Contentsquare brings AI agent journeys into experience analytics
Contentsquare's new experience platform adds AI-agent journey analytics and Sense-led testing, but broad access remains in beta.
Contentsquare has introduced a reworked experience analytics platform that aims to measure AI agents alongside human customers. The October 13 announcement extends conventional journey analysis into interactions involving large language models, autonomous agents, web and mobile sessions, and customer conversations.
That is a practical measurement challenge for marketing teams. An AI assistant may research a product, refer a buyer to a website or interact with the site itself. Existing dashboards can count some of those visits, but understanding what the agent did and whether it helped a human customer requires different classification and attribution.
Table of contents
Jump to each section:
- How the platform tracks human and agent journeys
- Sense connects experience analysis to experimentation
- Where Contentsquare fits among analytics rivals
- What marketers should verify before adopting it
Key Takeaways
- Contentsquare has unveiled a customer experience platform designed to analyze both human and AI-agent interactions.
- Its Sense layer is intended to connect detected friction with prioritization, experimentation and measurement.
- The new platform is in selected-customer beta; rollout timing and results at scale remain unproven.
How the platform tracks human and agent journeys
The new Experience Platform brings together human-session data, LLM and agent traffic, conversation intelligence, and journeys that span sessions and devices. Contentsquare says its analytics can distinguish agent interactions from human sessions, trace AI-referred activity and monitor autonomous agent behavior through WebMCP-related signals.
The distinction matters because an agent arriving at an ecommerce site may be carrying out a shopper's legitimate request. Combining it indiscriminately with human visits can distort bounce rates, conversion funnels and retargeting audiences. Treating every automated visit as unwanted bot traffic may also hide a new distribution channel.
Contentsquare also describes multi-touch attribution that includes LLM and AI-referred touchpoints. Marketers should verify which connections and identity signals are available in their own environments. An AI referral can be observed without proving which earlier recommendation caused the sale.
Sense connects experience analysis to experimentation
Contentsquare is presenting its Sense AI layer as a way to move from finding problems to taking action. The company says Sense can detect friction, estimate the commercial impact, raise alerts and create workflow items. It also announced agentic experimentation intended to generate hypotheses, help design tests and evaluate outcomes with human approval.
Jonathan Cherki, Contentsquare's founder and CEO, said that “AI is only as powerful as the context behind it.” The context in this case is behavioral data linked to the site experiences marketers can change. A proposed fix becomes useful only when the team can inspect the underlying evidence, authorize the test and measure whether a conversion improvement is incremental.
The release also describes new feature-level benchmarks and integrations with experimentation partners including Dynamic Yield, Optimizely and Kameleoon. Those integrations can be useful where teams already use separate testing stacks, but implementation details and entitlements will determine how much work is actually automated.
Where Contentsquare fits among analytics rivals
Experience analytics is already a competitive field. Amplitude, Fullstory and Quantum Metric address overlapping needs in product behavior, session analysis and conversion friction. Amplitude's 2026 annual filing identifies Contentsquare and these vendors across its competitive landscape.
Contentsquare's new emphasis is the proposed continuity between human and agent journeys, an AI assistant that ranks problems, and experimentation inside the same operating environment. A comparable feature checklist alone will not establish whether this is more accurate or useful than existing combinations of analytics and testing tools.
An independent report on the relaunch also frames personal AI agents as a potential new participant in customer experience. That is a credible reason to revisit measurement design, although the commercial scale and timing of autonomous-agent interactions still differ widely by sector.
What marketers should verify before adopting it
The reworked Experience Platform is available in beta to selected customers, with broader access planned in phases. Contentsquare has not established in its announcement when every capability will become generally available in all markets.
Marketing and product teams evaluating it should ask to see how the platform labels verified agents, crawlers, LLM referrals and ordinary visitors. They should also ask what attribution can be reconstructed across devices and conversations, how consent is respected, and which AI recommendations require human approval before a live experiment begins.
A sensible first test would compare an existing conversion journey report with the new human-and-agent view on a limited site section. Check whether the additional signals reveal a real friction point, whether a proposed intervention can be tested against a control, and whether the result is consistent enough to change budget or product priorities.
Contentsquare's announcement points to an emerging analytics requirement. Brands increasingly need to know which actions were taken by people and which were taken by software on their behalf. The beta's value will depend on how accurately it identifies that boundary and how much extra work its insights actually eliminate.
