AI Assistants Are Choosing Local Businesses For Your Customers via @sejournal, @MattGSouthern
All Paid Media PPC News Social MediaAdvertising Video Advertising Columns Ask A PPC ExpertNEW PPC Pulse Rundown Why AI volume alone can't deliver personalization Your content architecture may be the real bottleneck. Get the Rundown on what needs to change in your content ops. Digital Marketing All Digital Marketing Analytics & Data Ecommerce Lead Generation Content Marketing Social Platforms Google YouTube Reddit LinkedIn TikTokNEW WordPress Other CMS Platforms Webinar New AI Search & SEO KPIs: 4 Real Signals AI mentions and citations are benchmarks, not decisions. Get 4 traffic-predictive signals drawn from real bot data across hundreds of sites. SEJ Live Webinars Rundowns PodcastNEW Ebooks All Resources SEJ Live AI Search Visibility for Local Business August 26 | How ChatGPT, Gemini, and AI Overviews choose a single local recommendation - plus reviews, GBP, and attribution. AI assistants narrow local searches to a few names before customers reach a website. Here's what Google says about how those answers get built for local. SEJ STAFF Matt G. Southern 18 hours ago ⋅ 8 min read SEJ STAFF Matt G. Southern Senior News Writer at Search Engine Journal Bio Follow 1.2K READS AI assistants can now handle the comparison work local customers used to do themselves. Local search used to leave that work to the customer: someone typed a query, got a list of options, opened a few, and chose. Ask an assistant for a business recommendation now, and it returns a short list it has already selected. Google describes AI Mode as particularly helpful for questions that need exploration, reasoning, or complex comparisons, the kind it says might previously have taken several searches. In organic search, a business could check its ranking and know roughly where it stood. Google’s generative AI performance report shows how often links to your site appeared in its AI features. What the report doesn’t show is the query that led to each impression, or what happens on other assistants. Listings, reviews, and reputation can all affect whether an assistant considers a business in its responses at all. I covered those signals in Reviews, Reputation & Listings: The Local Signals AI Now Reads, and none of it has changed since. Google says its AI features in Search are rooted in the same ranking and quality systems as regular results. It names two techniques specific to the process of generating answers. Retrieval-augmented generation (RAG), which Google also calls grounding, uses existing ranking systems to pull relevant pages from the Search index. The AI generates its answer from what those pages say, while linking to the pages that support it. Query fan-out is a technique that involves running multiple related searches at the same time, gathering more results than the original question would return on its own. To get the content on a business’s website to contribute to Google’s AI responses, it has to sit on a page Google can retrieve. If someone asks for a quiet restaurant for a client lunch, that has to be described somewhere Google can read, such as in the reviews or the description. When Google pulls multiple sources with different information about the same business, it doesn’t document how it settles the difference in a given answer, so it helps to ensure the same information appears across every page you control. Uberall says its quick-service restaurant benchmark typically produced three to five recommended brands per query. That’s one vendor reading one industry, and it gives a sense of how small the field gets. Most local SEO efforts revolve around ensuring a business is well-qualified to be recommended. Keeping listings accurate, maintaining an active review profile, and ensuring consistency across sources all contribute to this eligibility. While meeting Google’s requirements is important, it doesn’t automatically guarantee that your content will be featured in its AI-driven features. Google folds AI Overviews and AI Mode data into the main Performance report, under the Web search type, mixed in with everything else. Those clicks are in the totals and can’t be separated out there. The Generative AI performance report does separate them, and Google is still rolling it out, so not every property has it yet. It counts impressions, meaning the times links to a site appeared in one of Google’s generative AI features, and it breaks those down by page, country, device, and date. There’s no query dimension, so an impression says a link to a page showed up and nothing more. It doesn’t distinguish an answer that recommended the business from one that recommended a competitor and listed the business underneath as a source. That report covers Google only. Nothing in Search Console reports what ChatGPT, Perplexity, or Claude told someone asking which local business to use. Whitespark’s analysis, which I covered in July, tested 540 queries across three U.S. cities and six industries. It found AI Overviews on 15% of direct local-intent queries, 92% of informational ones, and 97% of hybrid ones, questions like “should I hire a lawyer after an accident” that carry a purchase decision inside an information request. Those are the queries where an AI answer can shape which businesses a customer considers. As we wait for better reporting, teams are gauging performance two ways. Some run a fixed set of local queries against the major assistants on a schedule and record which businesses get named. That covers the queries on the days they ran, and assistant answers change with phrasing, location, and session, so the numbers are directional. Another way businesses are gauging performance is by separating AI referral traffic in analytics, where that data exists, and tracking volume and behavior over time. That shows what happened after someone clicked through, which is a smaller group than the people who saw the answer. Google’s guidance also warns against third-party tools that claim access to its internal ranking or AI systems. Each method measures a different part of the customer journey, leaving a different blind spot. Google’s optimization guide describes how to best position content for inclusion in its AI features. First, a site has to be included in Search generative AI features to be eligible to appear in them. Google is rolling that setting out to some site owners and not others. Include is the default. A child property inherits from the closest parent with a manually configured setting unless the child overrides it. If the setting is available, it’s worth confirming the site hasn’t been excluded, and businesses running each location as its own property should check those too, since a location property can inherit an exclusion from the domain above it. Google’s guidelines also recommend keeping crawling permissions open in robots.txt and at the CDN, and making important content available as text. They emphasize the importance of structured data that aligns with the visible page content, along with current Business Profile details. Google can process JavaScript content when it isn’t blocked, although its guide says JavaScript-framework sites are generally more complex to work with. Each step of rendering is another point where a business’s hours could fail to appear. AI responses can also pull from pages that a business has no control over. They might turn up an outdated directory carrying last year’s hours, an old article, or a competitor’s comparison page describing the business in the competitor’s terms. The testing I covered in July shows what can happen when those descriptions aren’t accurate. Make sure the hours, services, and service area are clearly listed in plain text on the site, matching what’s on the listings. This way, you provide Google with accurate information directly from the business itself. Google’s optimization guide also clears up some common misconceptions. Google Search ignores llms.txt and similar AI text files, and there’s no need to break content into small chunks for Google’s AI systems to understand it. Google describes AI agents as systems that carry out tasks for people, like booking a reservation or comparing product options. It says browser agents may access a site directly, analyzing rendered screenshots, inspecting the DOM, and interpreting the accessibility tree that screen readers rely on. That last one makes it important to check whether a site has accessibility gaps. A button built as an unlabeled div, a form field identified only by its placeholder text, or a phone number that exists only inside an image are problems for screen readers. Google hasn’t said those defects will stop an agent, but the accessibility tree is one of the layers it says browser agents may interpret.
Source: Search Engine Journal
This article has been carefully curated and reformatted for educational and informational purposes. Full credit goes to the original publisher.
📚 Visit more helpful articles on Joab Peters Blog
No comments