Artificial intelligence is becoming part of the ecommerce shopping journey without replacing search, according to Similarweb’s State of Ecommerce 2026 report. The research suggests consumers are stacking AI assistants and traditional search rather than choosing one over the other, creating a more fragmented path from product discovery to purchase.

The report, released September 10, finds that direct referrals from dedicated AI platforms increased more than 200% year over year but still generate relatively limited traffic for ecommerce marketplaces and online stores. The more important signal may be what happens before the click: Similarweb says AI recommendations can materially influence purchasing decisions, in some cases giving a recommended brand a 2-to-1 advantage over competitors.

The findings put pressure on ecommerce marketers to rethink what counts as product discovery. A shopper may ask ChatGPT or Google Gemini for recommendations, narrow the options through an AI conversation, then turn to Google Search before making a purchase. According to Similarweb, 89% of consumers who use AI during shopping research also use search.

That makes AI visibility less about replacing search engine optimization and more about becoming visible across multiple discovery environments.

The report, produced by Similarweb in collaboration with Statista, also examines the continuing migration from ecommerce websites toward mobile apps, as well as trends across marketplaces, beauty and cosmetics, consumer electronics, fashion and apparel, and home and garden.

Market Landscape

The ecommerce market is entering a period in which traffic and influence are becoming increasingly disconnected.

Search remains a major source of measurable visits, while AI conversations can influence a shopper without producing an immediate referral. Similarweb's findings therefore point toward an emerging distinction between traffic generation and demand influence.

That distinction is particularly important for brands competing for product recommendations. If an AI assistant helps a consumer narrow five possible products to two, the eventual website visit may not reveal which conversation shaped the decision.

Similarweb has previously reported rapid growth in generative-AI referrals and increasing use of AI for product research, reinforcing the idea that AI is becoming part of the discovery layer of ecommerce rather than simply another traffic channel.

The report's finding that AI users also rely on search complicates the idea that Google, ChatGPT or other platforms will simply replace one another. Instead, shoppers appear to be moving between them based on the stage of the journey.

That creates a more demanding environment for ecommerce teams. Product information must be discoverable, structured and authoritative enough to work across search engines, marketplaces, retailer apps and AI interfaces.

The broader market is also continuing to expand. Statista projects worldwide ecommerce revenue to reach about $4.91 trillion by 2030, compared with approximately $3.86 trillion in 2026.

Strategic Outlook

For retailers and brands, the immediate implication is that traditional acquisition metrics may provide an incomplete picture of how demand is created.

SEO, paid search, retail media and direct traffic remain measurable components of the funnel. AI introduces another layer in which consumers can receive comparisons, product explanations and recommendations before ever visiting a merchant's site.

That could change how ecommerce organizations approach content. Product descriptions, reviews, expert information and other authoritative sources may increasingly serve two audiences: human shoppers and AI systems interpreting product information.

The app-versus-web shift adds another complication. Similarweb says ecommerce app sessions are growing at roughly 1.3 times the rate of web visits, while 86.5% of U.S. consumers surveyed say they primarily shop on a smartphone or tablet. This makes mobile experience, app engagement and AI-assisted discovery interconnected parts of the commerce strategy.

For enterprise marketing teams, the emerging goal is therefore not simply to "optimize for AI." It is to build consistent product visibility across the full discovery ecosystem while measuring where possible how those interactions contribute to consideration and conversion.

Top Insights

• Similarweb finds AI referrals growing rapidly, but search remains central, suggesting ecommerce marketers must optimize for multiple discovery environments rather than replace SEO.
• AI recommendations can disproportionately influence purchasing decisions, making brand visibility inside conversational shopping journeys increasingly important for ecommerce companies.
• The 89% overlap between AI-assisted shopping research and search shows consumers are combining discovery tools rather than abandoning traditional search.
• Growing ecommerce app usage adds another layer to fragmented customer journeys, increasing pressure on brands to coordinate web, app, search and AI experiences.
• Statista's projected $4.91 trillion ecommerce market by 2030 underscores the commercial stakes as AI reshapes how consumers discover and evaluate products.

Get in touch with our MarTech Experts.