Chat with Softimpact
5 Major Insights from Our AI Search Research
1. AI Search Traffic Could Overtake Traditional Search by 2028
Our research suggests that by early 2028, websites focused on digital marketing and SEO may receive more traffic from AI-powered search platforms than from traditional search engines.
This shift likely reflects a broader transformation that will impact industries across the board.
If Google transitions its default search experience to AI Mode, the change could happen even sooner.
We anticipate that engagement with AI-powered search will continue to rise as the technology becomes more accessible and integrated into everyday use, gradually decreasing users’ dependence on traditional search results.
For instance, weekly active users of ChatGPT increased eightfold between October 2023 and April 2025, surpassing 800 million users.
At the same time, Google has begun rolling out AI Mode, a conversational, AI-driven experience that replaces the traditional search results page.
Google has also expanded the presence of AI Overviews, which generate summaries that appear above standard search listings.
As user behavior evolves, a growing share of clicks will shift from traditional search engines to AI-powered search platforms, and some clicks may disappear entirely.
In the short term, this could lead to a decline in overall combined traffic. Over time, however, traffic is likely to stabilize and gradually increase.
This shift happens because AI search:
Compresses the marketing funnel By delivering comprehensive answers upfront, AI platforms reduce the need for users to visit multiple websites to gather information.
Deprioritizes links AI-generated responses do not always include direct links to cited brands or sources. When links are included, they are often less prominent than those in traditional search results.
AI-driven search traffic could surpass traditional organic search within the next two to four years. Businesses that delay optimizing for large language models (LLMs) risk losing visibility to competitors who establish authority early in AI-generated results.
Although LLM optimization shares many fundamentals with traditional SEO, the two approaches are not identical.
The first step is to assess how visible your brand currently is within AI-generated responses.
This can be done using AI visibility and optimization tools that analyze brand mentions, prompts across the buyer journey, and competitive presence within large language models.
For example, an e-commerce brand selling athletic socks might monitor prompts such as “men’s running sock brands” or “best socks for yoga” to evaluate how frequently and prominently it appears in AI-generated answers.
2. The Average LLM Visitor Delivers 4.4x More Value Than Traditional Organic Search Visitors
Our research shows that the typical AI search visitor coming from non-Google sources like ChatGPT generates 4.4 times more value per visit than a user from traditional organic search, based on conversion rates.
As AI search adoption increases and reliance on traditional search gradually declines, we expect AI-driven channels to generate comparable economic value worldwide by the end of 2027, with the potential to significantly surpass it in the following years.
AI search visitors typically convert at higher rates because large language models provide users with comprehensive information, enabling them to make informed decisions more quickly.
When someone visits your site via AI search, they’ve usually already done their homework—comparing options and even understanding what makes your brand unique. That makes them much more likely to take action.
AI responses also feel more personal, like a recommendation from a friend, which can make them more persuasive than standard search results.
Visitors from AI search are often more qualified than those from traditional search. Even a small boost in AI traffic can have a big impact on your business.
The good news? A lot of the SEO practices you already use, like creating helpful content, making sure your pages are easy to navigate, and building brand mentions, also help you show up in AI search.
To get even better results, focus on:
If you want to track how your brand shows up in AI-generated content, there are tools that can monitor mentions and sentiment and give you clear guidance on how to improve your visibility.
You can also check how people feel about your brand in AI search results.
A simple way is to use tools that track over time so you can spot and handle negative mentions quickly.
3. ChatGPT Often References Lower-Ranked Pages
Our data shows that when ChatGPT Search points to webpages, nearly 90% of the time it’s linking to pages that would normally appear after the 20th spot in traditional search results.
Similarly, Perplexity and Google’s AI models also tend to reference pages that rank lower in standard search listings.
Some LLMs Still Favor Top-Ranked Pages
Certain AI models do cite pages that rank in the top 1–5 positions more often than pages ranked 6–20. In other words, doing well in traditional search can still improve your chances of being cited by AI.
But you don’t have to be a top-ranking page to get cited. Our data shows that AI can reference lower-ranked content, and we have three main reasons why:
1. More options to choose from – AI systems can draw from the vast pool of pages beyond the top 20.
2. Content-focused experiences – Unlike traditional search, AI search focuses on the most relevant pieces of content (“chunks”), not overall page metrics like load speed or length.
3. Smarter understanding of user intent – AI can:
o Process natural language more effectively
o Understand context through conversation
o Build a nuanced view of each user over time. This lets AI deliver the best answer for a specific user, rather than just the page that performs best on average.
What This Means for Your Strategy
Traditional SEO still matters. Ranking in search can help your content get noticed by AI systems. However, being in the top three organic positions isn’t as critical for AI search.
AI is more likely to cite content tailored to specific audiences or use cases. So, your strategy should include:
4. Quora Leads in Google AI Overviews
According to our research, Quora is the website most frequently cited in Google AI Overviews, with Reddit coming in second.
These platforms are rich in niche Q&A content, making them valuable sources for AI models. Reddit may also benefit from Google partnerships, which help its content feed AI systems.
These are high-authority domains that consistently rank well in traditional organic search, which likely contributes to their frequent citations by AI systems.
What This Means for You
Marketing on highly cited, high-authority websites can play an important role in your AI optimization strategy, as these sites are often referenced by AI systems.
Getting featured on other relevant, authoritative websites through digital PR or link-building strategies can also improve your visibility in AI-driven search.
You can identify the most impactful places to get mentioned by analyzing which pages are most frequently cited for your topics, and track whether your brand is mentioned positively, neutrally, or negatively.
5. Half of ChatGPT Links Go to Business or Service Websites
Our research shows that about 50% of links in ChatGPT responses point to business or service websites, even though some other domains may perform better on an individual site basis.
We believe this trend is similar across other AI models.
This suggests that large language models often rely heavily on your website when generating responses about your business, and generally view business websites as strong sources of information on relevant topics.
Your website has real potential to get cited in AI responses, but you need to create content that is LLM-friendly.
Here are some quick tips:
Notes on AI Traffic Projections
Traffic and value projections are based on historical data and adoption trends. These models account for:
Preparing for an AI-Driven Future
Optimizing for AI search isn’t just about visibility today—it’s a long-term investment.
LLMs learn from patterns across vast online content. By shaping how your brand appears now, you increase the chances it will be presented accurately, favorably, and prominently in the future.
Our research indicates that AI-driven search could become a major traffic and revenue driver by 2027, if not sooner.