Applying Machine Learning To Marketing
Three years ago, Gartner predicted that by 2017, CMOs would spend more on IT than their counterpart CIOs. Fast-forward to 2015; we are more than halfway there and can already bear witness to the shift in focus to the Chief Marketing Officer as the advocate and purchaser of technology of the future.
The role of the CMO has evolved from a traditional and tactical approach based on simple data capture and inefficient targeted campaigns to performance-led strategies based on rich data insights and measurement of business impact.
However, is there too much technology out there for marketers to handle? And how best should companies balance the adoption of technology with human capital?
Marketing today is very labour-intensive, often requiring marketers to dig through too much data that may not even be giving them the bigger picture they need to make impactful business decisions.
By 2020, the digital universe (source EMC) will grow by a factor of 300, from 130exabytes to 40,000exabytes, or 40tn gigabytes. However, the human brain can only hold the equivalent of 1m gigabytes of memory. With too much data for humans to sift through, machine learning is the act of a machine producing insights without being told what to do.
“[Machine Learning is the] field of study that gives computers the ability to learn without being explicitly programmed.” – Arthur Samuel, 1959
Many people say ‘data is everything’. It’s not; what you learn from the data and what you do with it is what matters most to the CMO.
Balancing what insights we can automate and what additional value human capital can provide is the key to marketing success. Marketing is often described as part art and part science, and in this case data provides the science and marketers provide the art. Machine learning helps marketers solve complex data-rich challenges, that are beyond the capacity and capability of the human brain, and use algorithms to initiate actions based on data. This type of machine learning is often referred to as ‘unsupervised’, as technology finds patterns, builds insights and automatically acts on those insights.
Machine learning has been part of a marketer’s everyday life for decades, without many realising it. I am using automatic spell check as I type this. Modern day examples are found with Google, Apple’s Siri, IBM’s Watson, Facebook recommendations, Quora and (related questions) and any technology that says ‘suggestions’.
In fact, only recently Google CEO Larry Page told shareholders on a webcast how “Google is using machine learning in a growing number of products and services, including automatic translation, voice-based searching, self-driving cars and the Nest connected thermostat”.
In B2B environments machine learning helps CMOs connect with consumers by providing recommendations based on insights about their interests, emotions and interactions with the development of ‘knowledge graph’ technology. Machine learning can help businesses with content marketing. Content performance marketing platform BrightEdge, for example, recently launched a technology that integrates with Adobe Experience manager and leverages machine learning to automate decisions on the content battleground (Adobe is CMO.
