The Shift From Information-Seeking to Agentic Task Completion
For decades, search engines have acted as a directory, pointing users toward the right website. However, we are witnessing a fundamental pivot. Search is evolving from a tool used for information-seeking into an “agentic” experience—one where the AI doesn’t just find a link, but completes the task for you.
This transition is evident in the rollout of features like Canvas in AI Mode, which transforms a simple query into a full, editable itinerary. Instead of bouncing between ten different tabs and a notes app, users can now organize flights, hotels, and attractions in a single side panel that integrates real-time data from Google Maps and the web.
Understanding the “Agent Manager” Model
Industry leadership, including Google CEO Sundar Pichai, has suggested that Search is becoming an “agent manager.” In this architectural role, the search engine acts as an orchestration layer, overseeing various agents to execute multi-step tasks.
This is supported by research such as the SAGE research paper, which focuses on training agents for reasoning chains over four steps. Further advancements, like CW-GRPO and SKILL0, are pushing agents toward reinforcement learning and the internalization of “skill packages,” allowing them to operate with less instruction overhead during inference.
The New Reality for Travel and Local Businesses
The move toward task-based agentic search (TBAS) is radically changing how visibility works for specific industries. For travel brands, the “early funnel” is shifting. When a user builds a trip via Canvas, the AI handles the selection logic for hotels and attractions.
the introduction of Flight Deals—a conversational, AI-powered tool—allows users to find bargains by describing their ideal trip rather than manually applying filters. Even as this improves the user experience, it creates a challenge for brands: if the planning happens entirely within the AI interface, traditional click-through rates to third-party sites may decline.
The Impact on Local Retailers
Local businesses are now facing a new discovery surface. When AI agents contact stores to verify stock, the “customer” is no longer a human, but a bot. This raises critical questions about how eligibility for these calls is decided and whether specific business signals influence which stores the AI chooses to contact.
The Visibility Gap: The “Invisible” Search Experience
Perhaps the most pressing issue for search professionals is the growing measurement gap. While AI capabilities are accelerating, reporting tools have not kept pace. Businesses currently have no way to track if they were included in a Canvas itinerary, if an AI agent called their store, or if their property triggered a price-tracking alert.
This lack of transparency extends across platforms. Whether it is OpenAI’s Operator using GUI vision to interact with pages or Perplexity routing across multiple models, the data on how businesses are surfaced remains largely hidden from the business owners themselves.
The Pressure on Digital Publishers
Publishers are feeling the impact of AI-driven summarization. Data from Index Exchange reveals a stark trend: 69% of analyzed publishers experienced year-over-year declines in ad opportunities, with an average drop of 14%.
- High Impact: Health and careers publishers saw ad drops between 40-50%.
- Low Impact: News and politics publishers saw smaller declines (around 7%), as users still seek trusted sources for critical international or local events.
Cross-Platform Agentic Trends: Microsoft and Beyond
The trend toward agentic AI is not limited to search. Microsoft has integrated similar capabilities into its Office Product Group, making Copilot’s agentic features generally available in Word, Excel, and PowerPoint. The goal is to move from “suggesting steps” to actually performing the operate—such as formatting, restructuring, and transforming data.
As these tools evolve, the standard for business data must also change. Schema.org was designed for crawling, but agents require real-time inventory, booking availability, and action endpoints to be truly effective.
Future Inflection Points
Industry forecasts point toward a significant inflection point around 2027, particularly for non-engineering workflows and agentic business processes. As platforms expand these surfaces, the demand for “agent-readable” business data will likely supersede traditional SEO strategies.
Frequently Asked Questions
Agentic search is a shift from providing a list of links to using AI agents that can execute multi-step tasks, such as booking a restaurant, calling a store to check stock, or building a full travel itinerary.
Canvas is a planning workspace in Google Search that allows users to describe a trip and receive a customized itinerary. It integrates real-time data for flights and hotels alongside Google Maps details.
AI-driven summarization allows users to get answers directly on the search page, reducing the need to click through to the publisher’s website, which in turn reduces ad impressions.
It is the lack of reporting tools that allow businesses to see if their products or services are being recommended by AI agents or included in AI-generated plans.
Join the Conversation: How is your business adapting to the rise of agentic AI? Are you seeing a shift in your traffic patterns or discovery methods? Let us know in the comments below or subscribe to our newsletter for the latest insights on the future of search.
