining was not even known outside of a small list of practitioners. There is no discussion as to how or why these are selected.  None the less, a great diverse collection

– – visualization guru – this has a large collection of blogs that comes with a sub-title, “connecting you to people, products, & ideas from SAS”. – this is a collection bloggers on R from around the world, specifically on R programming language. and  are two visualization sites, I visit, more so the first one for traditional visualizations while second one provides views of unusual insights and visual dynamics for big-data. ; there is a collection of authors contributing to different topics in information management, including analytics which is contributed by Steve Miller,  a leader in proprietary computational software,  I do not use mathematica but this list will be incomplete with out mentioning this.  At this stage this is a fun site for me. – a collection of top 10 social media blogs, not weighted by analytics – top 10 mobile media blogs, not weighted by analytics.

As a general collection of blogs, I visit blogs from list for any possible expansion of ideas or point of views; this site serves my purposes.

Have fun.

Footnote1:  A complementary reading for top 10 web analytics blogs is, on why the ranking method used may not be the best

From Data Monster & Insight Monster

Some Top Visualizaion Training Videos – The Basics, Advanced, Interactivity, and Real Time

I love these original thought leaders and contributors.

These are privileged videos which were paid for and presented in person and you can spend your time at home watching this for free!  All in a matter of 3 hours of focused time and a refreshing possibilities in your thought process.  Have fun.   I did.

The basics of why, what, how of designing visualizations is 

One of the important topics is interactivity of graphical objects; which are captured in the following two presentations.

Personally, I like D3 software.   Here is a take on introduction to D3

The Future of Interactive Graphics in R – A Joint Visualization and UseR Meetup

Information Visualization for Knowledge Discovery.  

Now combine the above in a paradigm like the following which is shown in simple ways for a very simple process.  This is wild and it is giving me creative ideas!

 Before, I close with the next interesting video, additionally you may want to read the following post on visualization tools and some top visualization sites that I plan to use it for Dashboard 3.0 


A real time visualization of Reddit

From Data Monster & Insight Monster

Leadership Qualities for Analytics Department (Leaders) to Lead An Organization

With the current attention and the t(h)rust in analytics, the key challenge analytics leaders still face is ‘gut based’ decisions by the management under daily pressures, while being expected of them to defend their daily analytics work in a fast and furious ways. These situations become lot more common the more stressful the decision points are and also leads to distrust in analytics to deliver on time for big decisions.  One might wonder that those are the situations senior management has to be careful and be pushy to use analytics, but then there is always  the practical world that challenges the time priorities.

Also, I get questions as to how to get others listen to you that analytics works and get right time to grow in side the organization.

The challenge is often people end up quoting rock stars of analytics or some popular book to say casually the importance of analytics and that is fine as a conversation starter.

However, that will not be enough for executives who are under stress every day and also because different organizations are at different stages of analytics and the management commitments are at different levels. 

To get people buy-in on a day in day out basis it is very important to show what measurements matter, and why they matter, when they matter, and how they matters for every business question, using the right data and building credibility on a daily basis. 

Also, there are systematic ways of understanding and applying leadership qualities in gaining confidence and followers, making change management easier, and getting the science of analytics helping lead your organization. 

Leadership is getting people to follow willingly.  These are every day’s intentions not because we want to become leaders but because it is the joy of living well.  The title leader is just a token of appreciation by others but the bigger benefit is the joy of living well.  The leadership principles are powerful and applicable irrespective of what one does.  However, I want to bring them together here to get your organization to be led by the process of analysis, process of information strategy, process of decisions that are supported by metrics.

The well understood traits of leadership qualities are:

  • Vision:
    • The first and foremost is the vision, a well defined and crisply stated vision which has only one measure  that tells you whether you are progressing in your vision or not on a daily basis, on an hourly basis.  This provides the platform for focus and provides the meaning to your life.  In our case this is getting the organization explain itself every day whether it is in control of its vision or not. Whatever you think or do should be related to your vision, nothing less, and so much so that the daily life becomes an art.
    • Be precise so that the slogan of your department fully reflects the companies vision,  market trends, and data intelligence specificity
    • We will call this data intelligence vision
  • Integrity:
    •  This is important because this is the only thing that will dispel the cliche ‘lies, damn lies, and statistics’.  A well trained statistician can easily see why it is a cliche and how in common usage people use statistics as a prop to support their ill-founded hypotheses and univariate based conclusions, with out proper control and test comparisons.
    • The way to get your team or your management to see the truth is relentless pursuit of statistical truth and consistent, relevant, and accurate (drive a CAR) interpretation of data and summary.  Some times, you end up in a difficult situation because you make a mistake in calculation or you did not see the right data assumption in the hurry burry of day to day life.  But never give untruthful reasons or interpretations, in the extreme, and not even in the cryptic messages.  For example, I find one of the crutch statisticians resort to is ‘not enough sample’.  This is not an issue of extreme untruthfulness, but do not slip into that quickly.  At least you can say how the confidence interval is becoming wider because of the sample size; that is a better way of communicating the challenges of limited data.
    • Always provide a list of assumptions used and addressed in defining measurements, inclusion/exclusion conditions in using the data, and supporting all your outputs with simple segmenting/grouping, ranking, weighting, appropriate sampling, and probabilities in the end, however sophisticated the methodology is.  I will bring out a template for standardized outputs managers need to ask for complete enumeration of quality of work.
  • Dedication
    • Dedicate your time and energy to be truthful about how the right metrics, right, data, right analysis, all the times, explain the challenges of correctly interpreting the statistical nature of our practical world.  The challenge here is most often getting the right data. For every company in every industry there is only one measurement that matters; it is not the money, even though it is tempting to say.  Commit to interpret and work out the complete processes  for any angle to any business problem for the company, day in day out, on the basis of that one most important measurement.  It reminds me the classic quote in Moneyball movie,”People do not understand. Your goals shouldn’t be buying players; your goal should be buying wins; to buy wins you should be buying runs – Peter Brand”
    • When you do this people want to follow you; no questions, because you support yourself the vision that you set up for your organization and your relentless pursuit of that one goal captured in your most important measure, sets clarity for others. Others in the company are behind you to dedicating their time and energy.
  • Magnanimity:
    • Give credit where it is due. Spread good words about people who are genuinely working hard dedicating their time and energy with integrity, in support of the data intelligence vision.  Alternatively, a good leader takes personal responsibility for failures; wow that means a lot of courage.  You are putting yourselves in a situation for people to naturally navigate to you as they feel comfortable in working for you because they get the best of the world; all the time and energy with clear directions from some one whom you can trust and they will get their rewards too.
  • Humility
    • Humility comes because you genuinely feel everyone has the same capacity and capability to be as good as you are.  It elevates every one’s initial stock of self-worthiness instantly and you never talk down any one or condescending with any one.  There is no better way to establish teams.  This is the foundation of believing in others and getting them up to you with daily guidance and teaching.  This is about open heart.  What is the indicator of open-mindedness?
  • Openness
    •  Great leaders have phenomenal listening quality and unless you are open minded you can not have that intense listening when people try to communicate.  I shall say that communication includes both hard verbal and soft body languages with empathy and go into conversations with no bias and conclusions
    • Openness builds trust and confidence in the leader
    • Surprised will you be, that when you listen very intensively you will actually get some interesting solutions with out even discussing the topic, because there is no noise in the exchange and this phenomenal focus on the process of solving problems keeps opening various channels of intuition and the person who communicates with may not even realize some of the inter-connectedness of the concepts that raises the quality of solutions. So what solves problems?
  • Creativity
    • The cliche here is ‘think different’.  But it does not help how to create, how to build the process of creativity in an intentional way. So I will say,  “Creativity is the ability to keep seeing all the resources and their properties and seeing alternatives as possibilities”.   This can be systematically built with in an organization if you follow the following.  Concentrate on the process, problems(requirements), devoid of people, biases, time, and space. Again it is demanding no distraction and seeing the most simplified form of the requirements in its purest form. Alternatively concentrate on requirements, concentrate on the pain points, concentrate on the limited resources – especially time, in our busy world. Some where one has to make the call on go-no go, yes/no, pull in resource or not, to be meaningful for our clients – internal or external- and properly negotiate time and resources. 
    • The author Tina Seelig says this is coming together of the internal combustion engine of creative energy and the external world of possibilities, where the fire in the belly is inherently available by culture(E) and attitude(I), supported by Habitat(E) and Imagination (I), and drawing energy and strength from Resources(I) and Knowledge(E).  The E means external possibilities and I means internal attributes.  This is a wonderful summary of external support system and internal nucleus, captured brilliantly by a double mobius strip.  
    • See all in this privileged short and a very effective free video presentation from Google.
    • So what helps to be decisive and correct in those calls?
  • Fairness
    • Deal with people in a fair and just manner.  Believe in the truth that every one has the same rights as oneself is.  Making decisions and the inevitable judgements are part of the job a leader end up doing every day.   But it requires having all the facts and right weights to each of the fact.  This is the one that makes people dedicate their time and energy and become loyal, because it democratizes the opportunities and challenges. 
    • Fairness not only to clients, which is our ultimate touchstone, but also to the resources and the joy of contribution.
    • This one more and more elevates a person to an attributed status as a leader, even when you make decisions which may look unusual and directionally a very strong one, a quality called, assertive.   Assertive – what?
  • Assertiveness
    • Assertiveness gives credence to your commitment and your full support of your own decision.  It is not aggressiveness but it is the depth of belief in what you decide.  Of course all decisions are not yes or no even though that is the result of a decision.  One way to carefully navigate the uncertain states of outputs of such decision is to introduce pilot programs and see incrementally the evidence needed to get more and more people to commit to initiatives and projects.  There are many other ways to get every one incrementally commit on assertiveness. All the other leadership qualities help too.  However, remember the proof of the pudding is ‘growth’ of the organization.
    • All models are not wrong; all models are approximations of truth; the question is know how much is approximated to believe in yourself and also to believe in your analytics
  • Sense of humor
    • Enjoy every day life and see the challenges of keeping up with all the above wonderful attributes with a sense of humor, because the daily variations that come as challenges can create some extreme interpretation and adherence to the above attributes.  So, know the fun of relaxing with humor.  This balances all the up-tight behavior that are likely to be popping up because of the strong commitment you have on the above list.  
    • Again it is not loosening up but seeing some funny side to the above qualities in a practical world.  This is difficult because of the seriousness of your vision and mission but let it pass with some smiling moments, as they say pass and get back to the focus on the process and requirements.

These are difficult for every one of us because every one is in some stage of growth and the practical world brings out all the humanity can offer to this person and having integrity and truthfulness to all individuals for all the data intelligence projects in a fair and just way irrespective of how they view you or treat you is the ultimate challenge every one goes through and only few survive that tough challenge. This is true of whatever field you are in.  We are bringing these concepts from the point of view how to get analytics in the driver seat.

    If you follow these in building your team, your department, your division, your followers will find you as a valuable person whom they can trust, and dedicate their life.   Every moment of your difficulty in developing and implementing data intelligence challenges will be cleared because the foundations of your actions are essentially spiritually driven by the above practical directions.

    Every project, every metric, every calculation, every justification, every communication, every appreciation, and every decision builds one more day of success for you and your followers.  The change management happens lot more easily and analytics is in the driving seat.

    Finally, I invite you to see the movie Moneyball and learn these in a true story where you can see the actual change management problems, and how Billy Beans handles them, dramatized for the movie, though, the author says.

    Do you believe this will help getting analytics lead your organization?  I do.  Write me your views.


    Accenture in its recent publication regarding leadership to lead analytically,, it brought out the following as the key dimenions to lead using analytics, from a macro group culture perspective.

    “Accenture research and experience shows that a culture of analytics can be built by focusing on five elements:

    • Respect for data. Organizations with analytical cultures demonstrate a profound respect for data and fact-based decision-making.

    • Pragmatic decision-making. On the other hand, organizations with analytical cultures know the limits of data and do not get stuck in “analysis paralysis.”
    • Drive to optimize. An analytical organization is fundamentally curious—about what others are doing in the market, about performance patterns and root causes, about new and better ways to do things.

    • Collaboration and transparency. An analytical culture is marked by collaboration and information sharing across organizational boundaries.

    • Rewards for analytics. Individuals are recognized and rewarded for their analytical capability, including not only the quality of analyses and insights, but also the breakthrough business results achieved by putting them into action.

    Analytical leaders are passionate about managing by fact. They insist on objective evidence before making decisions, they are agents of change—and they recognize that having data is not enough. They must put it to work in pursuit of business performance and competitive positioning.”

    From Data Monster & Insight Monster

    What are edges of Prediction? Areas to Be Careful About Hypes

    I am trying to be honest about my challenges in consumer marketing.  These questions help me understand and prioritize how to align the limited  resources we all have.

    * Embedded predictive modeling with in a database architecture – all modeling is wrong! and it only makes it more harder not to beat this cliche; embedded modeling or embedded catastrophe.  While embedded modeling is very useful, how do we make sure that it does not end up as embedded catastrophe.  Why hadoop structures are more reliable?  The other way of seeing this is that the Type I and Type II errors are mostly likely not that robust in the embedded modeling.

    * Consumer (micro level) centric media mix optimization – Technology makes this happen. The traditional media mix optimization is budget centric optimization; none the less a very useful tool and what does it mean to search for better methods that would be at least as good as budget centric optimization and better?  Also, which is more truer – multichannel marketing or media mix optimization or media attribution? are they same or one is more truer of than the other?

    * Social media analytics – Modeling privacy attributes? – can you guess why this would happen?

    * Black box modeling or black magic? The death of modeling paradigms that tout complete automation – Can you guess what these are?

    * The end and be all of media platforms? Why analytics has not caught on this?

    * What trends are important when consumers value more and more going green?

    * Hadoop (apache software project – developed initially by Yahoo) that leverages massively distributed computing architectures for structured and unstructured data as they are captured using reduction tools such as MapReduce to digest and summarize for visualization, ranking, and distributed decision, and also make it available for further lagged processing and create insights

    * The pervasiveness of Bayesian methods for automated customer intelligence and moderated automated business intelligence

    * Consumer (micro level) owned marketing intelligence or consumer owned privacy; can an individual consumer own and market his preferences – a revolution ready to be born if only it is done right – a marketing platform opportunity

    * Social media analytics – Modeling privacy attributes? – can you guess why this would happen?

    * Black box modeling or black magic? The death of modeling paradigms that tout complete automation – Can you guess which ones are likely to become sidelined – methodologies will not die? 

    * What is the most valuable media platforms for advertisers? Can you guess some of the big platform failures as of today and why they fail?  The ultimate platform is the one that is flexible with right analytics structures attached to it with out violating the  privacy concerns.  Do you know why structurally all plaforms are not the right ones. If agreed what is the key component of

    * What trends are important when consumers value and move more and more in the direction of going green?

    * Hadoop that leverages massively distributed database servers and MapReduce is one of the data reduction tools to digest and summarize as the data comes in for visualization and real time prioritizing and routing, and also load for further processing.  How do commoditize hadoop clusters for a web site?

    * The pervasiveness of Bayesian methods for automated customer intelligence and moderated automated business intelligence

    The last one is – figure this out.  It is such an opportunity and I am looking for partners to work on this.

     * Consumer owned marketing intelligence or consumer owned privacy; can an individual consumer own and market his preferences – a revolution ready to be born if only it is done right – a marketing platform opportunity

    Predictive Modelers are like, all knowing, OB GYN KENOBI – Predictive Models are Fair Game, but not Data !

    Since it is coming from comedy central, I thought this funny video could cheer us up any time.

    Colbert Report talks about predictive analytics and Target.  Colbert

    …”Now they know that you know that they know” – still they may not be sensitive. Not so, Colbert Report gives ideas on how to mix the sensitive messages as casual messages. 

    …Knowledge means profits

    …Barcode Mitzvah

    … Lots of talking about predictive science and imagine what could be happening if there is a ‘ … oops by mistake we used the data for selection’. That is why I say prediction is fair game but not data.

    That is you can use intelligence to figure out with the data, but do  not use the data directly to target. Big data is going to make predictions more accurate, if we do not fall for easy traps of everything becoming significant. 

    More by NY Times:

    From Data Monster & Insight Monster