This morning, I set out to write an insightful and well-researched article on AI content personalisation and its potential applications in B2B marketing. Like most writers I know, I make small efforts to romanticise the task at hand. “Bill Shakespeare may have inked his quill to the musical ministrations of a six-stringed lute,” I thought, connecting my earphones and opening Spotify, “but I have this personalised playlist called… ’wistful bubblegrunge early morning.’”
That is a real playlist title.
In September 2023, Spotify launched Daylist — a hyper-personalised playlist curated on a musical theme based on past listening behaviours, down to the weekday and the time of day. On Wednesday afternoon, my Daylist was called “glow up vocalist wednesday afternoon”. Based on my previous listening history, I learned, I tend to listen to Western and global pop music during Wednesday afternoons.
I am not the only netizen to be enamoured with Spotify’s unhinged Daylist titles. A quick trawl of Reddit unveiled increasingly hilarious hyper-specific musical moods, including:
- feel good coastal grandma sunday night
- delulu hopeless romantic tuesday evening
- paranormal dark cabaret afternoon
- nervous ocean thursday morning
Frankly, the copy reads like the half-baked Instagram poetry produced by a self-declared author whose trust fund is better established than their prefrontal cortex. The earnest efforts of Spotify’s new AI-driven personalisation feature to categorise millions of hours of listening histories into playlists studded with hip internet lingo (it’s okay if you had to Google “coastal grandma” just now) have been the subject of much hilarity.
Jokes aside, the efforts I took this morning to prepare for the production of this article became an ironic exercise in AI content personalisation. It’s impossible to escape — and I could just be a delulu romantic, but I don’t think that’s such a bad thing for B2B marketing.
What is AI content personalisation?
AI-driven content personalisation is the process of tailoring content and messages to individual users using AI algorithms based on factors such as their browsing and purchase histories, online behaviours, location, and other demographics. AI is being used to extraordinary effects to make personalised content recommendations on media sites like Spotify and Netflix, but its uses in digital marketing are where we see the true conversion power of AI-driven content personalisation.
Industry monoliths like Google use AI to personalise search results and advertisements based on browsing and purchasing behaviour. Amazon uses it to generate product suggestions, craft hyper-personalised email campaigns, and suggest website content likely to be relevant to the individual.
From a conversion perspective, it’s easy to see why this degree of personalisation in the purchasing journey is so effective. Although concerns about data privacy prevail, most internet users are flattered by the attention. That being said, it’s thanks to careful monitoring and readjustment that players like Google and Amazon know you like an attentive suitor rather than an obsessive stalker.
This means that audience segmentation just got a lot more detailed
By pulling an enormous wealth of hyper-specific data at speed, marketers can leverage AI-driven audience segmentation to paint a very precise picture of how individuals slot into different categories. This kind of analysis goes far deeper than traditional audience segmentation, which uses relatively simple demographic data to create well-informed assumptions about a target audience.
Unlike traditional segmentation, which is static, AI-powered audience segmentation with machine learning is dynamic and reflects real-time changes in buyer behaviour. Considering that people, groups, and markets are in constant flux, AI segmentation may provide a much more realistic view — even when it’s gazing into a crystal ball.
By upgrading audience segmentation with the predictive power of AI, marketers can begin to predict customer behaviours as nuanced and varied as churn likelihood or potential lifetime value. These predictions help craft more targeted campaigns and more efficient resource allocation.
The limits of AI content personalisation
The benefits of AI content personalisation for marketing seem endless — increased conversion rates, more efficient campaigns with higher ROI, better customer engagement, and improved customer retention are just some of the benefits to be attributed to the power of personalised marketing. However, there are limits to hyper-personalisation, especially as regulations about the ethical use of personal data are still developing.
Concerns about data privacy
Earlier this year, the highly-anticipated European AI Act was passed, marking the first comprehensive attempt to regulate AI globally. Among other things, the Act outlines provisions to ensure the quality and safety of training data used to develop AI systems, as well as requirements to protect individuals’ privacy and personal data. Furthermore, developers and providers of AI systems are required to uphold transparency and accountability by providing clear and understandable information about how their systems work.
The Act signifies growing consumer awareness and concern about the ethical use of their personal data, which could have significant implications for the use of AI in personalised marketing efforts. Spotify Wrapped may have made citizen surveillance sexy again, but all individuals should have the right to understand how their data is being collected, stored, and used (at the very least).
Hyper-personalisation is high-maintenance
Not only does successful hyper-personalisation require advanced tools and tech that typically require Google-sized budgets, but it also requires constant monitoring to remain effective. Marketing teams need to maintain constant vigilance towards changes in customer behaviour, market trends, and the overall effectiveness of personalised campaigns.
Creating a content echo chamber
Personalised content recommendations can’t really go awry in entertainment media, unless you suffer from acute FOMO and can’t understand why you’re not cool enough to have Baby Reindeer advertised to you on Netflix like it did to all of your friends. However, there are scenarios in which the hyper-personalisation and curation of content can cause legitimate issues.
The now-defunct app Artifact, founded by Instagram co-founders Kevin Systrom and Mike Krieger, was a free news aggregator that recommended individual articles based on your engagement with your personalised newsfeed. Colloquially coined “the TikTok of text”, the model essentially turned the news cycle into a For You Page.
That’s neat, but one of the dangers of personalising newsfeeds to that degree is creating an ideological echo chamber in which an individual only ever hears the news they want to hear. Not only does this create a hypothetical reality in which people are missing out on news that could be critically important to them but, on a more philosophical level, it could create a future in which news readers remain unchallenged by a different point of view. Finally, bombarding users with an endless stream of hyper-personalised content could also just turn them off altogether. Variety, after all, is the spice of life.
The last word on hyper-personalisation
We live in an age where most internet and app users are the recipients of marketing campaigns designed for an audience of one. Food delivery apps like Uber Eats use data from past orders to push meal recommendations and special offers around the time you usually order lunch; Netflix assesses your viewing habits to advertise shows it thinks you’d like in the topmost banner of your login screen; Spotify tracks your listening preferences and patterns to curate whimsically-named playlists for every time and day of your week.
Apps recommend when and what we eat, what music we listen to, what news we see first, and what to watch when we’re curled up on the couch. AI-driven personalisation also recommends what project planning tool we should purchase for our small business, which website design agency we should use next, and even what brand of bulk coffee pods to order for the office.
In many ways, personalisation is a win for consumers and marketers alike. Buyers get to the bottom of what they want faster, and we curate content that speaks to a more clearly defined target audience, and to a more satisfying effect. We’re watching and waiting to see how legislature guides the ethical use of personal data in the future but, for now, we’re sat back, relaxed, listening to some wistful bubblegrunge early morning.
