AI has generated a lot of excitement in the marketing community. The prospect of working with tools that could automate tedious tasks while simultaneously improving outcomes is certainly an appealing one.
For that reason, the Digital Marketing Institute launched an ‘Artificial Intelligence in Digital Marketing’ course – following “overwhelming demand” from its members. Unsurprisingly, in the institute’s annual members survey, participants ranked AI as the top skill they were hoping to learn in the following year.
While there’s no doubt that AI will prove hugely influential in the near future, there’s one thing that the institute’s announcement raises doubts about. It claimed that the best marketers today are “seizing the opportunity to harness the AI revolution”, and that course participants could learn skills that they could apply “the following day.”
While the latter may be true for those looking to use tools like ChatGPT for personal tasks, to truly harness an AI revolution at a business to influence and manage customer experiences at scale, marketers crucially need something many currently lack: solid data foundations. Simply put, without these, their skills won’t really matter.
Marketing Technology News: MarTech Interview with Susan Ganeshan, Chief Marketing Officer @ Emplifi
Also catch this: Fireside Chat Episode 2 – David Raab
Fixing data foundations to unlock AI
AI can be implemented in many ways, but one fact remains true: it is most powerful when supported by a comprehensive foundation of data.
For example, a customer marketing team that wishes to leverage machine learning to automate insights for their campaigns needs it. Large Language Models (LLMs) that automate and personalise content creation need it as well. But if the team’s customer data is flawed or incomplete, flawed or incomplete outcomes will follow.
One way to understand this is through the potential use of AI to personalise customer experiences at scale. Personalised content and recommendations can increase customers’ basket sizes by over 20%. This is something that customer marketing teams have been attempting to achieve for years. Yet, the labour-intensive processes involved in content creation and targeting have made it near impossible to deliver. ‘Unsophisticated’ was how half of all customer marketing professionals referred to their level of personalisation in our recent research.
With AI, however, this marketing utopia suddenly becomes possible (when implemented correctly.) For instance, AI can aggregate and analyse vast amounts of customer data from different sources, which can then be integrated with an LLM application to output highly personalised content. Instead of viewing generic marketing content, customers receive content that seems as if it were designed just for them. This is because it has been built on a set of real-time data about their cross-channel behaviour.
AI used in this manner has the potential to revolutionise marketing. However, it’s important to remember that AI models are only as effective as the quality of the data used to train them, particularly when tailored to specific customer profiles. Unfortunately, very few customer marketing teams currently have the necessary data structures.
Only 21% of participants in Plinc’s research “strongly agreed” that they could easily access and analyse customer data when required. When we asked how often their core data foundation is updated, almost two in every three B2C companies reported once per week or even less. These are not the right conditions to “seize” the AI revolution.
Unifying data to enable personalisation at scale
Without the right data foundations in place from the start, customer marketing teams will never be able to exploit the full potential of AI. To build them, they need to focus on improving access to their customer data with the right platform. This should combine all their customer data, from all sources, into a ‘Single Customer View’ (SCV).
The truth is that many marketing teams lack a robust SCV. This prevents them from creating personalised campaigns because they’re unable to access profiles that contain all the information they have about individual customers. Indeed, when the time comes for marketing teams to implement AI, SCVs will be crucial in allowing software to rapidly gather the relevant information it needs, and then develop it into personalised content. They will become the basis for the marketing model of the future.
With solid data foundations, marketing teams will also facilitate the delivery of more useful, real-time data for the rest of their businesses. As business leaders seek to expand the remits of their marketing teams, this will help to define what they become in the future: sources of business intelligence that understand their customers better than anyone else. It will also help marketing leaders prove their worth. Training and support for marketing executives will be crucial in ensuring its success.
Given the time that it will take for many marketing teams to make these changes, we might not see AI truly take off in the short term. However, I’m optimistic that once marketing teams realise the need to fix their unstable data foundations – and actually fix them – we will see the marketers fulfil the tech’s full potential, achieve personalisation at scale, and redefine how marketing teams operate.
Marketing Technology News: SGE Optimization: An Opportunity to Diversify Your Local Search Strategy
