AI Agents Outpace Measurement as Marketers Deploy Faster

AI agents are transforming customer interactions, yet a significant gap remains between deployment and accurate revenue attribution. Marketers are rapidly integrating conversational AI into their workflows, but the systems designed to track media performance often fail to capture the outcomes of these digital conversations. According to Invoca’s “2026 Broken Funnel Report,” this disconnect is creating a new layer of complexity in an already fragmented marketing landscape.

The adoption rate for conversational AI is accelerating. Forty-four percent of B2C marketers currently have live AI agents in at least one channel, with another 27% actively testing them. Despite this widespread usage, only 19% can reliably send conversion data back to advertising platforms. This lack of connectivity means that while an agent may successfully answer questions, qualify prospects, or nudge users toward a purchase, the broader attribution ecosystem remains largely unaware of the event. Consequently, nearly three-quarters of marketers cannot track leads from initial click-through through to final conversion. Only 26% can tie specific agent conversations back to the originating campaign, keyword, or ad.

The Measurement Mismatch

This technological divide influences how companies evaluate the success of their AI initiatives. Because direct revenue tracking is limited, most organizations rely on proxy metrics. Only 27% use revenue or booked outcomes as their primary measure of performance. Instead, the majority depend on indicators such as customer satisfaction scores, the volume of qualified leads, conversation counts, the percentage of interactions handled without human intervention, and cost-to-serve figures.

These metrics provide insight into operational efficiency and activity levels, but they offer little clarity on whether the AI is generating business value sufficient to justify the investment. The inability to connect agent activity to financial results leaves marketers operating with incomplete visibility into return on ad spend.

Fragmented Customer Journeys

The challenge extends beyond AI agents to a broader attribution crisis within the customer journey. Seventy-four percent of surveyed marketers cannot fully track leads from their first digital interaction through to conversion. The most significant point of failure occurs when a lead reaches a physical location, such as a dealer, franchise, or local branch, where 56% lose attribution. Losses also occur frequently in digital channels, with 46% dropping off during text or chat interactions and 40% vanishing during phone calls. Only 21% of respondents claim they can both see and influence leads all the way through to conversion.

AI introduces additional handoffs into a journey that was already difficult to connect. A typical path might involve a paid campaign generating a lead, an AI agent managing the initial conversation, and a local salesperson closing the deal. If these disparate systems do not share data, each segment of the journey may appear successful in isolation, masking the true contribution of marketing efforts to the bottom line.

This fragmentation also impacts the quality of customer experience. Only 27% report that their AI agents can access and utilize a customer’s previous digital journey. Without this context, an agent may treat a returning prospect as a new visitor, even if the individual has previously researched products or engaged with the company through other channels.

Marketers recognize this limitation as a critical area for improvement. The most requested feature for AI agents is the ability to tailor conversations based on the customer’s digital history, cited by 43% of respondents. Another 41% emphasized the need for agents to recognize returning customers and leverage their past interactions to personalize service.

Emerging Lead Sources

The shift in lead generation begins before the customer ever interacts with an AI agent. Marketers are increasingly identifying AI-powered search tools as high-quality sources of traffic. Forty-four percent stated that AI answer engines like ChatGPT, Claude, and Gemini produce their most qualified leads, outperforming Google Local Services Ads (40%), organic search (39%), paid social (37%), and paid search (36%).

Most marketers can identify at least some leads that originated their research in an AI answer engine, with 86% confirming this capability for most or some leads. However, visibility diminishes sharply as these customers move through subsequent stages involving calls, chats, agents, and local sales conversations.

This creates a complex environment for media optimization. While AI may influence the initial research phase and an AI agent may handle the next step, the eventual sale might occur elsewhere. If the final conversion data does not flow back to the systems responsible for buying media, campaigns continue to be optimized based on a partial view of performance. The race for AI agents has moved well past initial deployment; the current priority is establishing robust connections between agent insights, customer behavior, and ultimate revenue generation.

Invoca’s findings are derived from a survey conducted in September and October 2026 among 451 U. S. marketers at companies with at least 100 employees. The industries represented include home services, healthcare and dental, financial services and insurance, automotive, and telecom.