There is a popular marketing quote — “half the money I spend on advertising is wasted; the trouble is I don’t know which half.” That quote, which is more than 100 years old, has certainly aged well and the sentiment stills resonates with many even today. But it’s also true that marketers are among the most sophisticated users of data and analytics. In fact, marketing’s use of data and number-crunching goes back several decades. Marketers may not have branded it AI but bread-and-butter marketing use cases such as customer segmentation, buyer propensity and lead scoring models, and customer churn predictions have made heavy use of machine learning techniques. Marketers have traditionally been the biggest customers for software such as SAS or IBM SPSS and we just thought of them as statistical software packages.
If data-driven marketing has been in vogue for years, then what is different for marketers with the advent of the deep learning approach to AI? With deep learning being added to the marketing toolbox, the impact is at least two-fold. Firstly, the accuracy of existing marketing use cases is enhanced. Secondly, deep learning enables new uses that were not possible earlier.
Marketers have to go where customers are and consumers increasingly embrace digital avenues for content, commerce, and communication. As such, the impact of AI is being felt in the digital world and marketers are increasingly reliant on AI-driven technology for their digital (and omni-channel) strategies. But the world of Marketing Technology (or as it is popularly known as, MarTech) is quite complex and overwhelming. There are more than 8,000 MarTech software tools to manage the different stages of the customer life cycle such as consumer awareness, consideration, purchase, customer support, and loyalty. Most of these tools embed some kind of AI functionality. How do you make sense of all this complexity and better focus your efforts to leverage AI as a marketer?
One of the first marketing concepts you learn in B-school is the 4Ps or the marketing mix elements of Product, Price, Promotion and Place. This framework, which is 60 years old, has several variations to serve the needs of different marketing situations, but the classic 4Ps provide us a good frame to examine the AI use cases for marketing.
Product
Products are increasingly AI-inside. For example, fully autonomous, self-driving passenger cars are still far away from hitting the roads, but there are specialised self-driving vehicles — in factories, warehouses, mines, and fields. In passenger cars and commercial vehicles, there are driver-assist features enabled by AI. In financial services, lending and underwriting algorithms are being driven by deep learning. In online learning, we use AI for personalised experiences, exam proctoring, grading exams and more. Media outlets can use AI tools to automatically generate certain categories of news updates — like financial reporting, weather updates, and sports results. Smart speakers and wearables are possible because of AI. In these examples, you will find that AI functionality is at the core of a product or a service, or is enabling a better user experience.
Price
Pricing is a key lever of profitability. Real-time and dynamic pricing methods can help maximise revenues based on changing conditions, and AI can enable such approaches. You already see dynamic pricing use cases in e-commerce, retail, and hospitality industries where price is set based on fluctuations in demand and supply. AI also helps understand what product/features customers value most and help refine your pricing levels.
Promotion
AI can help with granular audience segmentation, automated content generation, better targeting of ads, personalised discounts and offers, customised marketing campaigns. By helping mine and analyse unstructured content, AI helps you surface new insights about your customers. These can be fed to marketing automation tools to achieve personalisation at scale. Not that taken to the extreme, such personalisation can feel creepy to consumers.
Place
AI can influence customer discovery of products and services across both digital and physical channels and plays an important role in shaping customer journeys and interactions. For example:
- AI-driven recommendations are at the heart of online commerce.
- Online catalogues are built in real-time.
- Personalised ads are programmatically displayed on apps and sites.
Key takeaways for marketers
AI is pervasive across the entire marketing mix, but marketers are primarily consumers of AI technology, not its creators. You don’t have to code but understand the workings of the AI tools and their capabilities. Your aim should be to become a super user of AI.
To do that you need to do the following:
- Gain digital and data fluency.
- Understand the MarTech landscape.
- Become proficient in tools, not just marketing frameworks.
- Be respectful of consumer privacy concerns.
While the possibilities are potentially limitless, we should distinguish between tactics and strategy. AI gives marketers turbo-charged tactical tools. But it is marketers who have to think of the big picture, use AI and data to better understand customers, and add value in a mutually beneficial way.
Finally, some food for thought: An early executive at Facebook lamented: “The best minds of my generation are thinking about making people click ads.” AI only recognises patterns. It neither understands people nor their motivations. Perhaps, there is a missing "P" then. We should add a fifth P of marketing — Purpose.