Remember Lycos? AskJeeves? Or using Yahoo search before Google launched in 1998? Some of us do! The search engine results page (SERP) has dominated our desire for fast answers, shopping, research, and even medical advice for 25 years. Then, in 2022, Generative AI and ChatGPT began to really change it all. Now, we can search OR “chat” and otherwise receive AI-powered results, usually in a conversational, interactive format.
Those bright blue links from 1998? A thing of the past.
Source: Business Insider
How Search is Changing
Google still, so far, dominates the internet, accounting for around 90% of the search engine market and vastly outranking direct competitors like Microsoft’s Bing. However, there have been signs of change for some time.
In 2021, TikTok briefly became the world’s most popular domain, and Google’s own data pointed to Gen Z preferring to discover information on platforms other than Google, like TikTok and Reddit. Google’s revenue from search advertising in the US is expected to decline to 54% from 67% in 2016.
Google announced updates earlier this year, including plans for AI-generated search results at its annual conference. Danny Sullivan, Google’s public liaison for search, says:
“We’ve continued to apply technologies like AI to enable new ways for people to search and make sense of what’s out there. Where there’s room to improve, we prioritize it, exemplified by our work to bring more helpful content from experts and everyday people into search.”
Trust in Google has been flagging. Despite constant updates to its search ranking systems, search results are plagued with spammy posts and content written purely for company SEO efforts. Internet users have increasingly been exploring other tools or amending their search criteria to find better results.
Businesses and publications that have invested millions into their online presence and getting it in front of customers, primarily through Google, are concerned about what the future might bring.
The SEO industry generated $68.1 billion in 2022. Expectations, before generative AI, estimated the sector would more than double in value to $129.6 billion by 2030. Now, the worry is the industry may be made rapidly obsolete.
Of course, it’s not so clear-cut. Generative AI results to date are often without clear source citations and can be provably false or inaccurate. Google is hardly going to shun completely a search advertising market likely to reach a near $300 billion figure in 2023. Then, there are copyright and data privacy concerns, as well as the broader risks associated with AI.
But, arguably, the internet is primed for generative AI-enabled search, and Google and Microsoft are rapidly iterating in a new battle to get ahead of tomorrow’s search behaviors.
A recent survey shared by Search Engine Journal found just under half of those asked were interested in AI-powered search, AI-generated recipes, AI-powered smart assistants, and even AI-designed products based on consumer trends.
AI in search will mean new ways of discovering information and, as yet, an undefined evolution of search ads and search optimization.
Google’s Search Generative Experience
Baron Insights, in October, noted that Google would likely maintain its lead in search because it has the largest set of consumer data.
To date, Google’s Search Generative Experience (SGE) has been an experiment in Google Labs only available in certain countries by opting in. Google considers Bard and SGE complimentary products, and both are now incorporating Gemini, Google’s new suite of large language models (LLMs), to help the search giant compete with OpenAI’s ChatGPT.
SGE focuses on delivering quick overviews of search topics so users don’t have to click on individual web pages. Here’s an example from Semrush:
Image Source: Semrush
On the right, Google displays the source which users can visit:
Image Source: Semrush
SGE is also designed to prompt follow-up questions. Here’s an example that follows a search for a cookie baking recipe.
Google’s AI search offering is conversational and more engaging, with logical follow-ups and useful summaries, but it has the same risks of inaccuracies as ChatGPT and other models, and for avid researchers, the sources and citations might be a little weak.
Image Source: Semrush.
Microsoft Bing Chat and Copilot
Microsoft has incorporated ChatGPT into Bing Chat, available right next to its search box, since February. In November, the chatbot was rebranded to Copilot and shares its name with all of Microsoft’s AI platform companions.
Copilot is much more conversational; responses have footnotes that link back to the bot’s sources.
Microsoft’s Copilot suite for businesses is comprehensive, and with the head start on a wide release of AI-powered search, the company is also hinting at future developments.
How Will Generative AI-Powered Search Play Out?
There’s no easy answer to the future of search. So far, AI appears to display more of its curated responses, potentially missing a multitude of other answers and sources that users might want to consider based on their preferences for sources of information.
SERPs list out reams of potential resources, where searchers can scroll and pick out sources of information that resonate with their needs. But has the SEO industry nullified the true relevance of these results, instead pushing content to the top that purely meets Google’s algorithm preferences, despite Google’s endeavors to rank “real” content higher?
The near future is likely to see savvy internet users use multiple tools and resources for the information they need. How LLMs are developed to overcome inaccuracies will be a significant influence in the long term. Then, there’s how big-spending businesses will be served with new opportunities to reach customers and how SEO or even the emergence of generative AI optimization to consider. Search Engine Journal asked ChatGPT if AI would kill SEO back in March 2023. Here is the response:
We’ve said it before: 2024 could be very interesting indeed.
