The old adage that marketing is half art and half science is about to change. I’m not saying that the art of marketing is going away, in fact as more science is used the need for quality storytelling and creative expression becomes even more important. I’m just reading the tea leaves that the science of marketing is about to tip the balance of this equilibrium in a big way, and that’s largely due to advances in artificial intelligence and machine learning technology.
It’s important to understand that you can’t have artificial intelligence (AI) without machine learning (ML) so I’ll briefly explain each for context. Artificial intelligence, the most known example being IBM’s Watson, is basically the ability of machines to develop cognitive capabilities; to think, infer, and connect the dots like humans. Just like humans interpret multiple data points and evidence and then apply reason to formulate insights and make decisions, artificial intelligence can do the same at mass scale with exponentially larger data sets and variables as long as it can learn fast enough.
This is where machine learning comes in, by rapidly consuming and analyzing enormous quantities of unstructured data to identify patterns, solve problems, make predictions, and subsequently learn from them to make even better predictions, all without explicit programming. For example, machine learning is both the brains and muscle behind natural language processing (think virtual assistants like Amazon Alexa and Apple Siri interpreting spoken requests and answering in kind), speech and image recognition, and predictive analytics, all of which are key functions in the enablement of artificial intelligence.
We have already seen some examples of early adopters testing artificial intelligence in the digital marketplace with mixed results. Sure, there are well publicized stumbles like the corruption of Microsoft’s AI chatbot, Tay, but there are also plenty of successes like The North Face’s ecommerce recommendation engine, and Campbell Soup Company’s Chef Watson cognitive advertising on The Weather Company’s online properties.
Behind the scenes, artificial intelligence is currently being used by marketers to augment various operational functions and tasks. These include optimizing online media buying and programmatic advertising, modeling lookalike audiences for consumer targeting, performing sentiment analysis on social media mentions and customer service audio (via ML powered natural language processing), and automating tag curation for digital asset management.
These examples are just the beginning though, as artificial intelligence is expected to not only help marketers work smarter and faster but more importantly increase consumer reach and engagement by fully automating touchpoints between brands/companies and consumers at massive scale. While these are obvious benefits to the discipline of marketing, new considerations and challenges will surely emerge. For instance, as artificial intelligence becomes the gatekeeper to reach consumers, marketers will need to craft messaging and content that not only satisfies a user’s needs, but also satisfies the machine medium (i.e. website, chatbot, smart speaker, digital assistant, IoT device, whatever comes next, etc.) which will deliver that messaging/content to the user. Some more food for thought…with the recent debunking of purposely disingenuous fake news websites and clickbait articles in social media, will artificial intelligence become so enlightened and efficient at some point that it overzealously and accidentally produces and disseminates fake marketing?
We can deal in future hypotheticals all day long but the bottom line is that as AI becomes more pervasive, professionals will need to be prepared for dynamic changes in customer expectations and how they utilize technology to design and implement customer experiences that satisfy those expectations. Companies that harness the speed and power of artificial intelligence and machine learning will surpass and move farther ahead of their competition. Will you be ready or be left behind?