Published on: May 9, 2025 / Updated on: May 9, 2025 – Author: Konrad Wolfenstein
Artificial intelligence: Why Salesforce's Agentforce isn't (yet) taking off – independent alternatives are better – Image: Xpert.Digital
Salesforce Agentforce: Between Vision and Reality in the Age of AI
Autonomous agents in the enterprise: Why Salesforce is struggling
Since its initial unveiling in September 2024, Salesforce Agentforce has been considered one of the most promising AI platforms for creating autonomous agents in the enterprise environment. Despite grand announcements and ambitious goals from Salesforce CEO Marc Benioff—including the vision of activating one billion agents by the end of 2025—the reality paints a more nuanced picture. The platform, which aims to go beyond traditional chatbots and independently manage complex business processes, is struggling with adoption hurdles. While Salesforce has achieved some successes, revenue forecasts remain below expectations, and widespread market penetration has yet to materialize.
What is Salesforce Agentforce?
Agentforce represents Salesforce's vision of a "third wave of AI"—following Predictive AI and Copilots. It's a platform that enables companies to build and deploy autonomous AI agents that go beyond the capabilities of traditional chatbots and assistants. These agents make independent decisions, optimize processes, and operate autonomously without requiring constant human oversight.
Suitable for:
Core functions and architecture
The Agentforce platform is based on several key components that together form a powerful ecosystem:
- Agent Builder: A low-code/no-code tool that allows companies to configure their own AI agents using Salesforce tools such as Flows, Prompt Templates, Apex, and APIs.
- Atlas Reasoning Engine: This acts as the "brain" of Agentforce, where users can define parameters for natural language dialogues. The engine breaks down complex tasks into smaller steps, evaluates each step, and proposes a plan for further action.
- Einstein Trust Layer: Effectively protects data from Salesforce from unauthorized access by third parties, thus enabling the secure use of various Large Language Models (LLMs).
- Integration with the Salesforce ecosystem: Agentforce is natively integrated with the entire Salesforce Customer 360, enabling agents to leverage the full customer context from CRM applications.
The AI agents can be deployed in various business areas, including customer service, sales, marketing, and commerce. They can handle tasks such as answering customer inquiries, qualifying sales leads, and optimizing marketing campaigns.
Challenges and adoption hurdles
Despite its great potential, Agentforce faces several challenges that are slowing its adoption. These can be divided into different categories:
Complexity and learning curve
One of the biggest hurdles is the platform's complexity. Users report that implementing and mastering Agentforce requires a steep learning curve. Even after issuing correct instructions, the output often contains too much unnecessary information. Configuring an agent is described as time-consuming due to the excessive number of instructions required.
Performance problems
Users report slow loading times and occasional unresponsiveness. The user interface is sometimes perceived as overwhelming, with too many features and tabs, which can negatively impact efficiency.
Limited customization options
Advanced customization features are partially unsupported, limiting the platform's appeal for businesses with specific requirements. The agent marketplace is described as "thin" compared to competitors, forcing teams to build many features from scratch.
Integration problems
Problems with third-party integrations can lead to data synchronization errors. Documentation is often perceived as inadequate, especially for complex integrations, forcing developers to rely on forums to find solutions.
Costs and unpredictable pricing
Total operating costs can be prohibitive for smaller businesses. Users report unpredictable bills as usage increases, with some months exceeding the budget by double, making quarterly planning difficult.
Governance and Compliance
The governance tools are described as only mediocre, especially for companies in regulated industries. Compliance teams often have to create workarounds for functions that should be standard.
Market performance and financial impact
The challenges in adopting Agentforce have a direct impact on Salesforce's financial forecasts:
- In February 2025, Salesforce announced that its revenue forecast for fiscal year 2026, at $40.5 billion to $40.9 billion, fell short of analysts' expectations of $41.35 billion. This disappointment was primarily attributed to the slower adoption of the Agentforce platform.
- Rebecca Wettemann, CEO of the analytics firm Valoir, commented on the situation, saying that after bad experiences with generative AI experiments, companies are not prepared to “write blank checks until Salesforce proves that Agentforce actually works.” The next quarter or two will be crucial for Salesforce.
- Parker Snook, Senior Research Analyst at M Science, noted that Agentforce's monetization is "perhaps starting more slowly than many had expected".
Positive developments and success stories
Despite the challenges, there are also positive signs for Agentforce:
Growing customer base
By February 2025, Salesforce had completed 5,000 Agentforce deals, including more than 3,000 paid deals with well-known customers such as Singapore Air, Equinox and Jacuzzi.
Successful implementations
Some companies report positive results with Agentforce:
- OpenTable: Within three weeks of implementing Agentforce, the company manages 73% of restaurant web queries, a 50% increase over their previous tool.
- Goodyear: With Agentforce for field service, the tire specialist aims to reduce repair times by providing technicians with vehicle insights and automating appointment scheduling.
- Pfizer: The medtech company uses Agentforce to improve its customer-centric processes and supports 20,000 employees in sales and engagement.
Internal successes at Salesforce
Salesforce also uses Agentforce internally. Since the platform's launch in October, Agentforce has autonomously resolved 84% of 380,000 service requests, with only 2% requiring human escalation. Furthermore, quote generation cycles have been accelerated by over 75%.
Strategies to accelerate adoption
To accelerate the adoption of Agentforce, Salesforce has implemented several strategies:
Integration with existing platforms
Salesforce has integrated Agentforce with its existing platforms such as CRM, Slack, MuleSoft and Tableau to lower the barrier to entry.
Customization options and flexibility
The platform offers customization options that allow companies to create AI agents tailored to their specific needs.
Usage-based pricing model
Salesforce has introduced a usage-based pricing model for Agentforce, where each conversation costs $2. This approach is intended to encourage users to interact with the platform more frequently.
Continuous improvements
In December 2024, Salesforce announced Agentforce 2.0, which offers significant innovations such as a library of turnkey skills, the integration of AI agents into Slack, and advanced reasoning capabilities and RAG functions.
Will Agentforce take off?
The future of Agentforce depends on how well Salesforce can overcome current challenges. Experts agree that Salesforce's return to double-digit growth rates hinges on the success of the Agentforce platform.
The next one to two quarters will be crucial in demonstrating that Agentforce can indeed deliver the promised benefits. If Salesforce can successfully address concerns regarding complexity, performance, integration, and cost, the platform has the potential to become a game-changer in the field of autonomous AI agents.
Marc Benioff's vision of activating one billion agents with Agentforce by the end of 2025 remains ambitious, but not impossible if adoption gains momentum in the coming months.
Suitable for:
Autonomous AI agents: Why companies are still hesitant to introduce them
Salesforce Agentforce represents a significant advancement in the evolution of AI platforms for businesses. The combination of autonomous decision-making capabilities, deep integration into the Salesforce ecosystem, and a low-code development approach offers a compelling value proposition.
However, the slower-than-expected adoption demonstrates that the path to widespread deployment of autonomous AI agents is fraught with challenges. Companies are hesitant to invest in new AI technologies without seeing clear evidence of their effectiveness and return on investment.
For Salesforce, this means they not only have to overcome the technical hurdles, but also win customer trust by showcasing compelling success stories and measurable results. The coming months will show whether Agentforce can reach its full potential and deliver the promised breakthrough.
Suitable for:
Your AI transformation, AI integration and AI platform industry expert
☑️ Our business language is English or German
☑️ NEW: Correspondence in your national language!
Konrad Wolfenstein
I would be happy to serve you and my team as a personal advisor.
You can contact me by filling out the contact form or simply call me on +49 7348 4088 965 (Munich) . My email address is: wolfenstein ∂ xpert.digital
I'm looking forward to our joint project.
