The amount of digital trace we leave behind us every day is no doubt already large, growing and increasingly unstructured. In other words, it is big, as Big Data term proponents like to say. At the same time humans have a need to create information, knowledge, understanding and wisdom out of any data (Ackoff 1989, Rowley 2007, Bellinger et al 2015), big or small. Luckily, technology which enables us to do it at scale is there. So things seem to be all right and a lot of value seems to be generated on a daily basis. But is it really so? In search for fresh answers I attended the Aalto Digital breakfast on Big Data in the Industry where data experts from Yle, Smartly, M-Brain, Supercell and Tieto presented their uses of Big Data to create value for their customers.
It is everywhere, and the companies benefit from it
Based on Rossi et al (2015) big data has found its applications in nearly every field of business (see Figure 1) through digitalization of services. An example is gaming industry which remains the frontrunner with using the data to figure out who, when, where and how is playing their games. Another example are media and retail industry which are finding out which content the customers are likely to like and purchase next based on their earlier preferences. Further example is market and media intelligence which uses massive datasets to identify trends and assist and simplify decision making within companies.
It therefore appears that both businesses and end-user customers benefit and realize value from Big Data, big time. Business users mostly benefit by being able to more precisely target their offers to customers that are most likely to (dis)engage, or validate whether the existing offer has sufficient engagement. End customers eventually benefit from the offer that might best fit their own aspirations and needs. Most use cases are, however, still largely business driven. It is still not that often that end customers fully consciously give up their digital-trace in order to be offered tailored data products and services. The general feeling is still that of companies being in the driver position and end customers merely following them. This does not look like co-creating value with the customer but still very much trading goods for money. Humans’ behavior is considered to be mere sequence of numbers, and a lot of them. The general focus is on quantity of data since it gives sufficient statistical significance. The assumption is thereby that humans are statistically predictable creatures, which will behave the same in the future as they have behaved in the past. But is that really true? It is almost commonly accepted that most of the human decisions are based on emotions at least as much as on facts and numbers.
But how to really bring it back to the customer?
While being powerful, Big Data analytics technologies still seem to mostly generate information and at best knowledge, at least based on definitions of Ackof (1989). They seldom stretch to answer the question of “why (does the customer behave like she behaves)“, based on pure numerical data. The data which, when brought together, could eventually answer these questions is locked into corporate silos. As suggested by Vakkuri (2015) efforts similar to data.gov and Helsinki Region Infoshare could be extended nationwide to bridge this gap. Even if available, in order to answer the “why” question a lot depends on the domain knowledge and intuition of a data scientist. As summarized by Valtonen et al (2015), the huge amounts of data can be analyzed automatically to generate information and knowledge which gets outdated fast, but it is still human touch that is needed to make sense of it and turn it into longer lasting wisdom.
Some of the key skills to reach to level of information and knowledge mentioned by Rossi et al (2015) are statistics, scripting, software development, parallel computation platforms, presentation skills and last, but not least, domain knowledge. While these may be sufficient to communicate with the customer indirectly, i.e. through data, one has to remember that the gained insights are thereby bound to be “thin”. In order to collect “thick” data and get to the level of wisdom one requires ethnographic research methods as well (Madsbjerg & Rasmussen, M.B. 2014). This still seems to remain out of the big data scientist toolkit, without obvious reason.
The customers nowadays offer their digital existence to businesses, and pretty much for free. But is their story understood by the businesses? Are customers getting in return what they really value? We might be just a conversation away from finding out.
Rossi A., Ojala M., Kärkäs P., Valtonen K., Vakkuri M. Digi Breakfast on Big Data in the Industry, http://digi.aalto.fi/en/aalto_digi_strenghts/data_science/, Accessed on 14.Dec.2015
Ackoff, R. L. 1989. “From Data to Wisdom”, Journal of Applies Systems Analysis, Volume 16, 1989 p 3-9.
Rowley J. 2007. The wisdom hierarchy: representations of the DIKW hierarchy, Journal of Information Science, 2007, 33(2), p 163-180
Bellinger G, Castro D., Mills A. 2015. Data, Information, Knowledge and Wisdom, http://www.systems-thinking.org/dikw/dikw.htm, Accessed on 14. Dec. 2015
Madsbjerg, C., Rasmussen, M.B. 2014, The Power Of “Thick” Data, http://www.redassociates.com/press-1/2015/8/18/wall-street-journal-the-power-of-thick-data, Accessed on 11.Dec.2015