What is Thick Data?

We live in a time in which Big Data seems to be the ultimate solution for data collection. The possibility of obtaining massive and segmented information through the Internet, the fact of being permanently geo-localized through our smartphones or the huge amount of metadata we produce in our virtual chores (getting to popularize the saying that “Google knows you Better than yourself “) have been a real revolution in every conceivable sector. Organizations and companies desperately seek to find Big Data professional profiles to incorporate into their companies, being one of the most promising job of the future.

This situation has led to a certain devaluation of the qualitative data, which is seen as subjective, insignificant, “small-scale” and therefore unreliable. What can anthropology, the discipline that studies the particular, do to confront the increasing popularization of the “Big”, the great tsunami of quantitative data generated by algorithms?

The truth is that there is much we can do. And it is necessary to name it and spread it. We present you the Thick Data.

Thick Data. The importance of context and emotion

The term Thick Data has been popularized by the anthropologist Tricia Wang and it refers to “dense data”, a more than evident nod to Clifford Geertz´s “dense description”. The Thick Data differs from Big Data by its qualitative approach, obtaining ethnographic data that allows to reveal contexts and emotions of the studied subjects. While Big Data requires an algorithmic process usually carried out by statesmen and mathematicians, Thick Data is the ground of anthropologists, sociologists, and social scientists.

The process of data collection carried out by Big Data requires a standardization and a grouping that strips the results obtained in context. The immense size of the samples makes it impossible to focus on particular stories, which are full of insights and emotions that are fundamental to understand the relationship between the consumer and the product:

When organizations want to build stronger ties with stakeholders, they need stories. Stories contain emotions, something that no scrubbed and normalized dataset can ever deliver. Numbers alone do not respond to the emotions of everyday life: trust, vulnerability, fear, greed, lust, security, love, and intimacy. It’s hard to algorithmically represent the strength of an individual’s service/product affiliation and how the meaning of the affiliation changes over time.

By focusing on a smaller sample, Thick Data is able to reveal the social and emotional context of users, key information to determine the success or failure of a strategy, product or process. The data obtained by Thick Data are based on human depth and contextual particularity, and the collection process can produce novel ideas not yet addressed. Undoubtedly, the quintessential technique of Thick Data is ethnography.

Thick Data is not shown as antagonistic to Big Data, but as complementary. It is not the classical and Manichean opposition between quantitative and qualitative, but rather to understand that both techniques are equally valuable to obtain a general view of the situation that we are studying. While the Thick Data loses scale, the Big Data loses resolution. While Big Data isolates variables to identify patterns, Thick Data assumes human complexity. Two sides of the same coin.

Nokia y Lego: Thick Data cases

Tricia Wang played a part of her career with the Nokia phone company. In 2009, Nokia was still a market leader in “emerging” countries such as China. The company had invested heavily in Big Data, and had determined that its business model would be to produce smartphones for elite users. However, throughout her extensive field work in China, Tricia Wang had realized that low-income users would be willing to pay for more expensive smartphones. She thought Nokia should not focus its product on the select minority, but should shift its strategy towards more massive audience. And so she told the headquarters of the company.

However, Nokia argued that her 100-person sample was “weak” compared to the immense amount of data it had obtained through Big Data, and that none of its quantitative indicators supported Wang’s theory. Something that was quite logical taking into account that the Big Data, with all its virtues, is incapable of understanding the cultural context and the human emotions.

Nokia’s product development model was a fiasco and the company was absorbed by Microsoft in 2013. It currently has only 3% market share, and its fall in China precipitated its worldwide decline. According to Tricia Wang, one of the fundamental reasons for this cataclysm was the blind faith the company put into numbers and quantitative analysis, omitting more complex dense data such as those provided by the ethnographer. The issue is, in her own words, that “what is measurable isn’t the same as what is valuable“(Tricia Wang).

In this video you can see Tricia Wang explaining this case in more detail. You can also consult her article “Why Big Data needs Thick Data?”.

Thick Data: future for anthropology

There are numerous examples of applicability of Thick Data and we fervently encourage you to continue to snoop around the concept. Anthropologists, sociologists and social scientists of tomorrow must learn to defend what they know how to do and the added value that their studies can bring to companies and organizations. Ethnographic techniques are most suitable for Thick Data, as they delve into a myriad of contexts, meanings and emotions that other methodologies, perhaps more settled, difficult to perceive. Thick Data is a cover letter, a concept that we should commit to and popularize among all.

As always, there are many questions that open up in the applicability of Thick Data. How to show results that are not quantifiable? How to explain to the agencies that the use of Big Data, for many opportunities offered, is insufficient and even ineffective to address certain studies? And besides all this, we should not omit the “moral question”: Is it legitimate that anthropology serves the interests of companies, or should we remain in the comfort of university cloisters?

These questions, and many others, must be approached from the collective work carried out by the anthropology of tomorrow. An anthropology charged with the future, deprived of the myth of “lack of exits”, proactive and committed to sectors as diverse as technology, organizational culture, design, health, public policies. An anthropology that does not understand Big Data as a threat to its own existence, but is able to position itself as a complementary and necessary element to quantitative studies.

Only in this way will we reach the desired goal of professionalization.