If AI crawlers copy your content, are they giving anything back? Microsoft Clarity’s new AI Scrape-to-Referral Ratio answers that question with a single, actionable metric: how many scrapes an AI operator performs for every referral it sends to your site.

What the new metric shows

The AI Scrape-to-Referral Ratio card compares recorded scrape activity to referral visits from AI sources. Instead of forcing teams to infer value from raw bot logs, the ratio highlights whether AI platforms are merely extracting content or actually driving users to your pages.

Microsoft illustrates the point with an example: one AI source produced 41 referrals but registered roughly 6,000 scrapes per referral. That kind of gap flags a high extraction volume with limited traffic return — a clear prompt to investigate further.

How Clarity surfaces the data

The Bot Analytics view in the Clarity AI Visibility Dashboard consolidates related signals so you can move from detection to validation:

  • AI Scrape-to-Referral Ratio card — a single metric that compares scrape counts and AI referral traffic at a glance.
  • Operator-level breakdown — a ranked list of referring operators so you can see which AI sources send visits and which scrape heavily with little referral return.
  • Mapped-domain coverage safeguards — ratios are only calculated across accurately mapped domains to avoid misleading rollups when CDN and Clarity coverage differ.
  • Coverage context — Clarity notes when AI bot data and referral traffic cover different domains so you can interpret the ratio correctly.
  • Direct session-recording links — click a referral source to open session recordings filtered for AI-referred visits, making it straightforward to validate attribution and inspect visitor behavior.
  • Traffic-quality signals — metrics on scrolling, engagement and conversions let you judge whether AI-referred visits are meaningful.

Microsoft’s recommended workflow is straightforward: open Bot Analytics, review the Scrape-to-Referral card and operator list, confirm mapped-domain alignment across data sources, pick a referral source and watch session recordings to evaluate visit quality and conversions.

Practical value for marketers and site owners

Site operators have long debated whether to allow AI crawlers free access. The Scrape-to-Referral Ratio changes that debate from abstract to measurable: it gives you an evidence-based signal to prioritize follow-up.

Use the ratio to triage efforts. Start with operators that show high scrape volumes but few referrals. Watch their filtered session recordings to determine whether the small number of referrals drives real engagement or conversions. If referrals prove low-value, consider tightening crawler access, adjusting robots rules, or refining attribution and CDN mappings to ensure data alignment.

Importantly, the metric is diagnostic, not prescriptive. It identifies candidates for action but doesn’t set a universal threshold — the acceptable scrape-to-referral rate depends on your site’s business model, content strategy and tolerance for extraction versus visibility.

What to watch next

Two immediate signals matter for teams deploying this metric: whether the ratio drives changes to robots rules, CDN or attribution configurations, and whether AI platforms with favorable ratios continue to send measurable traffic. Watch for shifts in operator rankings over time: a persistent high-scrape, low-referral profile is a stronger case for intervention than a one-off spike.

Clarity’s new card gives analytics teams a clearer way to quantify the return side of AI visibility and a direct path from metric to verification. Treat the Scrape-to-Referral Ratio as a prioritized alarm bell — it shows where to look and what to validate, not an automatic policy decision.