Is there any such thing as a truly universal IT product, service or data stream? Or is cultural baggage inherent to anything done on a global scale?
I’ve made a career out of curating “messaging” gaffes around the world and teaching companies how to avoid them. So how could I not be interested in the cross-cultural implications of one of the most-hyped IT trends in history, the Internet of Things (IoT)?
Considering the technical expertise of the audience reading this blog, I need not go into the weeds about the principles of IoT. Let’s just say the basic idea is sensors sending data to other sensors, humans or a central repository, in order to gain insight from the aggregated or “a la carte” feed data.
While the sensor technology may be universal, the “things” that create data can have considerable cultural implications, and this is especially noticeable if data from around the world is aggregated at global headquarters.
Over the past 20 years, I’ve written about how companies need to develop international communications strategies that have a “global chassis and a local body.” In essence there needs to be a core corporate brand message containing universals, with sub-messages that pertain specifically to a target culture and language. One only need to look at the McDonald’s menus from around the world to see this delicate balance between the global chassis and local body of “things.”
This challenge is by no means new. Enterprises have had to normalize data and localize user interfaces well before there was an Internet. The most rudimentary example is normalizing the variation in decimal point or comma in international currencies. Another has been accommodating and integrating double-byte character sets from Asian, Arabic, Hebrew and Cyrillic text feeds.
But what is different in 2016 is the age-old 4 V’s (volume, variety, velocity and veracity) of massive amounts of data coming from such a wide range of sensors. How does IT develop the skills to derive insight from country data that defies normalization but is critical in the overall global analytics strategy?
Here are four examples of “local body” peculiarities that may increase the challenge of incorporating meaningful data into the central repository:
You can’t always get what you want!
There is tremendous variation in how countries allow or limit IoT data collection and transmission. As you might imagine, Europe is a hornet’s nest of EU regulation, complicated further by local country rules. In this environment, it can be difficult to find a highest common denominator for certain data feeds derived from the individual citizen/consumer, especially given the velocity of changes in privacy codes at the central and local levels.
Machine-to-Machine (M2M) can only go so far without humans
While the quality and reliance of M2M technology has made tremendous strides, history tells us that human-to-machine intervention is needed to understand important cross-cultural data nuances. For example, machine translation of languages has made tremendous strides. But a simple Google search on “machine translation blunders” gives enough examples of errors a human would have caught to give the M2M evangelist heartburn.
Gestures are becoming the latest “Thing”
Just when we think double-byte text character normalization is enough, now we find there is an emergence of international gestures for IoT input. Any undergraduate student taking a cross-cultural communications can tell you how gestures add an exponential challenge to international interactions. What we think of as a ubiquitous and innocuous “OK” symbol here in the U.S. can be translated as being called a “zero” in others.
Emotion and sentiment matter
Finally, emotion has become a key predictive analytics indice. Not unlike language translation, machine-to-machine algorithms have become better at determining emotion and sentiment in various IoT feeds. But even the cross-cultural subtleties in language make analyzing emotion and sentiment coming from sensors and social media listening posts exponentially challenging.
Collecting and analyzing data from IoT sources around the world has tremendous potential. But normalizing understanding, gestures and emotion across cultures requires a cultural anthropology dimension to staffing resources that few IT shops are prepared for.
Has your organization run into a cross-cultural challenge for data, and if so how have you handled it ?