AI Search is Driving Significant Signups, But Its True Impact is Largely Invisible

More than one in five individuals who sign up for Ahrefs report discovering the platform through AI chatbots like ChatGPT or Claude. This significant influx positions AI as a more influential source of new users than YouTube, second only to Google in its referral power. However, the true extent of AI’s impact on customer acquisition is considerably underestimated when relying solely on traditional referral traffic metrics. A substantial portion of AI-driven discovery occurs "out of sight," where users might encounter a brand via an AI assistant, subsequently search for it on Google, and then appear in analytics as organic search traffic. Therefore, to effectively advocate for AI search strategies within an organization, its influence must be identified through various data points beyond direct attribution.

The core challenge lies in the fact that a significant portion of the value generated by AI visibility remains untracked by conventional revenue metrics. This guide outlines seven key signals to measure AI search visibility and details where to find the necessary data. While a couple of these signals capture the directly attributable "visible" aspects like clicks and referral traffic, the remaining signals provide crucial evidence to support AI’s broader influence.

1. Tracking AI Referral Traffic

Sessions originating from AI assistants are treated similarly to other referral traffic. For businesses utilizing Ahrefs Web Analytics, a user-friendly alternative to Google Analytics 4, this data is readily accessible. Instead of complex manual configurations, users can simply navigate to the "Traffic sources" section and select "AI search" to filter traffic by this channel. This allows for an immediate understanding of which AI platforms are driving the most visitors. For instance, on Wordcount.com, an Ahrefs-owned website, ChatGPT accounts for a dominant 92.4% of AI search visitors, followed by Perplexity (4.6%), Gemini (2.7%), and Copilot (0.2%). This breakdown clearly indicates ChatGPT’s overwhelming contribution to attributable AI traffic.

Further analysis within the "Pages" report reveals which specific content and formats are attracting the most AI traffic and their associated engagement levels. On Wordcount.com, tool pages are the primary beneficiaries of AI traffic, demonstrating consistent user engagement. This granular data is invaluable for prioritizing AI platforms and commissioning content in formats that resonate with AI-driven discovery. It also provides essential top-line traffic figures for reporting, such as total AI visitors, AI’s share of overall traffic, and quarterly trends.

2. Tracking AI Referral Conversions

AI referrals represent a small fraction of the overall attention generated within AI platforms, and attributable AI-influenced conversions represent an even smaller subset of this attention. Despite their diminutive size, these conversions offer the clearest link between AI visibility and revenue, forming the foundation for any AI search return on investment (ROI) estimation.

Ahrefs Web Analytics’ "Tracked events" report can be used to imperfectly attribute conversions to AI. By applying the "AI search" channel filter, businesses can monitor conversion rates for events such as form fills and demo signups. For example, Ahrefs Free account signups from AI search visitors stand at 6%, a rate four times higher than that of organic search. Multiplying these conversions by the average customer value provides a tracked revenue figure that can be integrated into ROI calculations. It’s important to note that for free offerings like Ahrefs Free, the value is derived from the conversion of free signups to paying customers. This conversion data is crucial for justifying further investment in AI search, enabling businesses to strategically optimize high-converting pages for enhanced AI visibility.

3. Direct Inquiry: "How Did You Hear About Us?"

Instead of relying on complex data analysis and inferences, directly asking customers how they discovered the brand offers the most accurate attribution data. While self-reported attribution is not without its imperfections, it provides the closest approximation to the truth. The increasing use of AI in analyzing qualitative feedback at scale makes it easier to identify quantifiable patterns from open-ended customer responses.

Tools like Gong can facilitate direct questioning about AI discovery during sales calls. Alternatively, incorporating AI tool options into the "How did you hear about us?" dropdown during onboarding can capture this information. This direct approach supplements analytics data that might otherwise miss AI’s influence, allowing for the tally of attributable revenue. For instance, Ahrefs now knows that Claude and ChatGPT collectively account for 21.7% of its signups across all tools, surpassing YouTube and trailing only Google. Furthermore, prompting customers to share the specific AI prompts they used to find the brand during lead forms or surveys can provide valuable insights into which products, categories, and areas to invest in for optimal returns.

4. Monitoring Direct Traffic for Dark Funnel Insights

A consistent portion of AI-driven visits will inevitably be classified as direct traffic, rendering them unidentifiable through standard tracking methods. Many AI platforms strip referrer data, and users often search for a brand directly after receiving an AI-generated answer rather than clicking through from the AI. This behavior results in direct traffic in analytics reports.

It is advisable to record direct and no-referrer traffic volumes before initiating any AI-focused efforts and then monitor these metrics after each AI content push. A sustained increase in direct traffic following AI content publication, without other identifiable causes like campaigns, PR, or seasonality, can be reasonably attributed to AI. The value of this traffic bump can be estimated by applying the existing direct traffic conversion rate. For example, if direct traffic typically converts at 2% and an unexplained increase of 1,000 visits is observed, approximately 20 additional conversions can be estimated from AI.

5. Measuring Brand Demand Generated by AI Search

A rise in branded search demand, occurring concurrently with or following an increase in AI visibility, serves as a directional indicator of AI’s ability to generate interest. This metric provides a rough estimate of the extent of this generated interest, supplementing the limited scope of trackable clicks and conversions.

Brand Radar’s "Search demand" tab allows for tracking brand name search volume growth. It is crucial to utilize the "Brand name only" toggle to exclude non-branded category queries. When reporting on AI search, it is beneficial to chart AI mentions and branded search volume data side-by-side. A dashboard integrating Letaido and Ahrefs Brand Radar data can facilitate this analysis. While Ahrefs has witnessed exponential growth in AI mentions for its brand name, branded search volume has not yet fully mirrored this increase. Continuous tracking of this correlation over time will reveal the relationship between growth in branded AI visibility and branded search demand.

6. Reporting Web Mentions as an Early Indicator of AI Visibility Growth

AI platforms offer limited data transparency, lacking a direct equivalent to Google Search Console. The most effective method for gauging visibility involves tracking a sample of prompts and observing which entities are recommended. Consequently, monitoring web mentions alongside prompt data is highly valuable. Web mentions are easier to measure and tend to precede increases in AI visibility.

AI models derive their brand-related information primarily from the broader web’s discourse about a brand. A study indicated that branded web mentions are a significant factor closely tied to AI brand visibility. A higher volume of third-party pages mentioning a brand, particularly in a favorable light, increases the likelihood of AI recommending that brand and providing accurate descriptions. Ahrefs observed this firsthand: as the number of webpages mentioning them in the context of AEO tools grew, so did their visibility in AI answers for that category. Web mentions serve as a useful leading indicator for reporting, signaling progress in PR, content, and outreach efforts before more concrete numbers emerge. This can be presented as a leading indicator, such as "mentions are up 30%, which tend to precede AI visibility gains."

7. Analyzing Traffic Share by Domain

Reporting on AI search extends beyond highlighting gains; it also involves contextualizing perceived losses. If a company’s traffic declines by 10% while competitors in the same category experience a 25% drop, this indicates a net gain in market share for that quarter. Raw traffic numbers alone cannot reveal this nuanced performance. Therefore, when addressing inquiries about traffic decline, "traffic share by domain" becomes a critical metric. Available in Keywords Explorer, this metric situates a site’s performance within the broader market context.

Final Thoughts

The definitive quantification of AI search’s value remains an ongoing challenge. AI-driven recommendations influence buyer decisions long before they result in trackable clicks, meaning no single metric can provide a complete picture. Businesses should select the signals that best align with their objectives, report them consistently alongside ROI estimates, and maintain transparency about the limitations of each metric. The current landscape presents an opportune moment to build a compelling case for AI search strategies.

The emergence of AI-powered shopping is shortening the path from recommendation to purchase. Publishers are exploring innovative models like pay-per-crawl, and Google is piloting "AI Contribution," a Search Console program that compensates publishers whose content significantly contributes to AI Overviews, AI Mode, and Gemini answers. This signifies that AI content not only creates demand and influences sales but also possesses the potential to generate revenue. The time to measure this value is now, without waiting for perfect attribution.