In a significant shift that is fundamentally altering the landscape of local search engine optimization (SEO), recent data and real-world case studies suggest that traditional star ratings are no longer the primary factor determining which businesses appear in artificial intelligence-driven search results. This phenomenon was recently highlighted by Annie Jackson, Director of Revenue Operations and Growth at GatherUp, and Jason Wertham, Vice President of Review Defense Operations at GatherUp, during an industry session focused on the evolution of local discovery. The presentation detailed a scenario in which a car wash with a 3.3-star rating outperformed higher-rated competitors in a Google AI search because the business’s specific attributes matched the user’s complex query more accurately than its reputation score.
The case study centered on a query Jackson performed in Norfolk, Virginia, where she asked Google for a "no-touch car wash that fits an SUV." Rather than returning a list of the highest-rated car washes in the area, the Google AI Overview prioritized a business with a 3.3-star rating. Crucially, the AI highlighted the business’s clearance height and 24/7 operating hours directly above the star rating, effectively answering the user’s specific logistical needs before addressing the quality of the service. This indicates a pivot in search engine logic: query matching and specific contextual data are now outranking aggregate review scores in the generative AI era.
The Evolution of Consumer Search Behavior
The shift in how businesses are surfaced is a direct response to a transformation in consumer behavior. According to GatherUp data collected in the fall of 2025, consumers are rapidly moving away from simple keyword searches like "car wash near me" toward conversational, multi-layered inquiries. The survey revealed that 55% of consumers have consulted Google or Bing AI summaries to find local information, while 48% have used ChatGPT to inquire about specific businesses. Furthermore, 31% of respondents reported using these AI tools multiple times for local discovery.
This trend toward "long-tail" conversational queries—such as asking for a specific vehicle fit or a pet-friendly policy—allows Large Language Models (LLMs) to leverage their vast datasets to find exact matches. Google, for instance, utilizes its database of over 300 million places and 500 million review contributors to synthesize answers. The AI’s ability to factor in user-specific context, such as the type of vehicle owned or past search history, means that search results are becoming increasingly personalized and dynamic.
Jason Wertham noted that the "slot machine" nature of AI search results means that visibility is no longer a static ranking. Factors such as the time of day, the user’s current location, and even the specific device used can influence which businesses are presented in an AI summary. This variability underscores the need for businesses to move beyond traditional SEO metrics and focus on a broader strategy of digital presence and data consistency.
The Technical Barrier: Why Reviews Must Be "Evangelized"
One of the most critical technical insights shared by Jackson and Wertham involves how LLMs access and process review data. Currently, major directory platforms such as Google and Yelp block LLM crawlers from scraping review content directly from business profiles. This creates a disconnect: while reviews on these platforms still influence traditional local rankings and consumer trust, they are often invisible to AI tools like ChatGPT or Claude unless that data is mirrored elsewhere.
"The major directory service providers… do not allow LLM tools like ChatGPT and Claude to scrape or crawl the review data on the business listing," Wertham explained. To overcome this, businesses must republish their reviews on crawlable surfaces. This process, which Wertham calls "evangelizing" reviews, involves embedding review widgets on the business’s own website or posting review highlights to public social media channels. Once review content is hosted on a public, crawlable URL, it becomes "fair game" for LLMs to incorporate into their summaries.
This strategy is particularly vital for winning queries that include subjective terms like "popular," "highly recommended," or "most reliable." If an LLM cannot crawl review text to find these descriptors, it will likely omit the business from its recommendations, regardless of how many five-star reviews the business has on a locked directory.
The Build, Manage, Defend Strategy
To navigate this new reality, Jackson and Wertham proposed a structured "Build, Manage, Defend" rollout designed to optimize a brand’s AI narrative.
Build: Establishing the Foundation
The first phase involves creating a consistent and comprehensive digital footprint. This includes ensuring that listings across all platforms are accurate and that the business is actively generating a steady stream of new reviews. Consistency is paramount; a single discrepancy in a phone number or address—such as a restaurant listing an owner’s personal cell phone on a secondary social page—can confuse AI models and lead to incorrect answers or lost leads.
Manage: Responsiveness and Velocity
Management focuses on the "72-hour window." Consumer data suggests that 70% of customers prefer to receive a review request within 72 hours of a transaction. Furthermore, 45% of users now prioritize the recency of reviews over the aggregate star rating. Wertham emphasized that a business with 1,000 reviews and a 4.0-star rating is often viewed more favorably by both AI and humans than a business with a 5.0-star rating based on only 30 reviews, particularly if those reviews are old. Review "velocity"—the speed at which new reviews are posted—serves as a proxy for the current quality of the business.
Defend: Protecting the Reputation
The defense phase involves monitoring for policy-violating reviews and employing tactics to mitigate the impact of unfair negative feedback. Wertham introduced the concept of "review smothering," where a business focuses on high-velocity positive review generation to bury older, less relevant negative content. Additionally, businesses are encouraged to actively dispute reviews that violate platform policies, as even a decade-old review can continue to influence AI summaries if it contains high-value keywords or has received engagement (such as emoji reactions) from other users.
The "AI Slop" Penalty and Content Quality
As businesses attempt to optimize for AI, Google has introduced countermeasures to maintain search quality. A significant recent update involves the "AI slop" penalty. Google is now actively penalizing businesses that populate their websites with low-value, AI-generated content intended solely to "game" the system. This includes generic blog posts, poorly constructed FAQ pages, and scraped content.
Wertham warned that these "glorified FAQ scraping targets" now carry a risk of decreased visibility. Instead, the focus should be on high-quality, original content that provides genuine value to the user. This aligns with Google’s broader "Helpful Content" guidelines, which prioritize expertise, authoritativeness, and trustworthiness (E-E-A-T).
Implications for Multi-Location Brands and Franchises
The challenges of AI search are amplified for multi-location brands and franchises. In these models, a "consistency gap" often exists between the corporate brand and individual franchisees who may manage their own local profiles. Because AI tools synthesize a brand’s reputation from a variety of sources, a single poorly managed location can negatively impact the AI-generated summary for the entire brand.
To combat this, Jackson suggested that franchisors should run regular AI audits on behalf of their franchisees. By using a standardized set of prompts—ranging from brand-name queries to location-specific spot checks—corporate offices can identify where the AI narrative is failing and provide coaching or white-labeled tools to help franchisees align with the national brand strategy.
Analysis of the Shift from Positioning to Presence
The transition from traditional search to AI-driven discovery represents a shift from "positioning" to "presence." In the traditional model, SEOs fought for the "top three" spots in the local map pack. In the AI model, the goal is "total citations"—the breadth and depth of sources that confirm a business’s attributes.
Jackson noted that positioning is no longer a reliable metric because LLM answers behave like a slot machine. A brand might lead the answer on one device but be entirely absent on another. Therefore, the strategic objective is to ensure the brand appears in as many relevant summaries as possible by feeding the AI a consistent, crawlable, and data-rich narrative.
Small facts, such as hours of operation or contact details, tend to update quickly in AI models—often within days. However, a business’s "reputation narrative"—what it is known for—takes longer to shift, generally requiring two to four weeks of consistent data signals to move the needle. This necessitates a long-term, disciplined approach to reputation management rather than a reactive one.
Conclusion and Actionable Outlook
As AI continues to integrate into the core of the search experience, the traditional obsession with maintaining a perfect 5.0-star rating is being replaced by a more nuanced focus on data accessibility and review recency. Businesses that succeed in this new environment will be those that treat their reviews as portable assets, ensuring they are visible not just to human shoppers on directory sites, but to the AI crawlers that are increasingly making decisions on behalf of those shoppers.
The immediate takeaway for local business owners is to conduct an "AI audit" using incognito or temporary-chat modes to see what ChatGPT and Google AI Overviews are currently telling customers. By identifying the gaps between the desired brand narrative and the AI’s current summary, businesses can begin the work of "evangelizing" their data to ensure they win the AI answer, even if their star rating isn’t perfect.
