The industry has adapted the selfie for its own ends, but has perhaps missed some of the deeper desires it reflects.
Asia leads the world in selfie-taking, according to a 2014 study by Time magazine. Four of the top 10 “selfiest” cities are in Asia, a find hardly surprising given that over half of the world’s social-media users reside here. Selfie culture has become so dominant in Asia that two global smartphone giants, China’s Huawei and South Korea’s Samsung, compete with front-facing camera specs and ownership of spin-off category terms like “groufie” and “wefie”, respectively.
In the latest of our semiotic series, Laura Oswald uncovers how we decode the meaning of pain.
By Steve Verba
Big Data is big news. Bigger than anything else in IT, except The Cloud. Bigger than anything else in its impact on Marketing and Market Research. So big that it seems Semiotics would hardly be expected to lay claim to relevance to Big Data immense hype and impact. And yet, on closer examination there are several powerful contributions semiotics can make to bridge the gap between Marketing, Marketing Science, Big Data and Data Science.
Importance of Big Data to Organisations
Tweets about a new product launch, posted blog comments criticizing a brand, news article posts critical of a manufacturer, job board ratings of an employer all constitute new varieties of marketing data, appearing at real time velocities and accruing into huge volumes. On these 3 dimensions alone, marketing and marketing science struggle to keep up with potential impact Big Data can have on Pricing, Promotion, Placement and Product.
With this tsunami of new data, comes a new philosophy from the Big Data community that is at the heart of its perceived benefits: “Big data represents a cultural shift in which more and more decisions are made by algorithms with transparent logic, operating on documented immutable evidence. I think “big” refers more to the pervasive nature of this change than to any particular amount of data.” (Daniel Gillick, Senior Research Scientist, Google)
The biggest appeal of Big Data application in business is that the sheer amount of data provides a sort of intrinsic credibility. Secondly, the appeal also comes from the notion that the tools used to analyze Big Data are positioned as applying pure logic to that data to make precise predictions.
This is both a great Brand Promise and a Unique Selling Proposition for Big Data. But that simplicity sells itself short. It is not so much that current use of Big Data by Data Scientists is not right, so much as it is self-limiting in terms of providing truly useful and breakthrough insights.
However, if we stand back and look at what is happening from a semiotician’s point of view this can be seen more clearly.
What’s the Meaning of Big Data? Ask Semiotics.
Big Data includes a vast array of analytical techniques borrowed mostly from statistics, math and operations research modified to operate on vast data sets. Semiotics, on the other hand, draws our attention to the existence of an underlying code system informing the choices made within the discipline (e.g. what is ignored, devalued or glossed, or what is valorized, optimized or rewarded).
Merging of Big Data with Semiotics can also successfully solve many brand conundrums. In an ad agency competition for a major American ladies undergarment account, all competing agencies already knew a prior campaign had failed and generated negative sentiments.
The winning agency took that data and reverse-engineered the meaning behind the failed campaign using a semiotic framework. They uncovered that the realistic settings and photography did not match the desired whimsical fantasy intent, and instead triggered common uncomfortable dreams of being in public partly clothed.
All agencies had the same social media data – but the interpretation of meaning using a semiotically informed marketing framework is what made a difference in understanding (and won the account).
Big Data is providing marketers with the richest sources of data ever available. The analytical tools today are likewise at their zenith. However, for marketing neither the IT folks working to manage Big Data, nor the Data Science folks seeking to analyze it, can fully leverage their data working only from within the code systems of their own disciplines. Doing so serves to “bake in” limitations of insight and impact relevant to the marketers and business stakeholders.
Unlocking Meaning of Big Data via Semiotics
So how does Semiotics help with these limitations? Simply put, by elevating Meaning back into the pure equations, logic and data troves in play today. We can see this in two dimensions:
Dimension 1: Semiotics of Big Data
We have already illustrated one aspect of how semiotics can help Big Data get it right for marketers – by shining a light on the fact that how you pick which data you analyze and how you pick what tools you use are dependent on implicit code systems you may not even be aware you are using.
A semiotician can therefore shed light on how Big Data itself creates its own meanings and cultural codes as a community. Providing this self-reflection can “open up” Big Data and Data Science to its own blind spots and help forge closer relationships with mainstream marketing practitioners.
Likewise, Big Data needs to reflect on whom it thinks it is analyzing when it deals with marketing data from consumers about brands and products. What is the implicit model of consumers? Are consumers also seen as making consistent “decisions …made by algorithms with transparent logic, operating on documented immutable evidence”? Are Brands obviated because consumers are just exercising micro-economic utility theory decisions based on features?
Marketing has long ago gone past these earlier notions of how consumption takes place. In fact, semiotics has itself helped marketing move past these earlier theories. If we have hope to have real dialog between data scientist and marketers, market scientists and market researchers, we must bridge this gap.
Dimension 2: Semiotics and Big Data
At a more tactical, granular level there are several practical ways semiotics can contribute to the analytics used in Big Data.
Social Analytics – sentiment and text analytics are an obvious place where semiotics can make a greater contribution. After all modern semiotics has considerable roots in linguistics. The use of semiotics square and the concepts of markedness, structural semantics and psycholinguistic engagement measurement are untapped tools that have been applied previously on social analytics data by semioticians to uncover far more meaning and insight than simple broad emotional reaction counts.
Data Visualization – Big Data analytics includes significant emphasis on data visualization. Recent work has treated data visualization and informatics as a form of semiotics engineering to ensure understanding of complex data by treating the code systems of the reader/recipient as key. Semiotics provides the framework on which to judge poor vs excellent data visualizations for a given audience.
Cognitive Computing – one of the most powerful developments in Big Data analytics can be seen in IBM “Watson” which uses hundreds of linked algorithms (neural nets) and an immense knowledge base to process and answer questions like a human. This constitutes what IBM refers to as Cognitive Computing. Watson explicitly exceeds limitations of hard-coded linguistic models using semiotics: “Semiotics allows for representation and synthesis of topological systematic models of different kind, including diagrams. Wherever needed, linguistic models can be converted into semiotic representation with transformational mechanisms using IBM Watson parsing.” Dr.Gary Kuvich – IBM Certified IT Architect
Moving Marketing Forward through Big Data & Semiotics
To consider the larger perspective of semiotics helping to bridge the gap between Big Data and Marketing, it is useful to get a sense of where marketing seems to be going. Paraphrasing Philip Kotler, here we can look at four changes that are going to occur in marketing over the next couple decades: need to co-create products, crowdsourcing ideas, marketing automation based on artificial intelligence rather than done by skilled marketers, and lastly learning how to produce “lovemarks” with customers and employees.
Firstly, Big Data provides the opportunity to co-create products through web interaction – semiotics provides the framework to understand the marketing meaning behind those interactions needed for proper positioning, packaging and branding. Likewise, artificial intelligence systems like Watson can begin to include semiotic frameworks relevant to marketing and marketers to begin to answer questions about campaigns and promotions.
Finally, as we start to look past traditional definitions of brands we see concepts like ‘lovemarks’ that explicitly valorize great stories, dreams, myths and icons. These are the very “stuff” of semiotics. Harvesting these dreams, myths and icons from social media big data and refashioning them to enhance deeper brand loyalty is an ongoing job for the semiotician sitting right between marketers on one side and data scientist on the other.
- Big Data carries its own brand promise and USP, which while appealing, limits its impact.
- Semiotics can provide a very useful perspective on the underlying code systems in the Big Data community. When seen from a semiotic perspective, Big Data can provide more value for marketers and market researchers, but also traditional IT departments.
- Semiotics also contributes to the evolution and use of Big Data analytics tools. Big Data and technology are changing marketing. Semiotics is already part of that path forward.
Steve Verba is a multi-disciplinary consultant with a track record in applied semiotics, technology systems, Big Data and IT. He is based in Ohio a can be reached at firstname.lastname@example.org.