There has recently been a fair bit of drama about Google\u2019s patent about generating landing pages instead of sending users to websites. While it\u2019s true that the patent is about creating landing pages it\u2019s absolutely not true that it\u2019s about doing it for all pages that are low quality. In fact, the patent only describes a shopping page feature.<\/p>\n
Google Is Not Replacing Sites With Landing Pages<\/p>\n
Patents are not always clear about what they are about because they are often broadly worded in order to apply to the widest amount of use cases. And that\u2019s the case with this patent that\u2019s titled, AI-generated content page tailored to a specific user. That title offers no context for how the invention is intended to be used.<\/p>\n
Glenn Gabe recently posted about the patent, repeating what someone else said about it on LinkedIn.<\/p>\n
He tweeted<\/a>:<\/p>\n \u201cIf you thought AIOs angered people, just wait for AI-generated landing pages from Google. Yes, Google could create new landing pages from the SERPs if yours isn\u2019t good enough (based on this patent). Great catch from Joshua Squires at Amsive on LinkedIn -> AI-generated content page tailored to a specific user<\/p>\n And sounds like it could be used for advertising AND organic\u2026 It\u2019s worth checking out the patent.<\/p>\n *The system would calculate a \u201clanding page score\u201d for the organization\u2019s existing website. This score evaluates how well the current page meets the user\u2019s specific needs. Glenn is correct that it could be used for advertising. But the patent is not about generating landing pages for general search results.<\/p>\n A careful reading of the patent shows that it is really about creating a landing page for shopping search results when the landing page presents a poor user experience. The patent indicates that this can be applied to shopping pages, with much of the it implying or specifically saying that it\u2019s intended for paid advertising use cases.<\/p>\n The paid advertising use case makes sense because it will help advertisers with low-conversion landing pages make more sales. The patent makes multiple references to that reason for generating a landing page.<\/p>\n What The Patent Says<\/p>\n The patent in places makes references to websites using generic words like \u201corganizations\u201d and \u201ccontent providers,\u201d but every example, metric, and UI feature aligns with:<\/p>\n E-commerce sites The specific examples of what will trigger the generated landing pages are:<\/p>\n Conversion rate, bounce rate, click-through rate as scoring signals Patents generally begin with a broad overview of the invention, a section called an Abstract. However the more you read the more specific it becomes and that\u2019s the case with this patent. Near the top of the patent, in a section called Background, the patent precisely describes the use case for the patent:<\/p>\n \u201cIn some instances, the landing pages may be difficult to navigate, which can reduce the user experience. For example, a user may struggle to navigate a landing page to purchase a product when the landing page has a user interface that is not efficiently designed for usability and engagement.\u201d<\/p>\n Elsewhere the patent explains:<\/p>\n \u201cIn some instances, the navigation link can be included in a sponsored content item. There is no ambiguity that this patent is focused on shopping related content and for sponsored advertising.<\/p>\n Not For Editorial Content<\/p>\n There are no concrete examples in the patent that are related to editorial content, news sites, academic pages, blogs, or informational content. The clues overwhelmingly point to transactional, commerce-oriented pages.<\/p>\n Read Google\u2019s Patent<\/p>\n I have written a guide that teaches how to read a Google patent<\/a>. It\u2019s not that hard to learn and is useful to read. Every SEO should give it a read because it will be helpful for understanding how Google\u2019s various algorithms work. Read it here<\/a>.<\/p>\n
*If the existing landing page is deemed insufficient (the score exceeds a certain threshold for improvement), the system triggers the AI generation.
*An AI model leverages the user\u2019s search context, location, and past preferences to pull data from the organization\u2019s site and re-package it into a new, intuitive interface.
*The user sees an updated search result page with a navigation link leading to this custom AI-generated page rather than the standard URL.\u201d<\/p>\n
\nProduct listing pages
\nRetail landing pages
\nPaid search environments
\nConversion-focused commercial websites<\/p>\n
\nCall-to-action buttons on product pages
\nProduct feeds
\nExample of a shopper struggling to navigate a page to purchase a product
\nSponsored content placement
\nLanding pages that lack product filters (which make it hard for consumers to find their product).<\/p>\n
\nIn some instances, the landing page score can be determined based on a conversion rate associated with the first landing page.
\nIn some instances, the AI-generated page can include a call-to-action button to a product page\u2026\u201d
\nIn some instances, the AI-generated page can include a product feed that provides an overview of a product\u2026
\nIn some instances, the AI-generated page can include a sitelink to a product detail page.\u201d<\/p>\n