natural language processing (NLP)

GAIO: The Next Generation Of SEO

Noa Eshed, owner of award-winning growth marketing agency Bold Digital Architects, co-host of the Real Life Superpowers podcast.

getty

Search engines have come a long way since they first appeared. In the beginning, search algorithms relied heavily on keyword matching, often leading to irrelevant results.

Then AI entered the stage, with models like Google's Bidirectional Encoder Representations from Transformers (BERT) providing a huge step forward. BERT, not to be confused with Google's AI search engine, Gemini, launched in 2018 and uses natural language processing to understand the context and nuances of queries in a whole new way.

Following The Progress Of Search Technology

BERT processes text bidirectionally. This means that it reads an entire sentence forward and backward in order to understand its context. Simply put, it's able to interpret user queries more accurately because it considers the relationship between words. For example, it can distinguish between different meanings of the word "bank" based on the surrounding words, leading to search results that are more relevant.

We've established that BERT made a big leap in contextual understanding. The next step in AI's evolution in search was retrieval augmented generation (RAG). By combining retrieval and generation models, RAG improves text quality. In order to generate contextually accurate and informative responses, it extracts relevant documents or passages from a large database.

RAG is able to fetch real-time information and inject it into search results. This means that instead of relying exclusively on pre-trained data, RAG can access up-to-date information from the web, making sure that answers are accurate and complete. For example, if a user asks about a recent event, RAG can pull information from news articles and provide a detailed, accurate answer.

Where is this felt? The answer is two-fold: The first area is in search generative experiences (SGE). Currently in its beta phase, SGE is expected to shift the way search results are presented (maybe even to an extent where those search results we are used to will become a thing of the past). The second area is in generative AI searches—people are searching for queries directly on ChatGPT, Gemini, Perplexity, etc.

So how can business leaders use this information to ensure their site is following the flow of new search technology and staying in front of consumers' eyes? I believe the answer lies in combining traditional SEO efforts with generative AI optimization (GAIO).

Understanding GAIO

Basically, GAIO is the process of optimizing content to show up in generative AI search queries. These language models create content that answers complex user queries in natural language. It is a conversation of sorts, and the content you provide needs to be understandable in a way that is relevant for these models to source from.

Meanwhile, traditional SEO is all about keywords, with advanced SEO also taking into account context and nuances. Here is a good place to clarify that Google uses human evaluators and behavioral analysis to improve "regular" search results. In fact, within the "traditional" search engines, "users tend to make more complex queries as search quality improves." So the differentiation isn't that straightforward.

Still, results in search engines are not conversational, and that is the key differentiator between GAIO- and SEO-driven search results to keep in mind. On top of all of the above, businesses still need traditional SEO in tandem with GAIO because even when we assume the differentiation between AI conversational results and standard search results will stand as-is, RAG will still search for the most relevant content within the top ten to twenty search results. So it's difficult to stand a chance of showing up in AI results if you're not ranked in the traditional results in the first place.

Now that we know the what, let's get to the how.

Steps For Integrating GAIO With SEO

Incorporating this next level of GAIO on top of your traditional SEO is all about creating content that answers complex user queries in natural language. This requires understanding the user's intent, providing in-depth answers and making sure your content is always up to date and relevant.

There are three main steps you can take to achieve this goal:

1. Focus on context. Make sure your content answers users' questions in context. Practically, this means evaluating what questions your content can answer so that you have an end game of showing up in GAIO queries related to those questions.

2. Stay current. Keep your content up to date so it's always relevant. This means that any content that isn't evergreen will need to be regularly revisited in order to make sure it's still the most accurate information available.

3. Optimize for natural language. Use a natural, conversational style that AI can easily understand; this helps make your content a more natural go-to for the GAIO model, reducing friction and becoming the answer to what the user and the AI are searching for. You can use tools such as the Flesch-Kincaude readability test to make sure your content isn't more complex than an 8th grade reading level. As the saying goes, "If you can't explain it to a 6-year-old, you don't understand it yourself." I would add on, "so don't expect AI to."

Conclusion

I think it's clear that the integration of AI in search will only deepen. As AI continues to transform search experiences, the integration of retrieval mechanisms with advanced language models shows the importance of producing timely, relevant and contextually rich content. I believe we can expect these models to become even more sophisticated and even better at understanding complex queries and retrieving more relevant results. This is why it's important to optimize your digital strategies to take into account both established and emerging AI-driven search technologies.

Forbes Business Council is the foremost growth and networking organization for business owners and leaders. Do I qualify?

Google Brings AI Agent Platform Project Oscar Open Source

Google has made a significant announcement in the realm of artificial intelligence (AI) by releasing its AI agent platform, Project Ocean, open source. This move is expected to have a significant impact on the development of AI-powered applications and services across various industries.

Project Ocean is a cloud-based platform that allows developers to build, deploy, and manage AI models easily and efficiently. It is designed to simplify the process of creating and integrating AI models into applications, making it more accessible to developers and organizations of all sizes.

The platform uses a combination of machine learning (ML) and natural language processing (NLP) technologies to enable developers to build AI-powered chatbots, voice assistants, and other conversational interfaces. With Project Ocean, developers can create highly personalized and interactive experiences for users, improving the overall user experience and driving business growth.

By releasing Project Ocean as open source, Google is empowering developers to modify and customize the platform to meet their specific needs. This move is expected to accelerate the development of AI-powered applications and services, as well as foster innovation and collaboration within the AI community.

"We're excited to bring Project Ocean to the open source community, and we believe it will help accelerate the development of AI-powered applications and services," said Tan May-Ann, a Google engineer and the lead developer of Project Ocean. "By making the platform open source, we're enabling developers to build on top of our research and innovations, and to contribute to the development of the platform themselves."

Project Ocean is built on top of Google's Cloud AI Platform, which provides a suite of AI-powered services and tools, including Cloud AI, Cloud Vision, and Cloud Natural Language. The platform is designed to be scalable and secure, with built-in support for data encryption and auditing.

The open source release of Project Ocean is expected to have a significant impact on various industries, including customer service, healthcare, finance, and e-commerce. By providing developers with a platform to build AI-powered applications, Google is empowering organizations to improve customer engagement, streamline operations, and drive business growth.

"Open sourcing Project Ocean is a significant milestone in our journey to make AI more accessible and usable for everyone," said Kaveh Sabr, a researcher at Google. "We're excited to see the innovative applications that our developers will build using this platform, and we're committed to working with the open source community to ensure that Project Ocean continues to evolve and improve."

The open source release of Project Ocean is a testament to Google's commitment to advancing AI research and development. The company has a long history of investing in AI research and innovation, and has made significant contributions to the development of AI-powered technologies.

With the release of Project Ocean, Google is opening up new opportunities for developers and organizations to build innovative AI-powered applications and services. As the platform continues to evolve and improve, it is expected to have a significant impact on the future of AI development and innovation.

7 Ways Medical Marketers Can Navigate The Rise Of 'zero Click'

It's a 'zero click' world and we're just living in it.

In May, Google launched its Gemini 1.5 Pro model in the form of a "zero click search," meaning that answers to searches appear without users clicking on the links. 

The AI-based model synthesizes information from different websites and sources to provide users with quick answers. 

Though this may seem like a promising idea on paper, its practical application has garnered considerable controversy.  

Many of these searches deliver hallucinations, also known as incorrect AI-generated results. On an anecdotal basis, social media users have identified instances over the past few months in which they were fed information from 'zero click' searches that were incorrect.

These include claims that Batman is a police officer as well as the assertion that glue must be used to get cheese to stick to pizza.

Despite the humorous nature of many of these hallucinations, the model also has the power to generate some potentially dangerous misinformation, especially when it pertains to health. 

As more companies integrate similar AI models into their search engines, it becomes more imperative for health brands and medical marketers to adjust their practices to optimize customer outreach and promote factual information. 

Kristin Ryan, the head of U.S. Digital and Innovation at GCI Health, a 2024 MM+M Agency 100 honoree, offered seven best practices for how healthcare marketers can combat misinformation in this evolving digital landscape. 

Here are the seven takeaways.

1. Focus on being discoverable 

Medical marketers should create and publish content that's accurate, clear and high quality in order to cater to the AI-driven search algorithms. 

Ryan said that all content should be decisive and communicate the core beliefs of the brand, noting that clarity is key to ensuring the algorithm understands the content and amplifies it. 

She also advised against using superfluous language – such as the word 'superfluous.'

Keeping it simple allows for a broader audience outreach given that shorter, basic words cater to natural language processing systems. 

The same goes for the use of Google ads: the content needs to be clear and consistent with the website itself. This will increase the score of a website, she said, which is the basis by which content is sorted by AI-powered search engines.  

2. Image quality matters

Since AI models incorporate images as a part of explanations, images and videos must be clear, high quality and relevant. 

Images are pulled from different sources across the web, so the written content discoverability rules apply to visual assets as well.

3. Prioritize credibility

Now more than ever, it's critical to build trust with consumers. 

The rise of 'zero click' has opened the door for even more misinformation online, so health brands should be citing and linking to reliable sources to combat noise and irrelevant content. 

4. Don't fall asleep at the wheel

Monitoring content and remaining responsive will be crucial to ensure optimal outreach to target audiences, Ryan noted.

That means brands and marketers must have a team continuously monitor how AI tools present content so that any misrepresentations or misinformation can be quickly corrected.

5. Incorporate a schema markup

For those who may not know, schema markups are the way search engines read information – these tools don't digest content the same way that users do. Instead it's in the format of code, so this requires a technical approach to content. 

The schema markup helps translate the content to code, allowing search engines to better understand and categorize the content more appropriately. 

6. Build an interconnected online presence

Ryan contended that each website should have an "about us" page, a feature that establishes it as a business entity and links back to other social media profiles. 

The creation and continuous update of social media profiles with consistent content and links back to the website caters to google's knowledge graph. 

7. Last but not least, don't panic

With the emergence of any innovation, certain challenges arise and adjustments must be made. 

The industry is no stranger to growing pains, Ryan said, but stakeholders need to take proactive steps to avoid being caught flat-footed.  

"If companies go into this strategically and make sure they are making the right changes it will ultimately improve the search outcome and also improve the user experience," she said. 

---