Is Your AI Too Personal?

AI personalization crosses the line when customers can't understand why they're receiving specific treatments. Kathryn Rathje, Partner at McKinsey, explains how marketers often expose too much data instead of focusing on relevance. She discusses the value exchange principle for ethical personalization and why context matters more than data volume. The conversation covers dynamic billboard targeting, spectrum-based personalization approaches, and avoiding the "mad libs of data" trap that makes AI-driven outreach feel invasive rather than helpful.
About the speaker

Kathryn Rathje

McKinsey

- McKinsey

Kathryn Rathje is partner at McKinsey

CMO's Comeback Report

Episode Chapters

  • 00:00: When Personalization Goes Wrong

    The discussion explores how personalization crosses the line when the value exchange becomes unclear and consumers don't understand why they're receiving specific treatments or messages.

  • 00:54: Consumer Assumptions About Data

    Examples of how consumers often assume companies know more about them than they actually do, particularly with location-based advertising and Bluetooth technology.

  • 01:19: The Data Laundry List Problem

    Marketers often feel compelled to showcase all the data they have about customers rather than using it strategically to provide value.

  • 01:59: Context Matters in Personalization

    Effective personalization requires co ecting data usage to the context of the communication and the reason for sending it, avoiding irrelevant data insertion.

  • 02:20: AI Personalization Gone Wild

    Discussion of how AI-powered personalization can produce absurd results when people game the system, leading to completely irrelevant automated outreach.

  • 03:17: Personalization as a Spectrum

    The conversation concludes that personalization should remain a strategic spectrum rather than defaulting to hyper-personalized approaches for every interaction.

Episode Summary

  • Is Your AI Too Personal? Finding the Right Balance in Marketing Personalization

    Introduction

    Kathryn Rathje, Partner at McKinsey & Company, brings deep expertise in data-driven marketing and personalization strategies to address one of marketing's most pressing challenges: knowing when AI-powered personalization crosses the line from helpful to creepy. With extensive experience helping leading consumer brands implement sustainable growth transformations, Rathje offers practical insights on creating meaningful customer co ections without overstepping boundaries.

  • The Value Exchange Must Be Clear

    According to Rathje, successful personalization hinges on transparency. "The value exchange needs to be clear, and the why behind the treatment that a consumer is getting needs to be clear," she emphasizes. When customers don't understand why they're receiving specific content or offers, they often jump to conclusions that make the interaction feel more invasive than it actually is. This disco ect between marketing intent and customer perception creates the "creepy factor" that damages brand trust.

  • Real-World Example: Dynamic Billboards

    Rathje illustrates this point with dynamic billboards across Europe that use Bluetooth technology to push content to nearby app users. While the technology simply detects phones with specific apps installed, consumers often believe they've been personally identified and targeted. This misunderstanding highlights the critical need for marketers to educate customers about how their data is actually being used, preventing assumptions that could harm the brand relationship.

  • The Mad Libs Problem: When Data Becomes Performative

    Marketing teams often fall into the trap of showcasing their data collection capabilities rather than using that information strategically. Benjamin Shapiro describes this phenomenon perfectly: marketers feel compelled to prove they have information by awkwardly inserting personal details into communications. This "mad libs of data" approach, as Rathje calls it, creates uncomfortable interactions that provide no real value to the customer.
  • The rise of AI has amplified this problem, with automated systems mindlessly inserting personal information without considering context or relevance. Smart marketers understand that having data doesn't mean you need to display it. Instead, personalization should enhance the customer experience by providing relevant solutions to their specific needs.

  • Personalization as a Spectrum, Not a Destination

    One of the most valuable insights Rathje shares is that personalization exists on a spectrum. Not every interaction requires hyper-personalization, and forcing it can actually diminish the customer experience. Marketing leaders must consider where customers are in their journey and what level of personalization serves them best at each touchpoint.

  • Strategic Implementation Guidelines

    To implement personalization effectively, marketers should focus on three key principles. First, always co ect personalization to the context of your communication - if you're not a pet food company, mentioning someone's cat is irrelevant and creepy. Second, prioritize utility over demonstration - use data to solve customer problems, not to show off your data collection. Third, match personalization levels to customer journey stages - early interactions may require less personalization than ongoing customer relationships.

  • Conclusion

    The future of marketing personalization isn't about achieving perfect one-to-one communication at every touchpoint. Instead, success comes from understanding when and how to apply different levels of personalization based on customer needs, journey stage, and context. As Rathje emphasizes, judgment and understanding of marketing objectives will always matter more than technological capabilities. By focusing on clear value exchanges, contextual relevance, and appropriate personalization levels, marketers can leverage AI and data to create meaningful co ections without crossing into creepy territory.
About the speaker

Kathryn Rathje

McKinsey

- McKinsey

Kathryn Rathje is partner at McKinsey

CMO's Comeback Report
Related Podcasts by Category

B2B

See More

AI Personalization

See More

Marketing Consultant

See More

Artificial Intelligence, Customer Experience (CX), Marketing Strategy

See More

Up Next: