Transforming Search with AI: Google’s Strategic Shift

Ugo Smith

30 May 2024

captura de pantalla 2024 05 30 a las 11 37 43

A year ago, there was speculation both inside and outside Google that OpenAI and Microsoft’s AI chatbots could dethrone Google’s dominance in the search market. OpenAI had launched its GPT-3.5 model and publicly released ChatGPT in November 2022, which Microsoft then integrated into Bing shortly thereafter.

It was said that conversational search, which had revolutionised the way people interacted with AI, would divert users from Google, forever changing how we search for information.

What followed was exactly what one would expect from a company like Google. In response to the advancements by OpenAI and Microsoft, Google debunked the speculations at Google I/O 2023 (the company’s annual event) by introducing the Google Search Generative Experience (SGE), which integrated AI and conversational search into Google’s search results page. This announcement demonstrated how their model would continue to advance rapidly to defend their market share in the search and advertising space.

A year later, Google surprises again at Google I/O 2024: SGE has been renamed AI Overview. But this is not just a change in name; it also signifies a shift in focus and functionality.

From Google SGE to AI Overview: Comparison and Changes.

Google launched SGE in BETA with the aim of enhancing the search experience using AI and competing with OpenAI. This system focused on providing generative answers to complex queries, using advanced machine learning models to generate comprehensive and contextual responses. However, it had some limitations, especially in terms of accuracy in certain contexts and difficulties handling very specific queries.

The integration of SGE into search results was gradual, initially focusing on areas where it could improve query understanding. Although it never reached Europe, Good Rebels analysed its main functionalities a few months ago.

Now, with the introduction of AI Overview, Google incorporates significant improvements in technology and approach compared to SGE:

  • A Model Based on Gemini: AI Overview is powered by Google’s new Gemini model (an evolution of Bard), designed to handle complex queries more efficiently by integrating multi-step reasoning and multimodal capabilities, combining voice, text, and images in search.
  • Improved Result Generation: The AI now provides more precise and contextual summaries of the information requested by users, enhancing the understanding of complex queries. It also introduces filters for interacting with the information.
  • Content Visualisation: There is a greater focus on visual content, such as images and videos, to provide a richer and more dynamic search experience.
  • Personalisation: The AI has been optimised to better understand user context and preferences, offering more personalised results.
  • Reduction of Interaction: AI Overview eliminates the more conversational component of SGE, dispensing with the follow-up question functionality.
  • Reorganised SERP: The search results page has been reorganised to present search results more efficiently and provide more structured and relevant answers to user queries.
captura de pantalla 2024 05 30 a las 11 22 07
Before vs after SERP

captura de pantalla 2024 05 30 a las 11 26 28

New Opportunities for Advertisers.

In SGE, Google partially tested advertising, raising concerns about how this could endanger its lucrative advertising business. Now, it has shown that integrating conversational AI can not only maintain but also enhance this business.

With AI Overview, Google incorporates ads into AI-generated responses, offering new opportunities for advertisers to reach their audience more effectively using the shopping graph and related products.

Additionally, it provides a set of AI-powered tools to help advertisers create assets and ads more easily.

captura de pantalla 2024 05 30 a las 11 41 51 2

The incorporation of advertising in AI-generated responses can increase the relevance and effectiveness of ads. This benefits not only advertisers but also improves the user experience. Ads will be clearly marked and positioned in a way that distinguishes them from organic results, maintaining transparency.

Greater Control and Visibility Over Results.

With the introduction of SGE, many digital marketing professionals wondered how brands would have visibility over the impact their content generated if the customer journey shifted from being distributed across various websites to being centralised in the SERP, and how this would affect our reporting models.

Fortunately, several players are already considering or adapting their technologies to address the new challenge of measuring AI’s impact on search results.

Semrush and SEOMonitor have introduced new functionalities to track the positions of AI-generated results in the SERPs. These tools, similar to those that identify types of featured snippets or SERP varieties, respond to a growing market need. While much of the data on Google’s generative results also appears in Google Search Console, it is not separated and is mixed with all organic click data.

Additionally, as Google’s AI becomes more sophisticated, it becomes more important to understand how to control the information appearing in generative search results. Google has developed guidelines for removing content and links from AI Overview (General AI Overview), providing specific directions for webmasters. These guidelines include using tools like Google Search Console to manage the visibility of unwanted content in search results.

Technical Errors and Future Challenges.

As with any emerging technology (it is still in beta and only available in the U.S.), Google’s AI Overview is not without errors and challenges.

Since its launch, several cases have been documented where Google’s AI failed to generate accurate results or misinterpreted user queries. For example, suggesting using glue on pizzas to prevent the cheese from falling off, a response related to an old Reddit post that the AI used as a source.

These occurrences are part of the learning and adoption of the technology. However, it is crucial that the digital assets that feed the AI are of high quality and authoritative. This shows that SEO and content generation remain relevant. The question is at what pace AI Overviews will be deployed globally.

Conclusion: What Can Digital Marketing Professionals Do?

In digital marketing, we are used to rapid changes, but AI has accelerated this process even further. Specifically, the world of search is transforming at a dizzying pace, and the arrival of AI Overview is just one of many changes that will define how we interact with search engines in the future.

Changes that will undoubtedly improve the user experience, making it richer and more personalised, but also present new challenges and opportunities for digital marketing and SEO professionals.

In this constantly changing context, what optimisation strategies can we adopt?

  • Enhancing Traditional SEO: For certain search intents that Google seems not to prioritise in generative results, such as transactional or navigational searches (AI Overview generally appears more frequently in informational searches).
  • Focusing on YMYL Content: Topics that Google groups under the YMYL category (Your Money, Your Life, referring to health or personal finance content) will continue to be handled with special care in generative results, so brands needing to strengthen their positioning in these areas should continue to invest in traditional SEO.
  • Optimising Digital Assets: To ensure that our brand appears in generative results, adopting strategies to improve our website’s authority or simplifying content for easier comprehension.

At Good Rebels, we are working to provide a more complete view of the evolution of AI in online searches. Just as we did with the launch of SGE, in the coming weeks we will share a second study that will delve deeper into AI Overview. Stay tuned to Rebel Thinking!