Team Chemistry - Sports and Analytics

Late March is as good time to speak about teamwork. For basketball teams, the right group chemistry is essential, and this on display in the winners of the NCAA tournaments. You can also see how fragile team chemistry can be, in the shift that results when an NBA team adds a "superstar" yet changes their whole approach to playing. The coach has to decide how to best leverage the new individual talents while still maintaining a winning framework.

If you are growing or leading a sales or marketing analytics group, you have choices to make:

- the mix of hands on analysts versus consultants

- whether to staff in house or outsource, or even offshore

- having people steeped in your industry as well as other categories, for a fresh perspective


Then of course, the coaching and training are critical. Analytics is rooted in mathematics and algorithmic techniques which have been around for decades. Yet there are innovative digital, multichannel methods. Thus a leader must find the right mix of setting standards, yet delegating and allowing the group to develop some new innovative thinking.

Methodically removing balls from the air

Ever feel like you are working on so many responsibilities and immediate deliverables at once? It can be just like you are simultaneously juggling many balls in the air. This can happen in the "functional service areas" like analytics.

The key is to focus on getting items completed, one by one. Not on getting overwhelmed with the total set. Each completed and transitioned result will provide satisfaction for oneself and one's "clients."

But how to choose which one to complete first? You need some kind of "greedy algorithm:"
- first in, first out?
- nearest to the deadlines?
- easiest to get over with?
- task with the most pressure and stress attached?

The "greedy heuristic" you choose for getting the first one off your plate may be up to you, or handed to you.

As you plow through your list of tasks, some might be thorny, and need a new perspective, even a chance to put aside and "sleep on it" before solving.

I remember fondly my days on the high school mathematics team when there was a book called: How to Solve it, by George Polya It gave great problem solving techniques for seeing mathematical puzzles and theorems from a new perspective.
Examples of these techniques are:

* break into sub-problems, relate these to each other, and solve the core sub-problem.
* draw diagrams to represent the core issue
* apply a rotation, or a transformation, so the relevant dimension is more obvious
* find an analogous problem and solve that.

For the technically and mathematically inclined, this is a paperback investment well worth it.

Perspective in measurement and in life

In measuring time series data, one realizes that significant outliers or spikes will take on significance only over a certain duration and a certain time scale of granularity. A daily spike in website visits may seem signficant within last month's daily visitor plot. However, it may seem less so with a weekly plot. Or it may lose significance further compared to other larger spikes on a duration plot of an entire year.

This spring weekend has brought a new meaning to judging the relative importance of events, and in juding magnitudes of scale of importance. This humble overworked professional has tried to keep up with assignments over the weekend. He's also watched his Amazon rankings of his book on a daily basis, and wished they would nudge higher. These issues pale in comparison to the pleasure of riding bikes and playing baseball with kids as spring draws near. On the negative side, both hassles and joys are tempered by stuggles friends have with flooding nearby.

Furthermore, all of these are tiny in perspective when considering the natural disasters in Japan that happened within a day but whose duration of impact will be measured in years and tens of thousands of lives.

Nerds on TV -- the right way -- IBM Watson and Jeopardy

As a computer scientist, trained in artificial intelligence myself, I am delighted to see the development of the IBM Watson parallel computer competing on evening TV Jeopardy! episodes. See this Google search for the plethora of results on the Jeopardy shows.


Heck, as a parent, I love it that my kids are running to me with how awesome the smart computer is doing on their favorite quiz show!

See this great PBS News Hour video to find out more about Watson, Jeopardy,and the artificial intelligence background.

When are computers really "intelligent?" People have been pondering this for at least 100 years. Alan Turing developed a "Turing test" of the computer responding to a person's questions and having the answers be undecipherable from that of another human being. Then in the 1980's came the chess playing machines. First with algorithms, then with brute force statistics, and finally Deep Blue beating grand master chess champions by the turn of the century.

But now having smart computers compete on glitzy evening game shows and educating the world about parallel intelligence of computers? Love it.


See a great write-up by James Taylor on the Smart Data Collective community as to what the IBM Watson episodes mean to Decision Management and Analytics. In essence, computers can now make decisions on the fly leveraging large databases, and continually get smarter in our results.
This is after all what people do in decision making.

Think about that the next time you are creating your marketing and sales analytics systems. Rapid feedback loops, shorter decision cycles,and continual optimization.

The Trend of "Bad," Empowered Patients

While on a brief vacation at my Poconos hotel, I leafed through the free USA today that came by my door, and noticed an interesting headline article about an increasing trend of "Bad" or "Empowered" patients. What was described were guidelines in which patients, empowered by information available on the internet, and increasingly responsible for paying health costs, have been questioning physician decision making regarding test and treatment decisions.

Cited often was a new book by CNN's Elizabeth Cohen on being an empowered patient. Find out for yourself who the best physicians are, whether the empensive tests indicated are necessary, whether the expensive medications are required.

Concurrently, another story gaining publicity that features a desperate kidney cancer patient who has benefitted from Internet communities is that of E-patient Dave deBronkart. "E-patient Dave" has become a strong advocate for patients to seek online information for communal support, and to never give up, or accept a grim fate.

This all reinforces the need for healthcare companies to make sure they are online, participating in the discussions that empowered patients are having. It also emphasizies the benefits of digital CRM, to make sure patients are educated about the benefits of treatments you advocate.

Making decisions based on data

When developing a learning plan fo a new RM campaign, one of the most important elements to compelte in your plan are the business decisions to be made should the metrics flow one way or another. There should be an in-market test plan, where the decision making criteria are clear depending on the test results.

Then after the campaign has been running for several months, the difficult decisions have to be made. Shifting media spending, improving your website, changing value propositions or incentives.

Think to yourself: what kind of marketer, or campaign manager are you? Do you like to review all the data, weight the pros and cons, and then choose an alternative yourself? Or do you wish your consultants or agencies to make these choices for you? That can be just fine, as they have experience, and should have the brand's interests in mind.

A third option is to remain indecisive; this is the urge to resist. Metricsare not just for learning. Whatever choice you make, choose the path of optimization.

Also, be sure to record what you have learned: the circumstances and the results. This is what knowledge management is, and where "best practices" and "industry benchmarks" come from, after all: actual recordings of in market events. You may as well make your events part of those benchmarks