The Problem: AI Bots Are Invisible to Standard Analytics
AI crawlers — GPTBot, ClaudeBot, PerplexityBot, and dozens of others — now account for a significant share of traffic to many websites. Yet most site owners have no reliable way to compare this AI-driven traffic against human traffic. Standard analytics tools were built for human visitors; they were never designed to answer a basic question: how does AI’s attention to a page compare with a human’s?
To address this gap, we built EdgeShaping, an AI bot visibility tool, and within it implemented a metric called the AHQG Matrix. This article introduces its extension, the AHTG Matrix, and explains why the distinction matters.
AHQG: Measuring the Gap Within Search
The AHQG Matrix (AI・Human・Query・Gap) plots two variables:
- X-axis: AI-driven search traffic (crawler and AI-search-engine visits with search-like intent)
- Y-axis: Human search traffic (search clicks, sourced from Google Search Console)
Each page is placed into one of four quadrants based on this comparison — for example, pages where both AI and humans engage heavily (“ALIGNED”), pages where only AI shows search-like interest, pages where only humans do, and pages neither engages with through search.
By restricting the analysis to search-intent traffic, AHQG captures something specific: the position of each page within the broader AI-era search journey, in which AI systems capture early-stage interest and — in many cases — that interest eventually converts into a human visit. AHQG is a lens on that particular pipeline.
Where AHQG Falls Short: Search Is Not the Whole Story
AHQG’s precision comes at a cost: it only sees search. But traffic doesn’t only arrive through search. Paid ads, social referrals, direct visits, backlinks — on many sites, especially commerce-driven ones, these non-search channels dominate.
A framework that only measures search-intent traffic will systematically miss the AI-human gap on any site where search isn’t the primary channel. This is the blind spot AHTG was built to address.
AHTG: Widening the Lens to All Traffic
The AHTG Matrix (AI・Human・Traffic・Gap) generalizes the same underlying logic to traffic as a whole, regardless of channel. The core changes are:
- Q (Query) becomes T (Traffic): the search-intent restriction is dropped. Every channel — search, ads, social, referral, direct — is included.
- The quadrant logic and labels are unchanged (ALIGNED and the rest carry over directly).
- The Y-axis becomes total pageviews (PV), representing all human traffic rather than search clicks specifically.
AHTG is not a competing framework or a simplified substitute for AHQG. It’s the same underlying question — where does AI and human attention diverge? — asked at a coarser resolution, across every channel a page receives traffic from.
How AHQG and AHTG Relate
| What it measures | Traffic scope | |
|---|---|---|
| AHQG | Search intent gap | Search only |
| AHTG | Overall traffic gap | All channels |
Both frameworks are answering the same structural question about AI-human divergence — they simply operate at different resolutions. That shared structure is precisely why AHQG’s quadrant classification transfers to AHTG without modification: the axes measure different inputs, but the underlying pattern they’re classifying is the same.
A Hypothesis Worth Flagging
Here’s a claim worth stating explicitly, even before it’s fully tested: e-commerce sites tend to be driven more by paid acquisition than by organic search. If that’s true, then AHQG — built around search intent — may be the wrong primary lens for e-commerce. AHTG, which looks at traffic as a whole, may turn out to be the more natural framework for understanding the AI-human gap on commerce sites specifically.
That’s a claim we intend to test directly, not just assert.