As AI assistants move from answering product questions to influencing — and potentially completing — purchases, brands are facing a new advertising and commerce problem: visibility inside AI is difficult to measure. NIQ and Similarweb are developing a new Agentic Commerce Measurement solution designed to show brands and retailers where products appear in AI-driven discovery, how consumers interact with those recommendations, and whether that activity ultimately contributes to traffic, conversion and sales.
AI assistants are becoming another place where consumers discover products, compare options and make purchasing decisions. For brands and retailers, that creates a measurement gap that traditional search, web analytics and retail reporting were not built to handle.
NIQ and Similarweb are looking to close that gap.
The companies announced a collaboration to develop a new measurement solution that connects NIQ’s product intelligence, consumer behavior and retail sales data with Similarweb’s digital signals from generative AI platforms and the broader digital journey. The initial solution is expected in the fourth quarter of 2026, beginning with selected categories and markets.
The premise is that AI is becoming a commerce channel that needs its own measurement layer.
Traditional ecommerce analytics can tell a brand how many consumers visited a product page, where some of that traffic originated and whether a purchase followed. But that becomes harder when a consumer asks an AI assistant for a product recommendation, evaluates several options inside the conversation and then follows an AI-generated path toward purchase.
In that environment, the brand needs answers to several new questions: Was its product recommended? Which competitors appeared alongside it? Was the product information represented accurately? Did the AI interaction generate a visit? And did that interaction eventually contribute to a purchase?
NIQ and Similarweb are designing their solution around those questions.
The companies say the platform will initially focus on five areas: consumer intent, agentic shelf visibility, product content readiness, AI-driven traffic and AI-driven conversion.
Consumer intent is intended to show what shoppers are asking AI assistants and which questions or needs influence their decisions. This is potentially valuable because prompts can expose purchase intent in a way that conventional keyword data may not fully capture.
Agentic shelf visibility applies a familiar retail concept to AI. Instead of asking whether a product ranks on a traditional search results page, brands can evaluate whether AI assistants recommend their products when consumers ask for relevant products or solutions.
That creates a new competitive metric.
A brand could have strong visibility on Google, Amazon or retailer websites but still be poorly represented inside conversational AI. Conversely, an emerging brand could gain disproportionate exposure if AI systems frequently recommend it in relevant product conversations.
The third component, product content readiness, addresses the data foundation behind those recommendations. AI systems need structured, complete and understandable product information to identify products and accurately describe them. Missing specifications, inconsistent descriptions or poorly structured catalog data can therefore become a visibility problem rather than simply a content-management issue.
This could make product information management increasingly important to both ecommerce and advertising teams.
The traffic component then attempts to connect AI discovery with measurable digital behavior. Similarweb will contribute digital intelligence designed to identify traffic originating from AI platforms as well as visits influenced by AI-driven discovery.
The final piece is conversion.
NIQ brings retail sales measurement into the equation, allowing the companies to move beyond visibility and traffic toward the commercial outcome brands ultimately care about. The goal is to connect AI-influenced interactions with verified omnichannel purchasing behavior.
That distinction is important because AI visibility alone is not necessarily valuable.
A product can appear frequently in AI recommendations without generating meaningful sales. Similarly, a brand could receive AI-driven traffic that converts poorly because the landing-page experience, pricing or availability does not match the consumer’s expectations.
Connecting discovery, traffic and sales gives advertisers a way to evaluate whether AI exposure is actually contributing to business performance.
The development comes as technology companies build infrastructure designed to let AI agents participate directly in commerce. Google and OpenAI are developing protocols that can support agent-mediated product discovery and purchasing, potentially compressing the traditional path from search to product page to checkout into a conversational workflow.
That could fundamentally change how digital product visibility works.
In conventional advertising and ecommerce, brands optimize across search engines, marketplaces, retailer sites, social platforms and paid media. In an agentic commerce environment, they increasingly need to optimize for the systems that interpret product information and decide which products to recommend.
This is where the NIQ-Similarweb collaboration intersects with AdTech.
The emerging discipline resembles a combination of search visibility measurement, product intelligence, audience analytics and attribution, but the decision-maker is no longer necessarily a consumer navigating a results page. It can be an AI system interpreting a natural-language request and selecting products on the consumer’s behalf.
For enterprise marketing and commerce teams, that means existing measurement stacks may need another layer.
Brands will likely need to monitor AI recommendation share, competitor visibility, product-data quality, AI-originated traffic and downstream conversion. They may also need processes for identifying inaccurate product descriptions or recommendations and feeding corrected information back into their commerce content infrastructure.
The challenge will be attribution.
As AI assistants become embedded across browsers, shopping experiences, search products and standalone applications, separating direct AI traffic from broader AI-influenced behavior will become increasingly difficult. A consumer may see a recommendation in an AI conversation, return through a retailer site later and complete the purchase without an obvious referral path.
That makes NIQ’s retail measurement and Similarweb’s digital intelligence complementary, at least conceptually.
The competitive landscape is likely to expand quickly. Search analytics companies, ecommerce platforms, retail media networks, product-information providers and marketing intelligence vendors are all potential participants in the emerging AI-commerce measurement market. Google and Amazon have particularly strong positions because they control large portions of both product discovery and transaction infrastructure, while independent measurement providers can potentially offer a cross-platform view.
NIQ and Similarweb are betting on the latter approach.
Their planned solution does not attempt to become another AI shopping platform. Instead, it is designed to help brands understand what happens inside those platforms and connect AI discovery to measurable commercial outcomes.
That distinction could become increasingly important as agentic commerce develops.
If AI assistants become a meaningful gateway to product discovery, marketers will need to know not only whether consumers saw their advertising but whether AI systems considered their products relevant enough to recommend in the first place. The companies’ planned measurement platform is an early attempt to make that new layer of digital influence visible.
Market Landscape
The advertising and commerce ecosystem is moving from search-led discovery toward AI-mediated discovery. Google’s AI experiences, Microsoft’s AI ecosystem, OpenAI’s commerce initiatives and Amazon’s shopping technologies are all contributing to a landscape where consumers can increasingly research products through conversational interfaces.
For brands, this introduces a new optimization layer between product content and consumer demand. Traditional SEO and ecommerce analytics remain important, but AI recommendation visibility, structured product information and AI-influenced attribution are becoming emerging measurement categories.
The key industry challenge will be distinguishing AI visibility from AI value. Measuring how often a product is recommended is useful, but connecting those recommendations to traffic, conversion and verified sales will determine whether agentic commerce measurement becomes a meaningful enterprise discipline.
Top Insights
- NIQ and Similarweb are developing Agentic Commerce Measurement to connect AI product discovery with traffic, conversion and omnichannel sales outcomes.
- The solution will track AI recommendation visibility, consumer intent, product content readiness and AI-driven traffic across emerging commerce environments.
- NIQ contributes product intelligence and retail sales measurement, while Similarweb adds digital signals from generative AI platforms and online behavior.
- The partnership reflects a shift from measuring traditional search visibility toward understanding how AI assistants influence product consideration and purchasing decisions.
- Enterprise brands will need new measurement and optimization workflows as AI assistants become increasingly important gateways to product discovery and commerce.
Get in touch with our Adtech experts
