AI and, more specifically, machine learning, have created new ways to personalise dynamic content specific to an audience. This confers advantages to many companies, including those that sell directly to the public. To do requires segmenting the audience based on their demographics, psychographics, behaviours, and goals. AI technologies can assist with this.
However, the possibilities offered by AI are not always realised since retailers and brands risk falling into a trap of over-personalising their offerings. This is according to Michelle Bacharach, CEO & co-founder FindMine.
Bacharach’s firm provides AI-powered dynamic content creation for brands including Anine Bing, Land’s End, Lululemon, Rent the Runway, and Rogers Media. This gives Bacharach a unique perspective.
In addressing the problem further, Bacharach states: “Retail by its nature is aspirational – consumers shop for a better sense of style, fashion, beauty etc. Personalization is great at giving the customer more of who they are, but brands can risk losing their editorial vision by over-personalizing.”
Taking the personalised approach further, Bacharach thinks: “If everything looks like “me dot com” then every brand is simply competing on price. In order to maintain the brand identity and position they’ve cultivated in the minds of consumers consciousness, staying true to the brand’s editorial vision is an important differentiation and closes the gap between consumers and the aspirational vision of themselves.”
Where companies often make mistakes is by not seeing issues through to their required point of completion. Here Bacharach says: “Personalization shouldn’t be the end all be all, but rather an opportunity to learn something about your customer.”
Illustrating this with an example, Bacharach says: “A new mom trying to get back in shape is a different shopper than a marathon runner, even though demographically they might be the same. We train our AI to split the “running” campaign into all the different flavours – marathon runner, your first post-partum run, etc.”
This means: “They each contain the unique expertise of the brand but didn’t have to be manually set up, so the brand can create every flavour of the running campaign, rather than just one generic campaign.”
Consequently, the brand can “see who’s clicking on what and store that information to their CDP so they know that a certain person should be in the “new mom” segment and tailor a more custom experience to that shopper in the future. In this way, personalization and aspiration combine to get a better view of the consumer, while helping her become a better version of herself at the same time.”
Therefore, in terms of lessons to be learned Bacharach notes the dangers of over-personalisation and how retailers and brands can best maintain their identity while still tailoring custom consumer experiences.
Linking this back to her own company’s approach, Bacharach indicates that FindMine uses machine learning to develop custom digital marketing assets. According to Bacharach this eliminates the content bottleneck and unlocks “the full promise of personalization while protecting and enhancing a brand’s POV.”
