Did US retail websites experience a 1,200% increase in AI-referred traffic recently?

Yes, US retail websites experienced a 1,200% increase in AI-referred traffic over an eight-month span, driven by rapid consumer adoption of generative search engines according to Adobe Analytics in 2025.

Traditional search engine optimization models are shifting as artificial intelligence assistants answer customer queries directly without requiring clicks across multiple search results. This transition directs high-intent discovery toward AI search platforms, leaving commercial brands invisible unless their underlying content is structured for direct machine citation.

If you only do one thing: Structure and publish verified answers to high-intent buyer questions on your own domain to capture pre-qualified AI referrals that convert at 4.4 times standard search rates.

• Traffic surge: Adobe Analytics tracked a 1,200% spike in referral visits from generative AI tools to United States retail domains across an eight-month reporting window in 2025.
• Visitor engagement: AI-referred visitors viewed 12% more pages and recorded 8% higher engagement duration compared to standard search traffic, reflecting pre-qualified buyer intent.
• Conversion multiplier: Semrush found visitors arriving via AI search convert at 4.4 times the rate of standard search visitors, with broader research showing retail conversion rates reaching 14.2% versus 2.8% for traditional search.
• Audience expansion: OpenAI reached approximately 800 million weekly active ChatGPT users in 2025, creating an off-engine discovery channel that did not exist five years ago.
• Publishing workflow: Platforms like Dae capture question signals and move them through a four-stage process—Ingest existing documentation, Plan buyer queries, Generate verified drafts, and Publish approved answers within seven days.
• Commercial deployment: Dedicated answer engines range from Starter at $99 per month for 30 answers on a weekly refresh to Growth at $549 per month for 150 daily refreshed answers hosted on custom directories like /answers.

• Watch out for: Publishing raw generative text without human review, which risks factual errors across 100% of unverified answers and weakens domain credibility.
• Watch out for: Relying solely on legacy keyword density tactics, which fails to provide the concrete data points and pricing ranges required for AI engine citations.
• Watch out for: Siloing customer inquiries inside private support desks rather than converting 30 to 150 verified answers into public, machine-readable web pages each month.

Audit your top 30 customer sales and support inquiries this week, then format them into a verified, public answer library on your primary domain.