SEO and the AI-Powered Future of Search

by | May 17, 2024 | blog, Business, Uncategorized

The concept of search engine optimization (SEO) has always been a dynamic one, as online marketers strive to stay ahead of rapid internet growth and constant advancements in technology. Now, SEO seems to be on the precipice of another significant paradigm shift. And–in staying true to just about every other tech headline lately–this one is fueled by artificial intelligence (AI). 

More specifically, it’s about large language models (LLMs), the same type of technology behind OpenAI’s ChatGPT and similar natural language chatbots and virtual assistants.

Why Things Are Changing for SEO

In recent years, some internet users have been increasingly critical of the perceived lower quality of Google results, as more paid ads and even borderline spam content creep to the top of search results. As one common example, there are stories of music and sports fans running searches for event tickets and being funneled to up-charging reseller websites while official ticket outlets remain buried a half page down the search results. 

An influx of powerful AI–one that better understands language and what makes content useful to real humans–could be just what Google needs to turn this perception shift around.

Of course, there has been buzz about Google incorporating more AI into their search product for many years now, in addition to the ongoing debate about how (and how much) AI will change SEO. All that chatter has only increased with the advent of ChatGPT and other impressive natural language products. This isn’t just more dubious AI hype, however. Today, we’re looking at a specific patent recently filed by Google: United States patent US11769017B1.

Google’s New Patent and LLM-Integrated Search Summaries

Google’s new patent, US11769017B1, introduces a groundbreaking method for generating AI-powered summaries of search results, potentially ushering in a whole new type of search experience where a user receives a concise, coherent, and custom-generated summary of the results their query has generated. Presumably, this text summary would either precede or be integrated into a list of links to external websites, similar to what a traditional web search generates. Google has adopted the term “Search Generative Experience,” or SGE, for this new type of web search.

It seems a key element of Google’s SGE project will be creating these generative AI summaries or “snapshots” of search results. This will be achieved using large language models (LLMs). LLMs are powerful emergent technologies (think ChatGPT) that make a computer capable of interpreting both the content and context of natural human language to a greater degree than ever before.

Google aims to use the power of LLMs to generate these context-rich summaries of search results, potentially distilling dozens of different high-quality sources into a few key facts. This is nothing short of a complete reimagining of what a search results page is and how an internet user interacts with it. Implemented well, these summaries could give users a better understanding of a topic with a few brief sentences or bullet points, then corroborate those key points with links to the high-quality sources that were used to generate the summary.

Understanding the Debates (and the Lawsuits)

Of course, if these summaries are “too” good (a matter of perspective), they could circumvent the need for users to ever click through to other websites at all because they’re getting high-quality answers to their queries through generative AI summaries alone. This would maximize the time web searchers spend on the Google platform, potentially robbing traffic from the websites that were actually used to generate the summaries.

This sort of traffic-siphoning is already a topic of much debate (and a few lawsuits) because of the way Google currently serves up lists of flights, movie times, website excerpts, product prices, and all sorts of other useful data without ever sending users to the websites being used to source that data. Some worry that generative AI summaries could exacerbate this issue.

Evolving SEO Strategies for an AI-Powered Search Future

With Google’s next wave of search tech poised to embrace AI and LLMs in a major way, SEO and digital marketing strategies must adapt rapidly or risk losing traffic once this shift fully materializes. Somewhat ironically, this “AI takeover” could actually send SEO principles back to basics in a sense. You see, once Google’s products are able to parse and evaluate natural human language to a greater extent than they already do, the best way to get Google to rank your content highly should become fairly straightforward: have highly readable, genuinely useful content.

Here are some tips businesses and entrepreneurs can implement to ensure a seamless transition to SEO’s rapidly approaching AI era:

  • Prioritize Quality – Focus on crafting clear, authoritative content that offers comprehensive answers to relevant questions and does so in a readable and conversational (but still professional) way. Rather than spending hours trying to research the correct keyword combinations to poach traffic from your competitors, SEO will become more about ensuring that your content aligns with the actual intent of your business, organization, or website. Providing quality answers to relevant questions (ones that are legitimately within your area of expertise) will be the best way to increase your likelihood of being prominently featured in generative AI search summaries (and, in turn, getting those all-important click-throughs).
  • Write Clearly and Concisely – Make your content as clear and concise as possible while also providing the necessary context for a reader (or AI indexing bot) to interpret it coherently. Content written using straightforward, easily digestible sentence structures and simple formatting is more likely to be favored by LLMs for inclusion in summaries. Short sentences and bulleted lists can help.
  • Be Universally Accessible – Google’s AI routines are extremely powerful pieces of technology, but for your purposes as a website operator, it may be best to think of them as a member of the general public who has no preexisting knowledge of your field. For example, if you’re operating in a highly technical industry, marketing to an audience that reads at a very high level and doesn’t shy away from dense, complex posts, you may need to adjust your strategy for a better chance of showing up in generative search summaries. 

We’re definitely not suggesting that you “dumb down” your website to the point that it becomes a turn-off to your core audience, but you may also need to feature some content that is more plainly written, concisely organized, and universally accessible. This simpler content, designed with a specific eye toward web search, can then be used to funnel readers to the more niche technical content. This content will further establish your authority, but it may be beyond the grasp of today’s LLMs. 

  • Optimize for Conversational Queries – Keyword research will likely always have some importance, but LLMs are going to have more and more people running their web searches in a conversational, “natural language” way. Think of the questions that people looking for your products, services, or information would actually ask if they were speaking to an expert on that topic or if they were going to a librarian to ask for help finding information on it. Then, simply use your content to answer those questions correctly (and in an understandable and compelling way). This is often easier said than done, but nonetheless, it needs to become a core principle of your content strategy going forward.
  • Embrace Content Diversity – Search and AI are a pair of fast-moving tech concepts, and it’s anybody’s guess what the next big paradigm shift will be. With the future of search yet to be imagined, and today’s web users looking for info in all sorts of different ways, it’s a great idea to create content in multiple formats, including text, videos, and images. With podcasts continuing to be a hugely popular format, even audio content can play a key role in a comprehensive strategy.
  • Don’t Feed the AI Loop – It may seem counterintuitive on its surface, but one of the worst things you can do in response to an AI-powered Google search is to fill your own website with AI-generated content. While LLMs are great at generating something like a quick summary of multiple web pages, they remain extremely shaky at generating quality content. 

This is especially true when nuance, empathy, or even some types of specific technical knowledge are called for. This is because LLMs like ChatGPT and the forthcoming Google product cannot actually “understand” anything; they have merely been trained on the structures and patterns of human writing so that they can imitate them convincingly. So, when left to their own devices and asked to create something compelling, ChatGPT and other generative AIs tend to produce results that are predictable and mundane at best… and, at other times, bad enough to be downright scandal-inducing.

Future-Proof Your SEO Strategy with Better Content Matters

The integration of AI and LLMs into everyday web search marks a significant shift towards a more contextually aware, conversational search experience. For website owners and SEO professionals, adapting to these changes will be crucial.

The team of writers, editors, and digital marketing professionals at Better Content Matters can help. We will collaborate with you to design and implement a diverse and multi-faceted content strategy that prioritizes comprehensive, clear, authoritative, and accurate content, one that focuses on user intent and answering the actual questions your potential customer base is asking online.

In an increasingly AI world, BCM is still focused on providing world-class content custom-crafted by professional, well-trained human writers and editors. No matter what industry you’re operating in, we can provide you with content that supports your brand’s tone and boosts your industry profile while also aligning with the evolving landscape of Google search, SEO, and generative AI. Contact us today or follow our blog to learn more about what we can do for you.

Leo Siren

Leo Siren

Leo Siren is a freelance content creator from Michigan's Upper Peninsula drawing on his multifaceted experience as a public librarian, assistant harbormaster, financial software systems analyst, and forklift operator to deliver innovative, high performing content in a range of text and audio formats. His personal interests include the electric banjo, referencing Elder Scrolls lore in everyday conversations, attempting to identify wild mushrooms, and various other things that upset the people around him.