The Epidemiology of Innovation (An Introduction to Culture Mapping)

Scan the headlines of business journals and you might believe everyone is deeply invested in a transformative and dynamic long view of the markets they serve. The measure of a great company today is how innovative it is. Over $650 billion is spent on innovation efforts globally. But what do we really mean when we use this term? The efforts do not reflect the outcome. Scratch the surface of research and a dismal picture emerges. A recent McKinsey poll revealed that 94% of managers are dissatisfied with their company’s innovation performance. One of the key factors appears to be a persistent corporate mindset that rewards conclusions over evolving intelligence.

To better match the complex challenges innovation presents, it is critical to consider an approach and methodology more akin to epidemiology a method that accounts for the full picture of how trends in culture are evolving. A process rich with layers. One that recognizes often overlooked triggers to new behaviors. One that is exploratory, yet grounded by real data and structured by cultural parameters. An approach that unites consumer anthropology with data science.

We developed Culture Mapping into a patented semiotic algorithm for analyzing the patterns and migration of culture and trends. The result is a functioning API that collects and structures semiotic data from both open and closed data sources. This semiotic data consists of the signs and symbols put out into the world (online and offline), knowingly or not, by human beings. These signs are both words and images. And they come from diverse sources of both closed and open data. A closed data set could be books, magazines, songs and fanzines. Open data would be social media and other linguistic corpora. They all can be processed for key cultural signifiers using systems of natural language processing informed by analysts with expertise in unique subject matter areas.

It’s a balance from human to machine and back again. Our algorithms, based on our matrix structure, are intended to learn from the data collected, and query back to us. The process is akin to gardening. Cultivating and propagating to understand the relationship between unique strains. The process is repeatable and scalable. Is it an empirical technique? Is it quant? It’s a new gray area. We are quantifying language by plotting cultural language at data points. The task is not categorically sorting, but truly mapping the coordinates of cultural signs and symbols. Once we map and visualize the data, we can step back from it and begin to analyze and consider connections. This mapping keeps the creative process engaged in signals as they emerge and migrate over time. We become invested in a living system that we are designing to. This method affords us a way to shape our empathy to consider a variety of potential scenarios that may arise from the products and services we design. We can confidently consider a segmentation that is grounded in the reality and patterns of culture.

Current methods of data analysis are flawed because they allow bias to enter unexpectedly. It happens because we are too quick to look for finality in the data. When we look at information in piecharts and table graphs these visualizations impose an implied conclusion to the analysis. We think if it looks right, it must be right. It is only human to seek a single clear action from the information. But we need to build confidence in thinking deeper. The data masks the physics that shape these recorded outcomes. Methods such as sentiment analysis oversimplify the results to a binary of positive or negative and is recorded only if the signal becomes loud enough. These loud signals are muddied with cultural noise that is not taken enough into account. If we are not considering the triggers of the response, we are not considering the empirical truth of the data at all. Our methods must help us stay engaged with the data in creative ways. Our methods must help us be brave in our quest for innovation.

When visualizations gloss over important nuances that might lead to critical behavioral shifts, we all lose. It leads to findings that are impossible to integrate into the creative side of the innovation process. It leaves no room for inspiration. It dictates. Potential viability of product is interwoven with potential desirability of product. The only way of seeing that vision is to use a cognitive framework. We need our insight into culture to be fused with our way of imagining and making new things. As we create, we must be able to connect ideas to a living and evolving system that contours future potential.

Visualizing expressive linguistic data offers an inductive process of mapping cultural patterns, migrations, and evolution across genres. This inductive process is what separates semiotic thinking from design thinking, which follows a more deductive approach. Whereas design thinking lands on a new concept, semiotic thinking allows clients to see the cultural system unfold over time. Our empathy drives great ideas, but it cannot live in isolation. Ideas must stay connected to the way culture continuously works through its cognitive frameworks. The reality is that the successful products we design are made in the minds of the culture that consumes them. That consumption constantly adapts and integrates what is made, the way a human being sees as appropriate. That is the necessary living state of innovation, and our technology has to synchronize with that reality. Our ability to pattern the linguistic parts of the whole will better assure that products and ideas have long-term health in the culture they seek to impact. (full text of the paper is available at