The agent development lifecycle begins with three phases: build, test, and deploy. While many organizations have overcome the initial hurdle of creating their first agents, the real enterprise challenge starts immediately after deployment. 

Once agents begin handling real transactions, they often become black boxes. Teams can see the outcomes but lack critical visibility into why the agent behaved a certain way or which reasoning path it followed. Without this insight, tuning slows, trust diminishes, and scaling becomes an expensive gamble. 

This next phase requires continual refinement, performance optimization, and reliable monitoring of each agent’s health. To achieve enterprise-grade operational rigor for a growing network of agents, organizations need robust observability tools to close the loop. These work in tandem with testing tools, ensuring the agent lifecycle doesn’t end at deployment but continues through continual observation and iteration.

Today, Salesforce announced new agent observability tools, bringing deeper visibility, monitoring, and optimization to every AI agent within the Agentforce 360 Platform. By providing granular insight into agent behavior, the tools empower organizations to safely scale AI and continually improve agent performance. And by fostering better collaboration between humans and intelligent agents, these observability tools ensure that AI adoption is both reliable and trustworthy, accelerating every company’s journey to becoming an Agentic Enterprise.

With full visibility into every agent interaction, we can see exactly how our AI handles each request, step by step.

Ryan Teeples, Chief Strategy Officer, 1-800Accountant

The New Mandate: Why Observability Is Non-Negotiable

Fueled by a 282% surge in AI implementation, businesses are racing toward large-scale AI agent adoption, where one truth has become clear: seeing is believing. To succeed in deployment at scale, teams require a clear, continuous view of agent performance, learning, and improvement metrics. This need is critical as agents handle live conversations and make decisions that directly impact revenue and customer experience. Visibility into agentic behavior has become the new mandate, transforming uncertainty and guesswork into the trust, transparency, and measurable value necessary to scale AI.

“As AI adoption accelerates, the biggest enterprise challenge will no longer be about building an organization’s first agent; it will become how to best manage a fleet of agents that are making real-world business decisions. You can’t scale what you can’t see. Agentforce delivers the enterprise-grade operational rigor required for this new agentic world. We’re giving IT leaders the critical tools to continuously track performance, debug issues, and prove the ROI of their AI investments, ensuring every intelligent agent performs reliably, securely, and with total transparency.” – Adam Evans, EVP & GM, Salesforce AI 

What’s New in Agentforce

The new observability tools within Agentforce span three core areas: analytics, optimization, and health monitoring. 

Agent Analytics: Continuous Refinement

Agent Analytics provides a comprehensive view of how every agent is performing, translating performance into meaningful data, trends, and insights.

• Track Agent Performance: Monitor usage and effectiveness metrics across all deployed agents, giving teams a clear picture of how agents perform in real customer interactions.
• KPI Trends: Surface KPI trends over time, making it easy to see where performance is improving, declining, or requiring attention.
• Actionable Insights: Address ineffective topics, actions, or flows, enabling teams to take targeted steps to optimize agent performance.

Open Image Modal

Image Modal

Agent Optimization: Reasoning and Traceability 

Agent Optimization gives customers full observability into every agent interaction. By illuminating performance gaps, tracing session flows, and revealing exactly how agents make decisions, businesses can quickly diagnose issues, understand why an agent behaved a certain way, and take targeted actions to continuously correct and improve outcomes.

• Observe Every Interaction: Access end-to-end visibility of every Agentforce interaction to see exactly how agents respond, step by step, even across complex reasoning chains.
• Cluster and Analyze Sessions: Group similar requests to uncover patterns, friction points, and quality trends while scoring agent responses using intent, topic, and quality metrics.
• Optimize Agent Configuration: Identify configuration issues affecting performance and understand exactly where tuning, retraining, or guardrails are needed.

Open Image Modal

Image Modal

Open Image Modal

Image Modal

Agent Health Monitoring: Uptime and Reliability

Agent Health Monitoring ensures continuous agent uptime, reliability, and responsiveness by providing near-real-time visibility and actionable trust signals on your deployed fleet. This component is essential for operational rigor, ensuring your agents perform reliably and consistently even during peak loads.

• Monitor Agent Status Continuously: Track key health metrics in near real time, ensuring dashboard data is fresh and surfacing potential issues before they impact performance or trust signals.
• Resolve Failures Proactively: Receive high-speed alerts on critical errors, latency spikes, and escalations, enabling teams to quickly detect, investigate, and resolve problems to minimize downtime.
• Maintain Enterprise Reliability: Leverage a system built for high availability and operational robustness, giving you confidence that continuous monitoring scales reliably with your entire agent fleet.

Open Image Modal

Image Modal

How It Works: Unified Visibility and Intelligent Governance

Companies are building lots of agents and they need them to interoperate across their organization — from customer and employee interactions to the back-end systems. They also need visibility into how agents are performing and have the ability to govern them. To enable this, deep observability in Agentforce 360 is powered by two foundational building blocks:

1. Session Tracing Data Model: This data model logs every interaction, including user inputs, agent responses, reasoning steps, LLM calls, and guardrail checks, and stores them securely in Data 360. This foundation provides unified visibility and granular, session-level insights so you can ensure agents are behaving as intended and responding appropriately.
2. MuleSoft Agent Fabric: This new capability provides a single place to register, manage, govern, and observe every agent, regardless of where it was built. 

With these new agent observability tools, plus robust governance in MuleSoft and unified context in Data 360, every company can get full end-to-end control of all their agents, all from within the Agentforce 360 Platform.

Hear from Some of Our Pilot Customers

We’ve been working deeply with our pilot customers, like 1-800Accountant, Engine, and Nexo, to understand what they really need in an observability solution for agents. They need deeper access to data, governance of any agent, and a continuous improvement process. Their early success underscores the trust and transparency critical for scaling their Agentic Enterprises:

“With full visibility into every agent interaction, we can see exactly how our AI handles each request, step by step. This insight lets us understand not only whether tasks are completed correctly but how decisions are made along the way. With our agents projected to autonomously handle 90% of incoming service requests, this new observability tool allows us to streamline support, maintain high quality, and scale confidently, ensuring our customers receive fast, accurate service every time.” – Ryan Teeples, Chief Strategy Officer, 1-800Accountant

“Agentforce Observability has transformed how we deploy and scale Agentforce across our business. Engine fields over 530,000 customer inquiries a year, and with near-real-time visibility into every interaction, we can see exactly how our AI agents perform, learn, and make decisions. That level of insight not only ensures accuracy and quality for our customers, it builds trust across our teams, allowing us to confidently expand Agentforce into customer service, sales, and internal operations without constantly adding headcount. Observability is the foundation that turns AI from a tool into a trusted, continuously improving teammate.” – Demetri Salvaggio, VP, Customer Experience & Operations, Engine

“Handling up to 25,000 inquiries a month, Nexo’s deeply experienced Client Care team is greatly empowered by Agentforce. The platform gives us complete visibility into every interaction, showing how AI complements our human service — helping clients navigate complex crypto requests while giving our team the insights to continually raise the bar for customer experience.” – Nikolay Nedev, Team Lead & Product Management, Nexo

“By observing every Agentforce interaction, we can understand exactly how our AI navigates advertisers through even the most complex tools. This insight helps us understand not just whether issues are resolved, but how decisions are made along the way. Since launching Agentforce, we’ve deflected 46% of support cases, freeing our team to focus on higher-value work while ensuring SMB advertisers get fast, accurate guidance and a smoother onboarding experience. Observability gives us the confidence to scale these agents, continuously monitor performance, and make improvements as we learn from their interactions.” – John Thompson, VP of Sales Strategy and Operations, Reddit

Availability 

• Deep observability in Agentforce Studio, including Agent Analytics and Agent Optimization, is available now.
• Agent Health Monitoring will be generally available in Spring 2026.
• Regional rollout for Agent Optimization and Agent Analytics: EMEA customers is on November 20, 2025, and APAC customers on November 21, 2025.

Learn More

• Get more information on the Agentforce Session Tracing Data Model
• Find out why 95% of AI pilots fail (and what the other 5% do differently)
• Check out lessons from real-world Agentforce rollouts
• Read how developers can use observability tools to craft better AI agents

This article may include references to services or features that are still in development and are unreleased. Customers should make their purchase decision based on fully released and available features.