Ideas and Innovations: Unistack, Generative AI
Could profipoly marketing have emerged earlier? I don’t think so. It needs multiple innovations that have been in the works for some time to come together. Profipoly marketing is not just about taking consumer data and running algorithms to personalise recommendations. It is about a unified experience across channels and brand properties. It is about differentiation in terms of ease, exclusivity, and access for the Best customers. It is about frictionless journeys, perfect predictions, and inbox commerce. All of this requires thought and technology breakthroughs – just like the transformers idea and cheap computing enabled the Generative AI revolution. Twelve ideas and innovations provide the foundation for the profipoly revolution.
1. Unistack
Unistack, or a “unified stack,” aims to address the issue of integrating diverse point solutions. Over time, marketers have leaned on various point solutions for handling distinct facets of their marketing efforts, like data gathering, customer segmentation, and campaign orchestration across different channels. However, this approach has resulted in fragmented data and integration hurdles, obstructing marketers from achieving a comprehensive understanding of their customers.
Attempts to remedy these issues have been made through Customer Data Platforms (CDPs) and Application Programming Interfaces (APIs), but the fundamental problem persists. Marketers are still wrestling with disjointed databases and inadequate AI-ML effectiveness because of the siloed data, which hampers their ability to deliver optimum customer experiences. The first-generation martech solutions contributed aggregation and automation, but inadvertently created data silos and failed to provide a complete, seamless solution. Unistack consolidates customer data, engagement, and experience management, and full channel control into one platform. Adopting a Unistack approach allows marketers to gain a holistic, integrated perspective of their customers, thereby improving the effectiveness of their customer relationships and unlocking the potential of seamless omnichannel personalisation.
[Source: Martech 2.0: A New Profits Paradigm for Marketers and Vendors]
2. Large Customer Model
The first wave of Generative AI in marketing is helping drive content creation. Anna Anisin explains the uses:
Brainstorming and Idea Generation: Coming up with fresh ideas is essential to capturing customer attention. Generative AI models, such as ChatGPT, enable brainstorming sessions, offering creative suggestions and alternative perspectives.
Automating Content Creation: Generative AI can empower marketers to automate various aspects of content creation, saving time and resources for faster time-to-market. From generating social media posts and blog articles to crafting email campaigns, AI models can produce draft content that human marketers can refine and personalize.
Enhancing Existing Content: Generative AI can update existing content by providing valuable insights and suggestions for improvement. By analysing data patterns and user feedback, AI models can identify areas where content such as marketing copy, ad creative and customer messaging can be optimized.
Creating Visuals: Generative AI models can generate stunning visuals, including graphics, images, art forms and videos. Marketers can leverage these AI-generated visuals to enhance their storytelling, create eye-catching social media posts and produce visually engaging presentations.
The real opportunity lies beyond. Generative AI, with its ability to simulate and create data-driven patterns, is poised to revolutionise the marketing landscape. In the realm of predicting customer behaviour, this form of AI can sift through vast amounts of historical data to construct highly accurate models of customer preferences, behaviours, and potential future actions. By identifying subtle patterns and correlations that may elude traditional analytics, Generative AI can anticipate the ‘next best actions’ of customers with unparalleled precision. This not only ensures more personalised customer experiences but also enables marketers to be proactive rather than reactive, crafting campaigns and strategies that align seamlessly with evolving customer desires and needs.
In the world of customer engagement, Gen AI can be used to create a Large Customer Model (LCM), analogous to the LLMs we are seeing and using. From a previous essay:
LCMs…operate in a different ‘language’—the language of customer behaviour. They predict the next action in a customer’s journey by analysing extensive datasets of customer interactions and behaviours. The aim is to recognise patterns and sequences leading to specific outcomes. A prime example would be a good recommendation engine, which predicts what product a customer is likely to be interested in next, based on their browsing history and past purchases.
However, LCMs surpass LLMs in their ‘vocabulary’ breadth. They not only consider individual actions, like browsing a product or making a purchase, but also can consider a wide array of external factors. For instance, factors like the time of day, weather conditions, or significant events like holidays and festivals can influence customer behaviours.
What makes LCMs truly powerful is their ability to fuse learnings from vast datasets with real-time behavioural information. They capture the ‘customer language’ at an individual level, learning from each interaction to refine the predictive models. They anticipate behavioural changes, offering marketers the opportunity to adjust strategies proactively.
Think of the LCMs as next-generation Customer Data Platforms (CDPs) where data comes to life with predictions of next best actions.
