|BIG Data Provokes … 4|
How a critical simple regression model became central in defining the strategic metric for Oakland A’s is available only for members of this site on request. You can join by clicking “JOIN THIS SITE” link on the top right corner of this sit.
In simple terms, strategic metric is one that captures the competitive essence (“core competency”) of
an organization’s uniqueness in its products/services that strategically connects with the ramping up of technology and lifestyle.
Organizations, will not reveal what is its strategic metric, many times. Some times, the sr. management may confuse strategic metric with KPI(Key Performance Indicator)s or KLI(Key Leading Indicator)s. It happens all the times. What you can not understand and measure can easily derail your organization.
For most part, the core essence of the organization does not change for changes in technology and lifestyle. The following example shows the how flexes in the technology and life style dynamics redefines the processes and yet the core essence still remains the same.
Xerox, had the purpose of document processing but in an ideal world of strategic senior management, the greatest email services should be owned and hosted by Xerox, and possibly owning Adobe Inc. At the least having a strategic relationship with Adobe. After all, software was dominating over hardware and technology was changing fast. Adobe and Secured email services should have been felt as the greatest opportunity changes in its field.
Technology and life style will change but not the core competency of an organization’s uniqueness.
According to “http://www.opexcanada.com/learning/keytoprofits-corecompetencies.html”,
- “An excellent, reliable process
- Synergy with customers & suppliers
- Specific technical intelligence – “know-how”
- Corporate culture
- Unique product development
- Branding, Marketing etc.”
Honda’s pervasive dominance in cars, snow blower, boats, jet engines, power generators, water pumps, truck engines, race car, lawn mover, motor cycles, and robots, all emanate from that single focus to be the best in building the best engines for a given set of parameters.
Does the above definition fit in for Apple?
For all said, Apple’s strategic metric, until Steve Jobs was at the helm was a measure of customer delivery for a specific segment of people who are willing to pay that extra money because they are looking for that elegant laptop which has the latest technology as a leader, as technology ramps up. That passion became as “computing technology or interaction technology rather than laptops” as technology was fast evolving leading Apple into the iPhone and iPads, now moving with iCloud, thereby bringing along the software side also to support its vision. His customer intimacy, delivery, and focus on the strategic metric was so strong, he famously said, he did not need focus groups to tell what the customer wanted!
While I am positive that the above understanding will continue to drive Apple for a long time and possibly forever, as long as we have Steve Job’s spirit in the organization, some times I wonder whether that customer connection and delivery, and maniac appreciation for quality is softened.
How about Google?
Google, started as a search company, seems to be evolving in the direction of enabler and user of search. Android OS, Chrome (despite its weakness with viruses), and its foray into autonomous vehicles are all pointing in that direction. For our exposition of today’s topic, it is nicely interesting and consistently fulfilling to see Google did not start building computers but just stopped with strategic partnerships with manufacturers.
What can we say about Facebook?
Facebook’s uniqueness is its consumer peer to peer interacting platform. Facebook treads very carefully leveraging the consumer P2P platform and privacy concerns.
After the initial challenges on monetizing its huge consumer base, it is trying to increase the per consumer advertisement revenue; does so successfully that the market is rewarding nicely in terms of its stock valuations. Facebook more than doubled its price in the last 12 months. Should Facebook introduce its own phone or have a strategic relationships across the board to keep its consumer interacting platform better built, monetizing on the platform, is being watched by analysts carefully. The evil twin of consumer interacting platform is privacy concerns of consumers. Any let up here can be devastating.
Oakland A’s knew its strategic metric and it is well discussed, researched, articulated by all because it became a love story for all the sports fans and analytics professionals. It is a great case study.
Here are the five questions one should be asking and learning from the point of view of Oakland A’s as to how strategic metric played a central role in redefining its operations.
Do you know your strategic metric?
In the next article I will bring out the relationship between Strategic metric, KPIs and KLIs. Here is something to munch. KPIs are given undue importance. You really need to think about KLIs, more than KPIs.
Another point about strategic metric is that these discussions are interestingly true for personal life, governmental policies, and in fact it is about the vision…
Automation of knowledge work is the highest minimum value creation worth around $5 Trillion.
However, you slice and dice, pure big data and analytics is based on 5% of this automation of knowledge work and it will be at least $250 billion in the next 13 years. Remember that most of which will have to be automated work and so the computer scientists and new breed of data scientists will be at the helm leveraging these developments dominating this huge market developments. So here is one more reason why statistics departments have to reorient itself where at least there is a track for the big data statistical analysis.
There is no better time in the history of the world and history of USA that will create million millionaires, if only USA becomes an …. place. Fill in the sensible world. Analytics is going to be a significant component of it.
So, what is Web 3.0? Crisply,
Exercise: See how Brett King: BANK 3.0: Why Banking is No Longer Somewhere You Go But Something You Do fit, fits in WEB 3.0
S I R C L (soundex ~ circle)
You do not need Ph.D to do data analytics. Of course. But do not think correlations is all we need for all problems in life. The kangaroo court will decide and hold it as truth whatever it sees and interprets, until the predators like, lions, tigers, and even hyenas emerge in the field – the complex opportunities. In the same token, there are so many problems and opportunities for which you do not need a Ph.D. Let us not get stuck with letters here.
The news is that 1,000 Ph.Ds did not answer this correctly. The reference to that claim is here: http://predictive-models.blogspot.com/2012/07/even-math-professors-fail-in-this.html.
You have to see “Price is Right” to figure out how frequently general population picked it correctly.
If you do not know how to interpret the type I and type II errors you are a dumb data scientist: Do not get too geeky. Intelligent Mon(k)ey Managers do not care. It does not even matter even if the errors are big as long as they can beat the expectation by a penny. Just win today’s game and shoulder the Mon(k)ey managers, and you will be flying like rock stars. Ok, if it helps to satisfy ego, Dud(e) Data Managers are there only to support Mon(k)ey managers.
Statisticians are a dying race. This is a thick skinned Hippo. The Hippos probably do not want to come out of water. They can have their saga just in water, just working only on power data. As a different analogy, it all depends on how statistician caves can transform quickly and defend it, and thrive. I will never touch a graduate program with a ten feet pool, if that does not include at least two heavy computational courses, one of which has to be some thing like SAS/R/Python combination of courses. In some programs these are spread out across many courses. Going forward they should also include BIG data computational course if applied courses want to claim their affinity with opportunities or at least make it available as a track.
The opposite of big data is small data. You can not get sillier terminology than this. How ignorant and narrow understanding of big data, one can get? This means people neither understand why data explodes and what actually matters and what we actually do with data. The opposite of big data is not small data; it is power data. In the end all big data has to come down to power data; that is where decisions are made. Some think that if they know how to visualize data, they can interpret data, which is true, but that is not a substitute condensing big data to power data. The whole idea of Map Reduce is the early glimpses of where collected big data dross has to come down to. See: http://predictive-models.blogspot.com/2012/07/data-scientists-vs-decision-scientists.html
I will keep adding myths in this dragon world.
This is meant to provoke harsh look at some views. Have fun and flame it if your truth burns inside.