Editor’s note: This post was updated on March 25, 2025, with the latest information and resources.
As Salesforce Admins, we have clear core responsibilities that define our day-to-day roles. We manage users, data, and product, provide actionable analytics, and oversee security. But we’re currently in the third wave of the artificial intelligence (AI) revolution: contextual and generative AI, characterized by prompt-based generative AI, real-time AI applications, and autonomous agents.
This means admins all over are questioning, “Where does AI fit into our roles?”
AI is widespread and the technology continues to advance. At Dreamforce 2024, Salesforce announced the launch of Agentforce, its suite of autonomous and assistive AI agents. Agents are designed to help admins increase customer success to support their users and customers. To see this in action, watch the Agentforce Dreamforce Keynote: Build the Future with AI Agents. At TDX this year, we continued to see advancements in Agentforce, and learned how admins can become Agentblazers with Salesforce and discover how to bring humans and agents together to drive customer success.
Salesforce also announced that AI certification exams, currently including certified AI Associate and certified Agentforce Specialist, will be free for Trailblazers to attempt through December 2025. This demonstrates the importance of this knowledge base for Salesforce professionals and Salesforce’s investment in your professional growth and continued education.
The AI Associate certification is a foundational knowledge layer of AI as it relates to CRM. This exam is very high level and focuses on AI basics, Salesforce’s trusted AI Principles, and the ethical and responsible handling of data. The end goal of an AI Associate is being able to identify how AI can support their organization’s business processes and have discussions with stakeholders that consider ethical implications.
The Agentforce Specialist certification expands upon the basics tested by the AI Associate exam by putting these principles into practice, actually using generative AI within Salesforce. Admins are a key audience for this exam because they already have the foundational knowledge of Salesforce’s core capabilities as well as an understanding of their organizational processes and user needs. In fact, in the AI era, most admins will also be AI Specialists since they are the product owners and CRM specialists.
The good news is that admins have an advantage when it comes to this exam—they already have incredible familiarity with the Salesforce Platform including configuration, processes, best practices, testing, and troubleshooting. Therefore, the goal for admins venturing into AI is to understand how their existing knowledge functions within the context of Agentforce features such as Prompt Builder, Agent Builder, and the Einstein Trust Layer.
If you’re new to AI, first try out a large language model (LLM) outside Salesforce like ChatGPT. You can get a sense of how they work while building a study plan for the exam. You can simply ask the LLM to help you build a study plan for the Salesforce Agentforce Specialist exam. Your request is called a prompt, and you’re getting a sense of how you can structure your prompts. Now, this prompt is a very broad statement without a lot of perimeters. When I did this with ChatGPT, it gave me an 8-week recommended study plan broken into 2-3 day focus areas for the week’s topic.
But in my case, this exam is a priority and I want to fast-track my study plan. I can modify my prompt to inform the LLM how much time I have to study (maybe I prefer a 4-week study plan), what areas I need to focus on (include the percentages of the exam topic breakdown), and what my current experience/exposure level is.
Wow. Even though I shortened the time frame, I got more detailed results and even some study resources. This is a great way to practice your prompt-building skills as well as set yourself up with a study plan.
Keep in mind that if you use an LLM to create a study plan, you should review it yourself for practicality, relatability, and what works for your schedule. This is an important reminder when using AI tools—verify the output before using.
Now let’s take these prompt building blocks to Salesforce.
Now that you’ve seen AI in action outside of Salesforce, let’s take it to our comfort zone: within the platform. The exam is broken down into five objectives.
Let’s start with some foundational information and Agentforce concepts. Security is one of the core responsibilities of a Salesforce Admin, and the Einstein Trust Layer builds on our existing knowledge base. The Einstein Trust Layer adds on additional security components. In fact, this addresses several of the questions I get asked consistently as a Salesforce Admin, including the external accessibility of data. The Einstein Trust Layer allows you to assure stakeholders that there is a zero-data retention policy, ensuring no third-party retention, training, or access to your data is possible. Another key concept in this exam objective is implementing grounding features. This is a dream come true for admins that can specify which data points within Salesforce contribute to the AI response.
The Agentforce Concepts section also dives into Agentforce actions and using Agent Builder. Agentforce offers both standard and custom actions, and as an admin it’s our job to determine which to use and how to watch our agents as well as our user adoption. We’re often focused on how to create a more streamlined and better user experience so that we can increase CRM adoption. Setting up agents is a big win for your users because you can create conversational AI assistants to answer questions or help with common tasks for them, or create autonomous agents to manage automation and simplify your business processes.
This section also covers Agent Development lifecycle extending beyond just building an agent. It is important to understand how to test an agent using the Testing Center UI as well as how to deploy agents from sandbox to production. Admins are familiar with both testing and deployment, but it is important to note that testing agents is different from the testing we are familiar with, because we are no longer testing pass/fail, but for trustworthiness and accuracy.
Prompt Engineering is one of the largest percentage areas of the exam and, thankfully, Prompt Builder tends to be the feature admins are most drawn toward using. The reason admins tend to excel in this section is because we have all had situations where we wished we could just combine different data points for a summary, which helps us connect the dots and generate real-world use cases. Data management is also one of the core responsibilities of a Salesforce Admin, so we’re familiar with the data points in our organization and can easily identify which ones will be essential for properly grounding a prompt.
Plus, the best part about being a Salesforce Admin is building on the platform! Prompt Builder is a low-code tool and gives us a chance to do this using the different skills we already excel at, such as custom configuration and automation via Flow.
Using Agentforce requires Data Cloud be turned on in your Salesforce instance, so it is important that admins have an understanding of how Agentforce and Data Cloud work together. This section primarily focuses on configuring your Agentforce Data Library to provide your agents with your organization’s relevant data so it can answer questions thoughtfully and accurately.
Generative AI in CRM applications—specifically Sales Cloud and Service Cloud—is where you can start to specialize in your own organization’s functionality. Learn what features are available to you and identify use cases where you can make the most of these AI features. Agentforce and Sales Cloud encounters topics such as Einstein Lead Scoring, Einstein Sales Emails, and Agentforce Sales Agents (SDR and Sales Coach). Agentforce and Service Cloud encounters topics such as Agentforce Service Agents, Einstein Service Replies for Email, and Einstein Call Summaries.
Learning about AI in general is a great foundation for any aspiring Salesforce Agentforce Specialist. As admins, we have several other resources to use in preparing for this exam.
To learn how to use Salesforce AI products, take the Drive Productivity with Salesforce AI trail which offers videos, articles, and hands-on challenges to help you get familiar with using AI in Salesforce. Then, when you’re ready to schedule your exam, check out the Cert Prep: Agentforce Specialist trail which offers flashcards and sample exam questions. If you learn best in a classroom setting, join us at an Agentforce Now Tour event and participate in hands-on workshops with the Salesforce Trailblazer Relations Team. You can also get hands-on and dive into Agentforce by earning your Agentblazer Status.
And for more great content, join us during the Agentforce World Tour and TDX Global for exciting sessions, demos, training, and the chance to build your own agent and participate in an Agentforce Hackathon.
In conclusion, admins make great Agentforce Specialists because they already understand the Salesforce Platform, and they know the data points inside and out that make AI initiatives successful. Admins have the growth mindset to approach this exam by building on their current skill set with hands-on experience.
With Agentforce, AI is deeply embedded in Salesforce which means that as the product managers of their CRM, admins will inevitably need to at minimum understand AI features, even if they aren’t the ones writing the prompts and implementing agents.
By approaching this exam with your existing admin skills, understanding the core responsibilities, and doing hands-on preparation through Trailhead, Salesforce events, and referencing Help articles, you can become a double-threat admin + Agentforce Specialist. Who’s with me and plans on taking the Agentforce Specialist certification exam prior to December 2025?
Artificial intelligence (AI) agents are revolutionizing app development. Join us for the Agentforce Virtual Hackathon to push the boundaries of AI-powered agent technology and build groundbreaking solutions on Agentforce for the chance to win a $50,000 Grand Prize. With Agentforce, Salesforce is leading the way in the biggest technological revolution in decades. Agentforce is the […]
Just like that, TDX 2025 has come and gone! Whether you joined us in person or streamed sessions on Salesforce+, we hope you enjoyed all the insightful content designed to help you succeed in your careers and equip you with the knowledge and tools needed on your journey to becoming Agentblazers. With more than 50 […]
When you first create an agent—like a service agent for support queries—it can be tempting to deploy it right away. However, by following a test-driven development (TDD) approach, you write tests first, set clear expectations, and then build your agent to pass these tests. This process, combined with creating a robust test dataset and running […]
