What AI based Search like Chat GPT and Google Bard will mean for Search Behaviour and SEO – 8 Potential Outcomes

In the last few months, it seems the world has suddenly woken up to the AI revolution that is upon us due the general public discovering something tangible and usable for AI on a mass scale, in Chat GPT

It seems the race is on to grab AI Chat/Search market share with Microsoft first out of the gate with their investment in Chat GPT and Google seemingly having to rush out their announcement of their new AI Chat function in Google Bard with very few details or visual previews.

In the coming months and years there will be much speculation, theories, predictions and outcomes to how an online user’s interaction with information search and retrieval will evolve and this could also have a significant impact on the subject of SEO, particularly how websites are able (or not able maybe) to optimise for the evolution (or maybe revolution) of AI Search.

With this in mind, it got me thinking and wondering how this next level introduction will impact search behaviour and what it could mean for the SEO industry.

So here are a few points that could be potential outcomes, as the search world get to grips with the potential of AI dominating, as the internet continues to evolve……. 

Potential Search Behaviour Evolution

AI Search Functions will initially be a Tab Option on Search Interfaces

As Bing has already shown in it’s recent release on plans to include a “Chat” option as a tab in its main menu from the Homepage.

I suspect Google will do something similar to start with.

Eventually perhaps, we will see the Chat tab being the default tab on the homepage and the main search results we see today, listed in an “All” tab in Google right now being a secondary tab. “Classic Search” maybe?  

For the foreseeable future though, perhaps search engines like Google and Bing will recognise that a smooth transition is required between search engines and AI Chat Assistants, with a need to use them interchangeably.

More Long-Tail and Question Search Done in the Chat Function

Search queries used in search engines have evolved significantly over the last 5 years. Aspects like voice search have been introduced and also people have become more confident in the quality of results received when asking search engines more complicated questions. Based on this evolving search behaviour, AI Chat Search will result in further query evolution.  

As a recent example provided by Google, this is how Google currently displays information when using AI to respond to long queries with a long AI-created response: 

If the likes of Google and Bing clearly differentiate between the search engine function and the Chat AI function with separate interfaces and sections on their websites, then one scenario that may occur is that less long-tail and question search will be conducted in the search engine section and more volume of queries will occur in the Chat AI function for such search query types.

Bing will gain more market share worldwide and erode Google’s market share

It looks as if Microsoft may have stolen a march on Google with their involvement in Chat GPT, and so many are predicting that this may allow them to finally gain more market share as part of a paradigm shift in user behaviour. 

For many years Bing has tried to use it’s assistant Cortana as a unique selling point to get more people to use it’s search engine but it seemed that culturally not enough people searched and sought out information in the style and format that Cortana provides, but now with more people now becoming aware of AI Assistants and the phenomenal growth Chat GPT has seen, the world may be coming round to being suited more to what Bing offers.

Data from the last 3 months (November 2022 to January 2023) shows that Google currently has more than a 92% share of the worldwide search engine market, so it will be interesting to see whether Bing will increase it’s percentage share in the coming years and months at the expense of Google and other search engines. 

Microsoft has already started to implement measures to try and gain more market share for Bing and Microsoft as an ecosystem as a whole, with the new waitlist it has put in place for people to access the “new Bing”, which will provide an “AI-powered answer engine” with the full Chat AI mode only being available when you access to their new upcoming version of Bing.

It appears that they are offering incentives to users to get faster access to the new Bing version (with it’s new AI chat capabilities) if users set Microsoft defaults on their “PC” and install the Bing App on their phone. This seems like a strategy to gain a more ingrained market share for the Microsoft ecosystem.

Research being done in AI Chat environments and then bridging to more commercial search results

With the evolution of AI Chat/Search, an interesting development will be to see how people’s use specific functions as their search journey develops, for example if someone is starting at a research phase in their search journey but then wishes to move to a more commercial focus, such as wanting to find a website to make a purchase, is this where the more standard search engine results will still be relevant? 

Or will the AI Chat section/function incorporate search results in a more seamless way when people use prompts such as: 

• “Suggest some websites that provide more about this”
• “Where can a buy these things we have been talking about”

In some research journeys, it may be a better experience to provide some specific websites and web pages to view rather than a list of recommendations you then have to copy and paste separately into a search engine, like the recent experience I had with Chat GPT (see below).

Google and Bing are well positioned to integrate current search engine result snippets in various formats to AI Chat functions, so it will be interesting to see how they do this from a UX/UI perspective. Comparing the current Chat GPT interface and results to the new Bing “answer engine” will give us an early indication on how this could work. 

The above potential outcomes focus more on how user behaviour could work and evolve with AI Chat and Search. 

The below potential outcomes focus more on what it could mean for SEO

Potential SEO Industry Implications

Informative, Factual Content with Relevant Keywords will Still Matter

Following the mantra set by the likes of Google to ensure your website provides useful content to users will be a key factor to be referenced by any AI assistant service, just as it currently is to gain higher organic search visibility in the current core search engine results.

Optimising for SEO industry and Google defined concepts such as YMYL (Your Money Your Life) and E-E-A-T (Experience, Expertise, Authoritativeness, and Trustworthiness) defined in section 3.4 of the Google Search Quality Evaluator Guidelines, will stand a website in good stead to be referenced in future AI based answer engines. What we perhaps don’t know yet is whether they will have certain “go-to” sources they reference based on an authority based grading system which search engines like Google currently have through various signals including external links to a website and how often a brand name is searched for.

Machine learning may still require content to be referenced with metadata being processed and keywords being identified where required. A useful progression required would be having updated keyword research tools that measure search / request / command volume from new AI chat assistant tools to help SEOs identify the best terminology and content structures to include in content so that website content can adapt to how users will adapt their search query structures. 

Specific Markup Code to Highlight Specific Information to AI Chat / Assistants

One optimisation factor we may see help to increase visibility / reference frequency in AI based answer services is the use of structured data markup via schema code (or another web coding standard in future).

Google currently references certain types of schema markup as structured data to help identify specific content themes and also provide information in certain result snippet structures 

Google has been testing the use of “speakable” schema markup structures in website code to help identify certain information that may be suitable for voice assistants like Google Home to reference. So will search engines look to use existing or new schema code structures to help identify information suitable for AI based answer engines perhaps?

Some SEOs will still claim to have the secret recipe that optimises for all AI search

No matter what kind of new update there is in the world of search and information retrieval, there will always be someone who claims to have a magic SEO potion that ticks all the AI signal boxes and there is bound to be articles that will appear online telling you “How to optimize your website for AI Search” providing the answer to the secret recipe of AI SEO. So just be careful if you see such articles because every website has it’s own SEO DNA and one size does not fit all. This will ring true when it comes to how AI technology interprets each individual website.

SEO will not be dead

Obituaries have been written on the demise of SEO, almost ever since it was recognised as a discipline, art, science, skill (whatever you want to call it). 

The need for SEO and people that competently deliver SEO consulting will not cease to exist, but as always, SEOs will need to adapt to the new frameworks devised by the companies providing information retrieval resource solutions such as Google, Bing and others in future. 

SEO will adapt to include optimisation requirements for AI based information search based tools and it will continue to become more about making your website or app as informative and user centric as possible.