Answering 4 questions marketers have about personalization and generative AI

“The dream for marketing for many years has been mass personalization.” That’s what Drew Neisser, founder of CMO Huddles, said at the BrXnd Conference in May.

The explosion of generative AI has moved the “dream” of one-to-one messaging personalized for each consumer closer to reality—but concerns remain. Here are four questions surrounding personalization right now.

What does mass personalization mean?

• Personalization in marketing can refer to a lot of touchpoints, from recommendations to location-based targeting to creating user-specific products.
• Personalization also includes how a marketing message is delivered. For instance, brands can contact consumers via SMS, email, or app notifications depending on what they’re most likely to respond to.
• Retailers can use past purchase data to advertise the correct products to the right individuals.
• In a future state, marketers might be able to automatically create specific copy and images at scale for individual consumers using AI.

How do consumers view personalization right now?

There’s a disconnect in how B2C businesses and consumers think brands are doing with personalization. While 46% of B2C business leaders worldwide felt like brands were doing an excellent job at providing personalization as of 2022, just 15% of consumers agreed, according to Twilio data from March.

How does AI play into personalization?

• Generative AI streamlines personalization by automating tasks like copywriting and image creation.
• AI is also vital for combing through large amounts of data and feedback in order to provide insights necessary for personalization. In fact, 45% of US customer experience (CX) professionals said AI will impact CX through its advanced insights and analysis across data sources, according to June data from SurveyMonkey.

What concerns should marketers have about personalization?

• Personalization can be creepy. Hypertargeted ads can raise consumer privacy concerns. That said, more than 75% of internet users worldwide said they were willing to give away their email address, brand interest, and name to receive personalized interactions, according to Airship and Sapio Research.
• Personalized ads make omnichannel attribution difficult, since ad copy and even how ads are delivered (e.g., SMS versus email) may differ for each consumer.
• Humans can’t quality check AI-personalized outputs at scale. AI can help brands hyperpersonalize ads, creating a whole new level of scale. But that level of scale also applies to quality assurance, and without humans involved, the need to check images and copy for brand safety concerns is even greater.

This was originally featured in the eMarketer Daily newsletter. For more marketing insights, statistics, and trends, subscribe here.