What it shows is 13 months of data about LLM traffic, growth, and conversions

LLMs and their impact on brand website traffic are a big topic in our clients’ conversations. Everyone wants to know what’s going on, how they can do better, and what the best practices are.

My advice to brands right now is to start with data and focus on what they know for sure.

To gain insight into how LLM traffic influences key metrics, we analyzed our LLM referral traffic dataset in Google Analytics for all of our clients over the past 13 months (Jan. 1, 2025 to Feb. 7, 2026).

We focus on traffic from various LLM models to product sites and conversion events that are very close to real business results. In some cases, that’s buying. For others, it’s a generated lead.

Looking at this dataset, four major findings stand out:

• LLM referral traffic is still small.
• LLM traffic is growing rapidly.
• Sources referenced in answers vary.
• LLMs convert at a much higher rate compared to other channels

LLM referral traffic is still small

LLM referral traffic accounts for less than 2% of total referral traffic on average, according to our data set. In other words, less than 2 out of every 100 visitors to the site come from an LLM referral source.

The range is 0.15%-1.5% of referral traffic from various LLMs, including ChatGPT, Perplexity, Gemini, and Claude.

So while this is a big topic of discussion, it is not the most important in the near term impact on most businesses.

LLM traffic is growing rapidly

LLMs, as a referral source, are growing rapidly, according to our data. Comparing the first half of 2025 with the second half, we saw an average growth rate of 80% in LLM referral traffic.

There was a wide range across the dataset. Some companies saw only 10% growth, while others saw a 300% increase.

Below is the total referral traffic per month in 2025. It shows steady month-over-month growth, peaking at a 3-fold increase in traffic from January to December.

That means it’s not enough to understand your LLM traffic volume. You also need to monitor the speed of that growth.

LLMs are proliferating as consumer adoption grows, and agile algorithms keep changing. Between these two variables, you can see some dramatic changes that you need to watch out for.

Dive deep: LLM development in 2026: Tracking, visibility, and what’s next for AI discovery

Sources referenced in answers vary

Sources cited in LLM answers change rapidly.

Here’s a look at our dataset from September of last year. The data comes from monitoring over 5,000 prompts and their responses across various LLM APIs, including Gemini, ChatGPT, and Perplexity.

YouTube links and citations have increased over the past 30 days. Reddit has seen similar growth, although that traffic has recently slowed.

These changes in citations and links will affect the traffic that ultimately reaches your site, and may also affect your ad strategies and content.

If you don’t monitor this data, you won’t see these changes. LLMs do not provide this information directly — you can only access it through a third-party tool.

LLMs convert at a much higher rate compared to other channels

This is probably the most interesting and important discovery. When you compare conversion rates in terms of percentage of total traffic, the difference is clear.

LLM referrals are the highest converting traffic source for all our clients, with an average conversion rate of 18%. That’s higher than any other strategy, including paid shopping, SEO, and PPC.

However, they make up a low percentage of total traffic to a brand’s website, about 25 times less than SEO or directly.

Dig deeper: How to best measure LLM visibility and impact

What types should be made next

Based on these findings, you should take the following steps to prepare for a dynamic LLM situation.

1. Establish dedicated monitoring

Although the volume of LLM traffic is still low, its growth rate and volatility, including switching between sources like YouTube and Reddit, make monitoring very important.

• Track speed: Don’t just look at the volume. Monitor the growth rate of LLM referrals to understand if this channel is crossing a critical threshold for your business.
• Note the sources of the quote: Use third-party tools to understand which LLMs and which types of platforms, including forums, videos, and stories, are driving the most citations and follow-up traffic.

2. Make money from high value traffic

An 18% conversion rate suggests that users referred to LLM are highly qualified. They usually come with a clear objective or after their question has been answered or approved by the LLM.

• Analyze the most dynamic journey: Review user journeys for LLM referrals. What content are they coming to? What questions are answered that lead to conversion?
• Set a goal: Focus on content and landing page development on the high-end needs shown in the context of the LLM quote. Treat this traffic as a premium audience.

3. Plan for future growth

Given the rapid acquisition of LLM, today’s low traffic volume will not last.

• Develop an AI content strategy: Develop a strategy that anticipates how LLMs summarize, cite, and reference your material. This is not traditional SEO. It’s about being an authoritative source that LLMs choose to connect to.
• Allocate a budget: Although this may not drive immediate impact, dedicate a small budget to tools and resources focused on understanding and improving the LLM referral channel.

This space is developing rapidly. Hopefully, this dataset shows how things are going and inspires action within your organization.

This is a time of change. If you innovate, stay focused, and use data, you have a clear chance of outperforming your competition.

Focus: LLM consistency and recommendations: The new SEO KPI

From an emerging channel to a strategic signal

LLM referral traffic is still a small share of the overall volume, but it is growing rapidly, changing where it cites, and driving strong conversions.

Don’t be rash. Monitor trend lines, understand where quotes are coming from, and watch how these audiences behave once they arrive. This space is moving fast, and the closer you are to the data, the better you’ll be as it changes.