SEO For Search GPT
How To Optimise For Better Visibility
Quick summary:
Predictions On SearchGPTs Ranking Factors
Here is exactly what I believe will be the core ranking factors for SearchGPT. It’s actually very different from Google’s traditional SEO rankings factors based on the advancements and complexity of AI.
Similarity between ChatGPT and SearchGPT ranking factors
Currently, we have ChatGPT’s algorithm that we can study to see how SearchGPT will likely work. Based on the fact this will be heavily integrated into ChatGPT, I’m going to assume they are going to be closely related.
For a quick summary, I believe the core ranking factors are going to be essentially the same. This includes user intent and relevance, ranking the best answer to the query, and providing a useful, credible website as reference.
Here’s a table comparing the similarities and differences between SearchGPT’s ranking factors and ChatGPTs ranking factors for SEO.
Ranking Factor | ChatGPT | SearchGPT |
User Intent & Relevance | Summarises the best response based on its historical learning, without seeking new information. | Crawls websites to find the most specific, niche response to the query, prioritising fresh and relevant content. |
Content Quality | Generates high-quality, contextually accurate responses using a vast database of trained knowledge. | Ranks websites with high-quality, detailed content that aligns with user queries. Will train its answer from quality websites. |
Credibility | Provides information from credible and authoritative sources, ensuring accuracy. | Prioritises credible websites with authoritative content to rank higher in search results. This could be topical authorities. |
User Interaction & Feedback | Continuously learns and adapts based on user feedback to improve response accuracy. | Likely to consider user engagement metrics (clicks, dwell time) to refine search rankings and featured websites. |
Reference to External Content | References general knowledge and databases without linking to specific web pages. | Directly references and ranks external websites, displaying them in search results. |
Integration with AI Tools | Integrated deeply with AI-driven chat systems for interactive responses. | Integrated into ChatGPT to begin with but likely to become its own entire search engine. |
Natural Language Understanding | Excels at interpreting and responding to complex, nuanced language, providing contextually relevant answers. | Uses advanced natural language processing to understand and rank content that precisely matches user queries. |
Contextual Awareness | Generates responses by understanding the context of the conversation or query. | Ranks content that fits the broader context of a user’s search, considering related queries and topics. |
Trustworthiness Signals | Relies on data from authoritative sources and established knowledge bases to ensure trustworthiness. | Likely to prioritise websites that exhibit trust signals such as HTTPS, clear author information, and privacy policies. |
Semantic Search Capabilities | Capable of understanding synonyms and related concepts to provide accurate responses. | Uses semantic search to match queries with the most semantically relevant content, even if exact keywords differ. |
Adaptability to New Information | Updates its model periodically to improve accuracy based on new data and user interactions. | Crawls and indexes new content continuously, adapting to changes in web content to stay current with the most relevant information. |
Heavy focus on reliable brands with omnipresence channels
Based on what I’ve seen, ChatGPT seems to rank brands with a large omni channel presence. I believe SearchGPT will adopt this as well, however for many SEO experts, this proves extremely frustrating and offers the same problems happening with Google at the moment.
Brands that have good engagements across multiple channels like social media, youtube, Google will likely get featured more than others. Also, I believe those with a large search volume on their brand name will see a positive impact as well. When users are searching for your brand, this is a huge trust signal that your brand is trusted by many.
Less reliance on backlinks, instead detail of content
I also believe there is going to be less reliance on backlinks with SearchGPT. This is due to the nature of the AI’s extremely advanced natural language processing. It seems to have a much higher understanding of search intent and relevance.
Therefore, it can find a better response by crawling and understanding the context of websites content to match with the user query. It won’t need backlinks to confirm this content is helpful, instead it can use its advanced natural language to know that the content matches the searchers query indefinitely.
This is certainly a game changer, and going to allow the rise of helpful content so much more than Google ever can.
Data driven search preferences for more personalisation
One thing I believe we might see as well, is further personalisation. Right now, ChatGPT is collecting so much data on us as a user who are searching for different types of things in very specific ways, over and over.
They can take this data, and personalise it in a way based on our history of search, responses, clicks to external sources and ends of chats. If we find what we want, they know their answers and references are helpful and therefore keep showing these.
SearchGPT can then tailor its responses, results and references based on your historical data to provide more accurate, more personalised results. Although this might be a bit of time away from happening, I am sure it will be the future.
Schema markup rise in importance for AI
Now, the way SearchGPT will work is through featuring websites in snippets. By doing this, they will need Schema to read from websites to allow enhanced features.
For example, websites will need to add schema in for their titles, images and metadata so that SearchGPT can quickly grab this information and turn it into a featured reference.
I believe that this schema markup might make a difference in those that will get featured and those that will not. Although it will be a small ranking factor, I still believe that it will make a big difference.
How To Optimise Your Content For Search GPT
So, of course you want to get your website to get featured in SearchGPT and ChatGPT. From what I’ve seen and experienced, these are going to be the most impactful ways to get quick results.
I’ve actually highlighted what I believe to be the most impactful, short term action you can take to get fast results.
The importance of quality content
The entire system is built on advanced natural language processing, therefore content is going to be the biggest ranking factor overall. We’ve seen Google start to move toward this style of ranking as well making its search results tailored more towards user intent.
Content that is genuinely helpful is going to be featured much more. What I mean by genuinely helpful are things like statistics, studies, breakdowns, personal opinions, genuine reviews, advice or things to avoid in an industry, and giving away trade secrets.
My opinion on types of helpful content:
- Trade secrets
- Product comparisons
- Free SaaS tools
- Expert industry opinions
- Stats / studies with real data
Shift towards niche, more technical content (Most Important)
This is a big one, and I believe it will be the most impactful for a quick win and quick rank on SearchGPT. Now, the AI will basically be crawling its index of websites for the best suited answer.
Therefore, the more niche, more specific and more technical you can get your answer the higher the chance you will get featured. AI is going to look to summarise your content into an easy, more concise response.
So, it’s to respond with a technical answer that has been made simple by AI. You can see this happening now with ChatGPT. It will take a technical answer, and simplify it down. The better you can answer specific queries, the more likely you are to get featured.
Go into specific details
Go into as much detail on a topic as you can, to showcase your industry knowledge. This is going to be favoured over content that just brushes over a few things.
Here are two examples of content, one goes into more specific detail vs the other one.
Brushing over the topic:
Lighting a candle is simple: just use a match or lighter to ignite the wick. Once lit, the candle will burn and create a warm, ambient light. Make sure to keep the flame away from anything flammable and never leave the candle unattended.
Going into detail around the topic:
Lighting a candle is simple, but also it’s important to do it correctly to ensure the candle burns evenly and safely. First, trim the wick to about 1/4 inch before lighting. This helps prevent excessive soot and ensures a cleaner burn. Hold the match or lighter close to the base of the wick, allowing it to catch the flame gradually. Avoid holding the flame directly on the wax, as this can cause tunnelling—a situation where the wax around the edges remains unmelted, leading to an uneven burn.
Content that trains AI, but is also helpful for users
Now, this can touch slightly on what I’ve mentioned above. Creating content that AI can train itself on, and quickly summarise your answer. I do believe that this will become a trend in the future.
Content will purposely be created to help train AI on information around topics, especially up to date topics. This again means providing extremely in depth answers that are semantic in structure so AI can quickly train this information and turn them into vectors.
Using entity based SEO e.g. People, places, things
Entity-based SEO is becoming increasingly important as AI models like SearchGPT evolve. Google has already stated the importance of entities and their rise in their own algorithms.
So what are they? Entities are unique and unambiguous concepts that SearchGPT can identify and understand, allowing it to grasp the context and relevance of your content better.
For example, instead of just targeting keywords like “famous photographers,” you’d optimise for specific entities like “Annie Leibovitz”. This will likely have a better chance at featuring or referencing your content across many different SearchGPT queries.
How To Allow Search GPT To Crawl Your Website
It’s important as we start to see a larger shift towards SearchGPT that you allow OpenAI’s crawl bots to crawl your website and feature you in ChatGPT and SearchGPT’s responses. Here’s a quick breakdown of how they work, and how you can allow them to crawl your website.
How will SearchGPT crawlers work?
SearchGPT crawlers are designed to explore the web much like traditional search engine crawlers, but with enhanced capabilities tailored to the AI’s specific needs. These crawlers will systematically scan websites, indexing content to build a vast database of information that SearchGPT can pull from when responding to user queries.
SearchGPT crawlers will not only index text based on keywords but will also use advanced natural language processing (NLP) to understand the context and meaning behind the content.
This allows the AI to better interpret complex topics and provide more precise answers to user queries. For example, the crawler might distinguish between different meanings of a word based on the surrounding text, ensuring that the indexed content is understood correctly.
Enabling crawling of your website by
Just add this code to your robot.txt to allow ChatGPT and SearchGPT crawlers to access and crawl your website.
User-agent: OAI-SearchBot
Allow: /
User-agent: GPTBot
Allow: /
User-agent: *
Disallow:
Types Of Websites To Benefit From SearchGPT
As with the rise of anything, some websites will see benefits whilst others suffer. Now, like with many Google algorithm updates, the websites that benefit the most are the genuinely helpful ones.
Here’s my list with explanations of what websites I believe are going to really see a huge benefit from SearchGPT and should start to focus on getting some optimisation done as soon as they can.
- Technical, industry-specific sites: Those that specialise in specific, highly-focused topics, providing detailed and expert-level information that aligns with user queries.
- Educational platforms: Websites that offer comprehensive tutorials, guides, and learning resources that are valued for their accuracy and depth.
- Local business websites: Local websites with localised content, providing specific information about their service that is relevant to users in particular geographic areas.
- E-commerce sites: Product specific websites that provide extensive product descriptions, customer reviews, and buying guides, helping users make informed decisions.
- SaaS tools: Useful tools that users continue to use and keep going back too.
Types Of Websites To Suffer From SearchGPT
And on the other hand, we have those websites that will suffer when SearchGPT is released. Similar to when Google released their AI overviews, some websites saw a huge drop of traffic.
Here are the types of websites I believe are going to suffer the most when SearchGPT becomes more prevalent across the way we search and find information from the web.
- Thin content sites: Websites with shallow or poorly researched content that fails to provide in-depth insights.
- Overly broad directory sites: Business directories or listing sites that offer minimal value beyond basic contact information may see reduced traffic.
- Content farm websites: Sites that produce large volumes of low-quality, generic content designed purely for SEO & ads will likely see their visibility decline.
- Non-specific blogs: Blogs and websites trying to cover all topics and accepting guest posts will see their traffic taken by AI and ChatGPT.
Search GPT And Local Ranking Factors
Local search will still be very profitable for businesses such as cafes, local service based businesses, places and specialist trades. Surprisingly, I’ve found that ChatGPT actually quotes some fairly strong recommendations on local businesses.
When I asked ChatGPT on what ranking factors it used when telling me the “best roadworthy is Brisbane” these were the areas it said it focused on to provide it’s answer:
Identifying the key ranking factors
- Customer reviews and ratings
- Geographic coverage and reach
- Company size (number of locations, vehicles, employees)
- Professionalism and punctuality
- Industry reputation and longevity
- Service offerings and specialisations
Collecting the data for those ranking factors
Next, data is collected for each factor from available sources:
- Customer reviews and ratings: Aggregated from Google Reviews, Facebook, Yelp, etc.
- Geographic coverage: Information gathered from company websites, Google My Business profiles, and customer reviews.
- Company size: Assessed through the number of branches, employees, and fleet size, often gathered from company websites, business directories, or industry reports.
- Professionalism and punctuality: Inferred from customer reviews mentioning timeliness, behaviour, and thoroughness of the service.
- Industry reputation and longevity: Checked through business listings, years in operation (often found on the company’s “About Us” page), and any industry awards or recognitions.
- Service offerings: Based on the range of services provided, as listed on company websites.
Creating a weight to those factors
Each factor is assigned a weight based on its importance to the overall ranking. For example (Note* These are hypothetical figures, made up by ChatGPT)
- Customer reviews and ratings: 30%
- Geographic coverage and reach: 20%
- Company size: 15%
- Professionalism and punctuality: 15%
- Industry reputation and longevity: 10%
- Service offerings and specialisations: 10%
Calculate the overall score
Then assuming AI will rank the overall score and feature the brand / local business in the answer it gives. The businesses with the highest overall score will be featured first, then in numerical descending order from there.
Short summary of this
So, I’ve tested this and I would likely say that most of this is fairly accurate. All of the local businesses that ChatGPT recommended to me, had super strong reviews, had the largest coverage, were big companies with a lot of employees and had a fantastic reputation with large search volume on their brand names.
I would likely say this is a very close representation of the algorithm SearchGPT will use to reference and feature local businesses as well.
Will Websites Lose Clicks To Search GPT?
The question that we are all concerned about. Is there a straightforward answer… Likely not, because SearchGPT hasn’t been released yet we don’t have the data to tell.
However, I do have a hunch that yes, likely there will be less clicks on SearchGPT compared to Google because of the nature of AI summarising these websites and giving the answer directly to the user based on their query.
Now, I also think that certain terms and user journeys will lose more clicks than others. For example, if you ask a definite question, like “what is the weather in London” then your search journey is short, as you will get your answer straight away.
Other questions like “what is the best digital marketing strategy in 2025” might lead to more clicks through to see further details for your query.
Humanising your content might get better CTRs
Something to consider is trying to humanise your content. Using images of yourself, videos of yourself and adding in personal opinions could help a fair amount. This might lead users to click through to get a more human answer and read your content.
Fallbacks of SearchGPT
Like anything, of course there are some fallbacks of SearchGPT in my opinion. Although I do believe that it’s going to be more useful than the current search engines we have, I feel like it might have some teething issues to begin with.
Brand new, untested algorithm
It’s a brand new algorithm, which is likely going to have some ways it can easily be manipulated to get better features or rankings, although it seems somewhat harder than Google right now.
Still likely going to base itself from Google
Despite its advancements, SearchGPT will likely still draw from and align itself with some aspects of Google’s existing search engine methodologies. This means that some of the same limitations and biases inherent in Google’s approach could carry over into SearchGPT.
This is because SearchGPT isn’t getting any new information or data, it’s getting the same website and data that is already on Google.
Although, I do feel like it will break a lot of low quality content away from the surface of the web, but also push more traffic into the giants without giving small businesses a fair shot in the game.
It will take a very long time for users to change over
For a lot of users, Google is their bread and butter. It’s a search journey they have followed for years. Even if SearchGPT proves to be superior, getting users to switch from familiar search engines like Google will be a slow process.
My Final Thoughts…
I can see SearchGPT becoming a main player in the game, but it’s going to take some time. Overall when it comes to SEO and SearchGPT we can learn a lot of lessons from its algorithm.
Relevance, quality and expertise is going to become much more important along with factors like Schema for featured results and the credibility of your website.
I really see those brands who are absolutely everywhere, with reviews in different places, content in all channels, and a true specialist presence being the ones that will have greatest success with their SEO on SearchGPT.
Leave your thoughts in the comments below. I’d love to hear what everyone thinks about SearchGPT and how you intend on optimising your site and brand to get featured in the results.
