How AI Is Changing Shopping

Key Concepts

• AI Shopping Assistants: Platforms like ChatGPT acting as personal shoppers for consumers.
• SEO (Search Engine Optimization): Traditional method of optimizing content for search engines like Google.
• AI Agent Optimization: New strategy of optimizing content for AI agents to surface in search results.
• Direct Relationship with Customers: Retailers' goal of maintaining customer loyalty through their own websites and apps.
• Instant Checkout: Seamless transaction process facilitated by AI, bypassing retailer websites.
• Data Scraping: AI bots accessing and collecting data from websites.
• Large Language Models (LLMs): The underlying technology powering AI shopping assistants, requiring detailed and authoritative content.
• Metadata: Detailed information about products crucial for AI to understand and surface relevant results.
• Consumer Feedback: The role of platforms like Reddit in shaping AI-driven product recommendations.

Impact of AI Shopping Assistants on Holiday Retail

This discussion, featuring CNBC retail reporters Gabrielle Fonrouge and Melissa Repko, explores the significant and potentially revenue-driving impact of AI platforms like ChatGPT on the upcoming holiday shopping season. Analysts are predicting that shoppers will increasingly turn to these AI-powered "shopping assistants" for inspiration, akin to having a personal shopper.

Shift from Traditional SEO to AI Agent Optimization

For two decades, marketing strategies have been heavily focused on SEO, with efforts primarily directed at optimizing content for search engines like Google. However, the emergence of AI shopping assistants is fundamentally altering this landscape. The focus is shifting from optimizing for search engines to optimizing for "agents" – the AI systems that will be surfacing product recommendations.

• Key Point: Marketers need to adapt their strategies from writing for human searchers to writing for AI agents.
• Technical Term: SEO (Search Engine Optimization) - The practice of increasing the quantity and quality of traffic to a website through organic search engine results.
• Technical Term: AI Agents - Software programs that can perform tasks autonomously, in this context, acting as intermediaries for shoppers.

Challenges for Retailers and Brands

Retailers and brands invest heavily in their own websites and apps to foster direct relationships with customers and build loyalty. The rise of AI shopping assistants presents a challenge as consumers may bypass these platforms entirely, going directly to AI tools that then determine their purchasing decisions.

• Argument: Retailers risk losing direct customer relationships and loyalty if consumers skip their owned platforms.
• Real-world Application: The concern is that AI, not the retailer's curated experience, will dictate what shoppers find.

Elevating the In-Store Experience

The personalized experience offered by in-store employees has traditionally been a key differentiator for retailers. If AI can replicate or surpass this level of personalization online, it raises the bar for the in-store experience and puts pressure on retailers who fall short in customer service.

• Argument: AI-driven online personalization could expose weaknesses in traditional in-store customer service.

Data and Consumer Sentiment

While data suggests significant growth and potential revenue generation from AI in retail, consumer sentiment is mixed.

• Data Point: Adobe data indicates a substantial number of transactions originating from ChatGPT, showing an increase in conversion rates.
• Data Point: Salesforce data projects that AI is expected to drive billions in revenue this holiday season.
• Consumer Perspective: Some consumers find AI shopping assistants make holiday shopping more enjoyable, transforming a "chore" into something "fun."

Retail Industry's Technological Lag

A recurring observation is that the retail industry has historically been slow to adopt new technologies. Many existing systems, such as inventory management and point-of-sale (POS) systems, are described as being from the early 2000s. The introduction of AI further complicates these existing challenges, particularly in inventory management, which is considered a fundamental aspect of retail operations.

• Argument: The retail industry's outdated technological infrastructure makes it difficult to adapt to the rapid advancements in AI.

Mixed Success of Retailer-Developed Chatbots

Many companies have developed their own chatbots, but the success has been varied.

• Example: Target's chatbot, tested on Black Friday, was described as "not that good."
• Key Point: These chatbots often fail to provide personalized answers, leading to frustrating, scripted interactions, similar to traditional customer service chatbots.
• Example: A chatbot failing to recommend a book for a four-year-old niece, instead directing the user to a gift list, illustrates the lack of tailored responses.
• Conclusion: There is significant room for improvement in the functionality and personalization capabilities of retailer-specific chatbots.

Divergent AI Strategies: Walmart vs. Amazon

Major retailers are adopting different approaches to integrating AI and dealing with AI bots.

• Walmart's Strategy: Walmart has partnered with OpenAI and ChatGPT to enable instant checkout. This allows customers to discover a Walmart product and complete the entire transaction without visiting Walmart's website.
  • Technical Term: Instant Checkout - A streamlined purchasing process facilitated by AI, allowing immediate completion of a transaction.
• Amazon's Strategy: Amazon has taken a contrasting approach by actively blocking many AI chatbots from scraping its website and appearing in their listings. Amazon is developing its own proprietary AI tools.
  • Technical Term: Data Scraping - The automated extraction of data from websites.

The New Frontier: AI Version of SEO

The focus is now on how to ensure products appear prominently in AI-generated search results, essentially an "AI version of SEO."

• Key Argument: The strategy for appearing in AI results differs from traditional SEO.
• Interviewee: Target's Chief Information and Product Officer stated, "you used to write for a person, now you're writing for an agent."
• Methodology: To appear in AI results, content needs to be detailed and rich in metadata, unlike traditional SEO which focused on keywords for human readers who prefer less text.
  • Technical Term: Metadata - Data that provides information about other data. In this context, it refers to detailed product descriptions and attributes that AI can process.

The Need for Detailed and Authoritative Content for LLMs

Large Language Models (LLMs) require extensive and authoritative information about products to effectively serve user queries.

• Key Point: Keywords are less important for AI than detailed, authoritative content.
• Example: A soap brand created blogs with detailed skincare information. After implementing this strategy, they observed a double-digit increase in search traffic from LLMs.
  • Technical Term: LLMs (Large Language Models) - A type of AI algorithm that understands and generates human-like text, forming the basis of many AI assistants.

A New, Unregulated Playing Field

The current landscape of AI in retail is described as a "new playing field" with evolving rules and boundaries that are not yet established.

• Analogy: It's not a "Wild West," but a new frontier requiring the development of new norms and regulations.

The Potential for Consumer Feedback to Influence Results

There's a possibility that consumer feedback will gain prominence, influencing what AI surfaces. Platforms like Reddit are mentioned as sources from which this feedback is drawn.

• Argument: If consumer feedback is prioritized, the best products may naturally rise to the top.
• Perspective: This could lead to a refreshing scenario where quality and personalization are the winning factors.

The Unanswered Question: Impact on the Shopper

The ultimate impact of these AI-driven changes on the shopper remains a significant and unanswered question. It is still too early to determine the full extent of this transformation.

Synthesis/Conclusion

The integration of AI shopping assistants like ChatGPT is poised to revolutionize the holiday retail season, potentially generating billions in revenue. This shift necessitates a fundamental re-evaluation of marketing strategies, moving from traditional SEO to optimizing content for AI agents. While retailers face challenges in maintaining direct customer relationships and adapting their often-outdated technological infrastructure, the potential for enhanced personalization and a more engaging shopping experience is significant. However, the effectiveness of current AI tools, particularly retailer-developed chatbots, is mixed, highlighting a need for further development. Retailers are adopting diverse approaches, from seamless integration like Walmart's instant checkout to protective measures like Amazon's blocking of AI bots. The future of retail will likely depend on the ability of brands to provide detailed, authoritative content that LLMs can leverage, and the evolving role of consumer feedback in shaping AI-driven product recommendations. The long-term impact on the shopper is still unfolding, making this a dynamic and critical area to watch.