What’s the Difference?
Revolutions come and go. The American revolution was at one time considered the revolution to end all revolutions, at least until the French revolution, and the Russian revolution. There were also less definitive ‘revolutions’, like the Industrial revolution and the digital revolution, but there is a more subtle revolution happening right now, and while it might not have the political implications the American or French revolutions had, it is a very powerful one.
There are roughly 13.7 billion internet searches done every day or 1.664 searches for every person on our planet! Besides spawning companies Google (GOOG), Yandex (pvt), Baidu (BIDU), and Yahoo (pvt), internet search has created an ~$79b industry that looks to help companies improve their search results (SEO). It’s what we have become used to over the last 27 years since Page and Brin incorporated in 1998. During that period search engines have come and gone (Alta Vista, InfoSeek, Ask Jeeves, etc.) but the concept continued to evolve and get refined, even if Google continues to dominate.
Chatbots, like ChatGPT (OpenAI), Claude (Anthropic), and Perplexity (pvt), came on the scene at the end of 2022, but made no impact on internet search as they were trained and relied entirely on static data for their ability to answer queries. If you asked a question about something that happened after the model’s training end date, you got an answer like “As an AI language model, my knowledge is based on the data I was trained on, which has a cutoff date of [Month, Year]. Therefore, I do not have information about events or developments that occurred after that time”.
Things changed in 2023, with the release of Bard/Gemini, which included native internet search capabilities (Google Index), and ChatGPT’s Search (Bing Index), and while no tea was dumped in Boston harbor and there was no storming of any Bastille-like structures, a revolution began. First, it was summaries, where AI systems would put together the most logical answer to a query in the form of a one or two paragraph summary, and then came query answers directly from chatbots that were no longer limited to only using their training data as information sources, but without attribution. That did not last long, as it only took a few instances where AI chatbot answers were obviously wrong to push Ai designers to add sources and references to results in order to regain credibility. As of today traditional search engines still account for ~97% of all internet searches, with Google alone processing between 13.7 and 14 billion searches each day, but AI chatbot searches, as of April, have grown almost 81% y/y, albeit from a small base, but that is still a lot of growth.
There is a problem. When search engines are used for queries, the results are displayed in order of how the search engine ranks sites in its index, a science that many believe they understand, although only Google knows for sure. The entire industry functions around SEO (Search Engine Optimization), with billions spent to try to game the system enough to move one up in search results which can mean millions in advertising revenue. But with Google search now using query summaries, the quick answer for many queries is in that 1 or 2 paragraph summary and the ‘ranked’ search results only appear below the summary. In many cases (we don’t quite believe the numbers yet) the search goes no further and no site is selected..
In January of this year 6.49% of Google queries resulted in AI overviews, which increased to 7.64% in February and 13.14% in March. It peaked at 20.22% in April and sagged slightly in May (19.88%), although some contend that actual summary results were much higher in May, over 50% (doubtful). Is this good or bad? It depends on your perspective. If you are an advertiser it is not good, as some studies indicate that summaries (either search engine or chatbot) result in almost 35% less click-throughs than standard search results and many result in ‘zero-click searches’. For those that spend millions trying to improve their click-through results, this is the kiss of death, and almost all chatbot answers are either direct or summaries that don’t point to specific sites or links.
Of course, it is not as cut and dried as that might seem as studies have begun to show that a significant proportion of summaries are not factually accurate, reducing their credibility and usefulness, but unless that begins to affect user’s desire for summery results over listed results, its going to have an impact and it is doubtful that its going to be a good one for the SEO space or internet advertisers. Perhaps it will take the form of advertising discounting, or a consolidation in a very fragmented industry, but it would seem change is afoot. Will SEO go away, or lessen in its ability to affect advertising revenue for internet sites, or will GSEO (Generative Search Engine Optimization) take over, and if so, how will it work?
Since the monetization aspect of SEO could be impaired, businesses have to find new ways to offset a lack of click-throughs. Ideally the answer would be to be the ‘source’ for all AI chatbot or search engine summaries, or as many as possible, This would mean finding ways to get your content cited or used directly in summaries. Google has always emphasized EEAT (Experience, Expertise, Authoritativeness, Trustworthiness) as what drives search results, but we know that SEO is just another way of trying to game the search index system that Google is trying to protect There are certain ways in which content can be presented that make it easier for AI systems to parse and understand, and citations and backlinks are always helpful, but are we on the verge of a new industry, one that optimizes content for summaries?
Will the ads read “…our teams of editors will add the nuance that will boost your content into the stratosphere! Keywords, links, and everything else will optimize your content and help you squeeze out the summary competition!” But the real changes will likely come from the ability for content providers to provide 3rd party services with those services generating shared income between the content provider and the service provider. For example, the query “What LED lighting is best for Orchids?” yields a summary indicating that broad spectrum red & blue) LED bulbs are best for orchids, with a tag line asking whether you are interested in purchasing said bulbs or a fixture, and direct links to sites that have been ‘aligned’ with the content producer.
It’s a similar ecosystem to what exists today, but far more nuanced to specific products that are relevant to specific content, and that is going to spawn an industry of deal brokers, those who can connect a high quality content producer with top level summary exposure, to companies that provide the tools, talent, materials, or anything else that the content might reference. Sort of an intellectual influencer in a very small niche. It’s a complex structure but already seems to be one that is evolving from the search optimization world sites have been living in for the last decade. The times they are A-Changin’!
There are roughly 13.7 billion internet searches done every day or 1.664 searches for every person on our planet! Besides spawning companies Google (GOOG), Yandex (pvt), Baidu (BIDU), and Yahoo (pvt), internet search has created an ~$79b industry that looks to help companies improve their search results (SEO). It’s what we have become used to over the last 27 years since Page and Brin incorporated in 1998. During that period search engines have come and gone (Alta Vista, InfoSeek, Ask Jeeves, etc.) but the concept continued to evolve and get refined, even if Google continues to dominate.
Chatbots, like ChatGPT (OpenAI), Claude (Anthropic), and Perplexity (pvt), came on the scene at the end of 2022, but made no impact on internet search as they were trained and relied entirely on static data for their ability to answer queries. If you asked a question about something that happened after the model’s training end date, you got an answer like “As an AI language model, my knowledge is based on the data I was trained on, which has a cutoff date of [Month, Year]. Therefore, I do not have information about events or developments that occurred after that time”.
Things changed in 2023, with the release of Bard/Gemini, which included native internet search capabilities (Google Index), and ChatGPT’s Search (Bing Index), and while no tea was dumped in Boston harbor and there was no storming of any Bastille-like structures, a revolution began. First, it was summaries, where AI systems would put together the most logical answer to a query in the form of a one or two paragraph summary, and then came query answers directly from chatbots that were no longer limited to only using their training data as information sources, but without attribution. That did not last long, as it only took a few instances where AI chatbot answers were obviously wrong to push Ai designers to add sources and references to results in order to regain credibility. As of today traditional search engines still account for ~97% of all internet searches, with Google alone processing between 13.7 and 14 billion searches each day, but AI chatbot searches, as of April, have grown almost 81% y/y, albeit from a small base, but that is still a lot of growth.
There is a problem. When search engines are used for queries, the results are displayed in order of how the search engine ranks sites in its index, a science that many believe they understand, although only Google knows for sure. The entire industry functions around SEO (Search Engine Optimization), with billions spent to try to game the system enough to move one up in search results which can mean millions in advertising revenue. But with Google search now using query summaries, the quick answer for many queries is in that 1 or 2 paragraph summary and the ‘ranked’ search results only appear below the summary. In many cases (we don’t quite believe the numbers yet) the search goes no further and no site is selected..
In January of this year 6.49% of Google queries resulted in AI overviews, which increased to 7.64% in February and 13.14% in March. It peaked at 20.22% in April and sagged slightly in May (19.88%), although some contend that actual summary results were much higher in May, over 50% (doubtful). Is this good or bad? It depends on your perspective. If you are an advertiser it is not good, as some studies indicate that summaries (either search engine or chatbot) result in almost 35% less click-throughs than standard search results and many result in ‘zero-click searches’. For those that spend millions trying to improve their click-through results, this is the kiss of death, and almost all chatbot answers are either direct or summaries that don’t point to specific sites or links.
Of course, it is not as cut and dried as that might seem as studies have begun to show that a significant proportion of summaries are not factually accurate, reducing their credibility and usefulness, but unless that begins to affect user’s desire for summery results over listed results, its going to have an impact and it is doubtful that its going to be a good one for the SEO space or internet advertisers. Perhaps it will take the form of advertising discounting, or a consolidation in a very fragmented industry, but it would seem change is afoot. Will SEO go away, or lessen in its ability to affect advertising revenue for internet sites, or will GSEO (Generative Search Engine Optimization) take over, and if so, how will it work?
Since the monetization aspect of SEO could be impaired, businesses have to find new ways to offset a lack of click-throughs. Ideally the answer would be to be the ‘source’ for all AI chatbot or search engine summaries, or as many as possible, This would mean finding ways to get your content cited or used directly in summaries. Google has always emphasized EEAT (Experience, Expertise, Authoritativeness, Trustworthiness) as what drives search results, but we know that SEO is just another way of trying to game the search index system that Google is trying to protect There are certain ways in which content can be presented that make it easier for AI systems to parse and understand, and citations and backlinks are always helpful, but are we on the verge of a new industry, one that optimizes content for summaries?
Will the ads read “…our teams of editors will add the nuance that will boost your content into the stratosphere! Keywords, links, and everything else will optimize your content and help you squeeze out the summary competition!” But the real changes will likely come from the ability for content providers to provide 3rd party services with those services generating shared income between the content provider and the service provider. For example, the query “What LED lighting is best for Orchids?” yields a summary indicating that broad spectrum red & blue) LED bulbs are best for orchids, with a tag line asking whether you are interested in purchasing said bulbs or a fixture, and direct links to sites that have been ‘aligned’ with the content producer.
It’s a similar ecosystem to what exists today, but far more nuanced to specific products that are relevant to specific content, and that is going to spawn an industry of deal brokers, those who can connect a high quality content producer with top level summary exposure, to companies that provide the tools, talent, materials, or anything else that the content might reference. Sort of an intellectual influencer in a very small niche. It’s a complex structure but already seems to be one that is evolving from the search optimization world sites have been living in for the last decade. The times they are A-Changin’!
RSS Feed