Image of a woman at a computer overlayed by a UI of the RFP Agent in Slack

From RFP Scramble to Streamlined Workflow With Agentforce and Slack

Learn how an internal team built a Request for Proposal (RFP) agent with Agentforce that shaves hours off a tedious process.

作者:Hunter Reh, AI Architect at Salesforce2025 年 5 月 6 日

We’ve all been there. That sinking feeling when a particular email lands in your inbox: “Request for Proposal (RFP) Attached.” You know the drill— the RFP Scramble is about to begin.

It’s a process known for late nights, a frantic search for information, endless email threads, and versioned documents that often demands an average of 25 hours for a response. It’s high stakes, high effort, high tedium. It’s a perfect task for Agentforce.

The traditional RFP scramble

Historically, RFPs spend hours digging through scattered knowledge bases, locating old documents, and scouring email chains for relevant information.

And that’s just the beginning. There’s also the need to coordinate feedback from various cross-functional stakeholders operating under tight deadlines, and struggling to maintain version control over exchanged documents. This highly manual effort wasn’t just inefficient, it was draining and tedious.

Enter the Agentforce RFP Response Agent  

To address these challenges, the Salesforce partner architect team collaborated with the Slack partner engineering team to build and customize an agent to take on this burden. Meet the Agentforce RFP Response Agent, a dedicated agent that lives right where teams already collaborate: Slack.

Powered by the Agentforce platform, the RFP Response Agent leverages Salesforce’s next-generation Retrieval-Augmented Generation (RAG) technology to find, synthesize, and draft high-quality RFP responses right where the team is already working.

How it works

Instead of starting your next RFP with a sigh, imagine starting the process with a simple Slack message. Here’s how the Agentforce RFP Response Agent transforms the process:

  1. Initiate the RFP in Slack. Start a conversation with the Agentforce RFP Response Agent. You can use pre-defined prompts for common RFP sections or input your own custom queries based on the specific RFP requirements.

    Begin drafting your RFP response by selecting a common prompt or asking a specific question.

  2. Get answers to research questions fast. Engage in detailed discussions with the RFP agent for any questions about your RFP response data, improve your answers, and explore different viewpoints, all within your Slack workspace. Using Agentforce Data Libraries, the agent can find and synthesize information to provide comprehensive responses. Because it’s powered by advanced Retrieval-Augmented Generation (RAG), answers are contextually relevant and accurate.

    Interact with the agent to explore topics like product features, security protocols, or service details.

  3. Draft collaboratively with canvas. Once you’ve gathered the core information, simply ask the agent to compile the findings into a structured format within a canvas.

    The RFP agent creates a dedicated canvas, automatically populating it with a clear and concise synthesis of the essential RFP details.

    The RFP agent creates a dedicated canvas, automatically populating it with a clear and concise synthesis of the essential RFP details.

  4. Refine with human expertise. The generated canvas serves as your collaborative workspace. Share it instantly with your team members and subject matter experts. Everyone can then review, edit, add insights, and hone the response together, leveraging canvas’s built-in formatting and commenting features.

    The canvas becomes a central hub for team collaboration, refining the AI-generated draft with human expertise.

Why a work operating system like Slack was essential

Could you improve the RFP process without Slack? In theory, yes. You could replicate this with disconnected tools, such as a standalone AI web app, email for coordination, and a separate shared document. However, this would reintroduce the friction and inefficiency the Agentforce RFP Agent is designed to eliminate. Slack provides unique advantages that significantly enhance the end user experience.

  • Work, uninterrupted. The entire RFP process happens within Slack, eliminating the need to switch between apps and tabs. Finding information feels more like asking a colleague thanks to Slack’s conversational interface.
  • Teamwork made easy. Slack canvas offers a documentation environment built for real-time, fluid teamwork, a world away from emailing document versions back and forth.
  • The future of work. Ultimately, Slack acts as a central nervous system, a work operating system that integrates the power of Agentforce and RAG technology with the people you work with and the tools you use every day.

Gaining more efficiency, quality, and focus across the organization

The Agentforce RFP Response Agent significantly improves the entire RFP lifecycle, going beyond just saving drafting time. By using next-generation RAG technology, it ensures higher quality and more consistent proposals based on the latest, most accurate information. Centralizing the process in Slack also streamlines collaboration, making it easier for teams to contribute their expertise effectively. Furthermore, by automating information gathering and initial drafting, teams can concentrate more on tailoring the proposal strategy and value proposition for each specific client. By bringing intelligent automation and seamless collaboration together in Slack, the Agentforce RFP Response Agent offers a powerful way to transform one of the most challenging business processes into a streamlined, efficient workflow.

Get Started with your own Agentforce RFP Agent in Slack

Interested in bringing this agentic productivity to your own RFP workflow? Download our 3-step guide to configure your own Agentforce RFP Response Agent.

How is your team experimenting with Agentforce? Share your stories with us at @SlackHQ!

    這則貼文有幫助嗎?

    0/600

    超讚!

    非常感謝你提供意見回饋!

    知道了!

    感謝你提供意見回饋。

    糟糕!我們遇到問題了。請稍後再試一次!