Agentforce brings a new layer of AI into Salesforce by giving teams small helpers that handle routine work, answer questions in real time, and complete multi-step tasks inside the CRM. Each Agentforce agent connects to your Salesforce data, understands context across objects, and takes actions that normally require manual effort. More than 8,000 organisations already use the Salesforce Agentforce platform across sales, service, and marketing. They rely on it to shorten response times, improve data quality, and give users faster ways to move through daily tasks without switching tools.
The idea behind Agentforce is simple. Teams get clear instructions, the AI agents take care of the work, and users stay in control of final decisions.
The article explores how Salesforce Agentforce features work for sales, service, and marketing, what pricing looks like today, practical use cases across industries, and what you need to prepare before turning it on. You will also find practical Salesforce Agentforce use cases and notes from our team, who have already tested the platform in real projects.
What is Agentforce?
Salesforce Agentforce is an AI system built inside the platform that uses Agentforce agents to carry out actions, guide users, and support daily work. It works on top of your CRM data, which means every agent can read the current state of a record, understand related information, and perform tasks without extra integrations.
Agentforce platform uses structured instructions that tell each agent what it can do, what data it can access, and how to respond. These agents can update fields, create records, check information, summarise text, send messages, and support full processes like lead follow-up or case handling.
Salesforce Agentforce connects with the tools teams already use, such as Sales Cloud, Service Cloud, and Marketing Cloud. Agentforce Studio is the area where teams build, adjust, and test their agents. Studio provides controls for actions, rules, data access, and output checks, so every team can keep their AI work safe and predictable.
The platform includes ready-made agents for sales, service, and marketing tasks, and it also supports custom agents for more specific needs. The goal is to remove repeated manual steps and give users help at the moment they need it.
Agentforce for sales
Salesforce Agentforce for sales brings together several AI-driven agents that support every stage of the sales cycle. These Agentforce sales agents help qualify leads, nurture early conversations, manage pipeline hygiene, coach sales reps in real time, and simplify day-to-day tasks inside Salesforce. Each agent works as its own Salesforce user, follows permissions defined by your admin, and uses CRM data to take action or guide the sales team. The result is a sales process that moves faster, stays consistent, and reduces the amount of manual work for reps and managers.
Agentforce Salesforce for sales includes four main areas: lead nurturing, inbound lead generation, sales management, and sales coaching. Each plays a different role in helping teams grow pipeline, progress deals, and hold more effective conversations with prospects.
Lead Nurturing: always on outreach and qualification
The Lead Nurturing agent acts as a dedicated SDR that reaches out to new or updated leads, contacts, and person accounts. It sends a personalised first email based on record data, follows up when needed, answers questions using approved company content, and hands off interested prospects to the right sales rep.
The agent keeps communication active even when your team is offline. It also handles opt-out requests, off-topic replies, and can schedule meetings using the rep's connected calendar.
A single Lead Nurturing agent is included by default, and all outreach happens inside one “Conversation.” You can assign prospects through global rules, Automated Actions, or by selecting the Assign to Agent action on a record.
All emails appear in the Activity Timeline, so sales reps stay fully informed. The agent uses Salesforce features such as Sales Engagement, Einstein Activity Capture, Automated Actions, and Einstein Generative AI to generate context-aware communication.
Lead Nurturing also follows daily email sending limits depending on the connected email provider. It can send up to 1800 messages per day through Gmail or 9800 per day through Microsoft Exchange. The owner field must be populated before assignment, and Activity 360 Reporting must be turned off.
Inbound Lead Generation: capturing prospects on your website or chat
The Inbound Lead Generation agent engages prospects on your website or messaging channels. It answers questions, captures key information, and books meetings automatically.
This provides a 24-hour lead intake channel without manual data entry. The agent can create leads, contacts, or person accounts, depending on how you configure the topic.
Setup involves turning on the feature in Salesforce Go, assigning permissions, creating or selecting the agent user, verifying email settings, and configuring the data library the agent will use for accurate responses.
You can define which fields the agent must collect, what the default Status or Owner will be, and whether one-time password verification is required. After activation, you can connect the agent to Messaging for In-App and Web to make it available to customers.
When configuring inbound lead generation, required fields must be accessible to the agent, and dependent picklists cannot be marked required. If a required lookup field is not visible to the agent user, record creation can fail, so adjustments may be needed in the object setup or permission sets.
You can also expand your Agentforce Service Agent to use the same lead generation actions. This allows service teams to capture new leads during support conversations when customers show interest in your products.
Sales Management and pipeline support
The Sales Management agent helps reps keep opportunities clean and up to date. It reviews recent emails, notes, and call transcripts, then suggests updates to fields such as Stage or Next Step.
With Salesforce Agentforce Pipeline Management enabled, the agent can also make these updates automatically if configured to do so.
Pipeline Management uses data sources such as Notes, Einstein Conversation Insights, and EmailMessage records to understand recent interactions. It runs inside Pipeline Inspection, where reps can review or approve suggestions.
To configure it, admins enable the required features, set data sources, assign permissions, choose whether the agent should suggest or automatically update records, and create a Sales Management agent with the correct topics and actions.
This ensures that opportunities reflect the current state of each deal, without relying on manual updates. It also gives managers more accurate forecasting data and reduces the time reps spend on administrative tasks.
Sales coaching: a dedicated coach for every rep
Agentforce Sales Coach is an AI-driven coach that supports reps with pitch practice, presentation feedback, and realistic role-play sessions. It works on any Salesforce record where the component is added and uses CRM data to evaluate performance.
The coach can analyse what a rep says, reference Opportunity fields, and provide feedback based on the stage of the deal. It includes ready-to-use scenarios for qualification, needs analysis, proposal, and negotiation, each with its own topic and action.
Reps can practice presentations, run through role plays with an AI customer, or refine their negotiation approach. After each session, the coach generates structured feedback that includes a deal summary, strengths, areas for improvement, and next steps.
Sales Coach uses Data 360 to ground responses in company-specific content. You can create a data library with playbooks, objection handling guides, buyer personas, or sales methodology notes. By adding a retriever to the prompt templates, the coach can use this content to produce more accurate and relevant coaching feedback.
Admins can customise topics, actions, and prompt templates to match their selling style. They can define behaviour, rewrite instructions, update stage-based coaching criteria, and build custom scenarios for specialised roles such as account management or customer service.
Testing tools allow admins to preview role-play behaviour and feedback outputs before rolling them out.
All coaching sessions appear as Tasks on the Opportunity Timeline. Managers can track usage by building a Task report filtered by sessions that contain "Agentforce Sales Coach Feedback" in the Subject field.
Why sales teams rely on Agentforce for sales
Salesforce Agentforce for sales helps teams move faster, respond quickly to prospects, and base decisions on accurate CRM data. Lead nurturing increases early pipeline, lead generation provides 24-hour intake, sales management improves data quality, and sales coaching elevates rep performance.
Together, these agents create a unified sales experience that saves time, raises conversion rates, and builds consistent sales habits across the entire team.
Agentforce for service
Agentforce Salesforce for service introduces AI-driven support that handles early case intake, answers routine questions, gathers the details human agents need, and keeps every interaction consistent. The Agentforce Service Agent operates as a Salesforce user with its own permissions and follows topics and actions configured in Agentforce Studio.
Agentforce for service includes three core areas: case handling, inbound assistance through chat or web messaging, and lead generation during service conversations. With the Salesforce Agentforce Service Agent, support teams can cut repeated work, improve accuracy, and keep all interactions consistent.
Planning the use cases
A good Service Assistant relies on clear definitions of which case types it should support. Before setup, review the work your reps handle every day to identify where automation can offer the most value.
Focus on questions like:
- Which tasks are repetitive and predictable?
- Which cases have high volume?
- Where do errors or delays happen?
- Which scenarios require reps to search across many systems?
- Which cases would benefit from standardized handling?
Some teams prioritize simple, high-volume cases such as password resets, billing inquiries, warranty checks, appointment changes, or shipping updates. Others target more complex cases that benefit from structured guidance. The choice depends on the maturity of your processes and the availability of clear instructions for the agent to follow.
Decide whether Service Assistant will serve new reps, experienced reps, or both. New joiners often need detailed steps and more guidance. Experienced agents may prefer concise instructions and the ability to move quickly through the plan.
AI-assisted case handling
The Service Agent can greet customers, gather required details up front, and collect missing information before a human agent joins the conversation. It can answer common questions using CRM data or approved files, classify the issue, and update fields on the case to speed up routing.
The Service Agent uses the instructions defined in its topic to stay within the boundaries your admin sets, ensuring each reply follows your tone, process steps, and escalation path.
When a customer message requires more advanced help, the agent hands the conversation to a human agent with all collected information already added to Salesforce. This reduces back and forth and shortens resolution time.
Guided support inside Salesforce
The Service Agent can also assist human agents inside the console. It can suggest next actions, pull relevant information from case history, and help update the case record based on recent interactions.
These suggestions follow the topics and actions configured for your org, so the support team gets help that reflects your internal process.
Because the agent uses CRM data, the suggestions it provides are grounded in each case’s real history, including fields, related records, and notes.
Inbound support on your website or messaging channels
Agentforce for service works with Messaging for In App and Web, letting customers start conversations on a website or app.
The agent can:
- Answer questions;
- Collect information needed to create or update a case;
- Provide instructions from your documentation or uploaded files;
- Guide customers toward the right help path.
Customers get a 24-hour channel where they can get fast help without waiting for a human agent.
Admins can provide the agent with company-specific knowledge by uploading files to a data library. Data Cloud then indexes this information so the agent can use it to answer questions accurately.
Lead generation from service interactions
Support teams often receive questions from interested buyers. With lead generation enabled, the Service Agent can capture these prospects directly.
The agent can:
- Collect the required data;
- Create leads, contacts, or person accounts;
- Send a meeting link from the assigned rep’s calendar;
- Schedule the call automatically.
Lead generation uses topics designed to capture prospect information and generate the correct Salesforce record. This allows service conversations to pass qualified opportunities to sales without manual effort.
Admins can also configure the Sales Record Creation topic so it creates contacts or person accounts instead of leads, depending on the use case.
Setup and configuration
Configuring Agentforce for service includes enabling the feature, assigning permissions, preparing the agent user, setting topic instructions, and connecting channels.
Key steps include:
- Turning on the Service Agent;
- Creating the agent user with the correct license;
- Assigning permissions for managers and support reps;
- Uploading files to a data library if you want grounded answers;
- Adjusting topics and actions to reflect your process;
- Activating the agent in Agentforce Studio;
- Connecting the agent to a website or messaging channel
Once activated, the agent begins working immediately when a customer initiates a conversation.
Operational considerations
When enabling service features, several technical details matter:
- Required fields for case creation must be accessible to the agent;
- Dependent picklists cannot be marked required;
- Lookup fields must be visible to the agent user;
- Data libraries may take up to 24 hours to appear in guided setup;
- Meeting links require Inbox access for the assigned rep;
- Only one email address can be connected to the agent user;
- Messaging for In-App and Web may take a few minutes to sync after channel creation.
These checks help ensure smooth operation and prevent case or record creation errors.
How Agentforce for service strengthens the operations
Agentforce Salesforce for service removes many of the repeated steps support teams handle every day.
The Service Agent takes care of intake, routine questions, and classification while guiding human agents with accurate suggestions. It improves response time, increases consistency, reduces manual typing, and captures more valuable data on every interaction.
With lead creation added, the agent also supports revenue generation by handing off sales-ready prospects who reach out to the support team.
Agentforce for marketing
Salesforce Agentforce for marketing turns campaigns into conversational experiences that respond to each customer in real time. With Agentforce AI, marketers can plan, create, personalise, and optimise campaigns across email, mobile, ads, web, and messaging. The goal is to reduce production time, increase relevance, and give marketers more room to focus on strategy rather than manual execution.
Agentforce sits on top of a unified system shaped by three elements: data, engagement, and intelligence. Together, they allow marketing teams to connect customer information, activate it instantly, and guide customers along journeys that change based on their behavior and intent.
A unified foundation for customer data
Agentforce Salesforce for marketing relies on Data 360 (Data Cloud) to bring customer information into a single, trusted profile. Data 360 connects Marketing Cloud Engagement, Account Engagement, and external data sources through zero-copy integrations. There is no need for migrations, and the current org structures stay intact.
Once connected, marketers have real-time access to harmonized data that can be used for segmentation, personalization, and analytics without writing SQL. Unified profiles help the agent understand customer intent and update experiences as customers browse, click, reply, or convert. This strengthens every part of the journey and removes the delays caused by scattered data.
Conversational journeys across channels
Salesforce Agentforce replaces one-way messages with two-way interactions. Customers can ask questions, reply to emails, respond to SMS, start a conversation on WhatsApp, or engage with a website that adapts in real time. The agent answers based on your Data 360 profiles, your brand voice, and any insights available across your marketing, sales, service, or commerce systems.
Salesforce demonstrated this approach in its Road to Dreamforce campaign, where one email led to thousands of real-time conversations. Instead of clicking a landing page, recipients replied directly, and the agent handled bookings and questions with context from Data Cloud. Two-way email becomes available to all customers in February 2026, and two-way SMS and WhatsApp are available today.
Marketing becomes a set of conversations that update as customers move between channels.
Adaptive sites and journeys that react to intent
Agentforce Salesforce helps teams create experiences that adjust to each visitor. With Personalization and Data 360, websites can shift content, product recommendations, and messages as customers browse. These dynamic updates can also trigger actions such as scheduling visits, confirming appointments, or handing off to service with full context.
Brands such as Fisher and Paykel have already shown the impact. Their adaptive site reacts to customer behavior and intent, and this approach increased order conversion by 33%. Adaptive websites become available to all Salesforce customers in December 2025.
Campaign creation at the speed of ideas
Salesforce Agentforce reduces the time spent on journey planning, brief creation, content production, and approvals. In Marketing Cloud Next, agents generate briefs, draft campaigns, recommend audiences, and create content based on your instructions.
Inside the Agentforce chat window, marketers describe the goal of the campaign. The agent then drafts a full brief with key messages, audience descriptions, and campaign goals. After confirming the brief, the agent can generate a preview that includes a subject line, preheader, and paragraph copy. It can also draft a landing page and SMS message. You can iterate with follow-up prompts or adjust tone, length, and focus.
The agent actions that support this workflow, including Create a Campaign from a Brief and Draft a Campaign Brief, are powered by Flow. Admins can adjust these flows to meet business needs, add logic, or connect to other systems.
In the content builder, marketers can use the AI features to revise text, change tone, create sections, or generate full layouts. Content is grounded in the confirmed brief, the assigned brand, and the components already on the canvas.
The Campaign Designer for guided creation
Marketing Cloud Next Advanced Edition includes the Campaign Designer. Clicking Draft with AI opens a guided builder that walks marketers through creating a full campaign. The designer starts with a brief, then recommends a campaign flow with suggested channels and messages. Teams can revise text, adjust steps, set schedules, and add segments. The flow is based on the brief and, when selected, a reference campaign.
The entire experience shortens planning cycles and reduces the manual work behind journey creation.
Continuous optimization from real-time insights
Agentforce Salesforce helps teams monitor campaigns while they are running. Optimization agents track performance, spend, engagement, and creative outcomes. They surface insights in Slack so teams can act quickly. Paid Media Optimization can pause underperforming ads and shift spend to higher-performing audiences. Data 360 Clean Rooms allow brands and partners to improve targeting without copying data.
Unified Analytics, now available, and Multi-Touch Attribution, arriving in February, help marketers measure performance across channels and understand what drives results.
Agentforce inside Marketing Cloud Next
Salesforce Agentforce agents in Marketing Cloud Next streamline the entire planning and execution cycle. Marketers can:
- Open the Agentforce chat window and describe a campaign idea;
- Generate a full brief and adjust key messages or audiences;
- Produce multi-channel content;
- Refine content inside the editor;
- Create full email or landing page sections;
- Summarize performance based on campaign activity
If a brief is removed from a campaign, the generated segments and content remain, but the agent stops supporting that specific campaign. This design keeps teams flexible and allows manual changes without losing work.
Salesforce Agentforce pricing
Salesforce gives companies several ways to use Agentforce AI depending on their workload and teams. You can pay for actions, full conversations, or choose a user licence for unlimited internal usage. All activity is tracked in the Digital Wallet, so you can monitor costs as adoption grows.
Agentforce pricing overview
There are three primary models you can work with:
- Flex Credits for action-based usage.
- Conversation-based pricing for customer sessions.
- User-based licences for unlimited use inside the company.
Each model supports a different style of work, so teams can choose what suits them best.
Agentforce pricing model comparison
Salesforce Agentforce cost structure and inclusions
Each model includes different features and cost patterns:
Prices are stated as for December 2025. Check with official Salesforce resources for updates.
Agentforce buying options
Agentforce supports several payment paths:
- Pre-purchase allows you to pay upfront and draw down usage over the full contract.
- Pay-as-you-go gives you a simple way to start without any initial commitment.
- Pre-commit lets you lock in a baseline of usage and pay monthly while still getting better value. This option becomes widely available later this year.
These options support different Salesforce Agentforce use cases, from customer chat to internal automation.
When to choose each model
Flex Credits work best for teams with many small actions, such as service reps, back office teams, and marketing teams that generate repeated tasks. Conversation pricing fits predictable volumes of customer chat. User-based licensing helps large sales and service teams that interact with Agentforce throughout the day and need full flexibility.
Conclusion
Agentforce gives teams a faster and clearer way to work by supporting them at the exact moment they need help. It guides sales reps through real conversations, helps service teams solve cases with fewer mistakes, and gives marketers tools to plan and build campaigns in minutes. When all parts of the system work together, teams respond quickly, maintain higher accuracy, and improve the full client experience.
Our company has the skills to help clients get real value from Agentforce Salesforce. We are a Summit-tier partner with more than 140 delivered Salesforce projects, over 400 certifications, and 8 certified architects. Our team also holds 11 Agentforce Specialist certificates, which shows our focus on the new AI tools. We have tested Agentforce in our own products and in client projects, so we know how to set up topics, prompts, flows, data grounding, and testing in a way that brings steady results.
If you want support with planning, setup, or improving Agentforce, we can guide the full process. We help you define use cases, prepare clean data, build strong agent instructions, test all scenarios, and connect Agentforce to the rest of your system. The goal is to help you adopt AI in a way that is safe, simple, and aligned with how your team works.
FAQs about Agentforce
What is Salesforce Agentforce AI and how does it work inside Salesforce?
Salesforce Agentforce AI is an intelligent layer inside the salesforce agentforce platform that uses small helpers called agentforce agents to complete work inside the CRM. These agents read live Salesforce data, understand the context of a record, and take action based on clear instructions set by admins. They do not guess. They follow rules, topics, and permissions, which keeps their work safe and aligned with your processes.
Agents can update fields, create records, answer questions, summarise long text, prepare next steps for users, interact with customers, or guide employees through their tasks. Because Agentforce AI is built inside Salesforce, teams do not need any extra tools, custom integrations, or migrations. All work happens in the same interface your team already uses.
What can the Agentforce Sales Agent and Agentforce Service Agent do for my team?
The Agentforce Sales Agent helps sales teams keep their pipeline clean, nurture prospects, qualify inbound leads, and prepare for real conversations through pitch practice and role-play coaching. It checks emails, notes, and call transcripts to suggest updates that make records more accurate. It can also send personalised emails, collect meeting requests, and route prospects to the right rep.
The Salesforce Agentforce Service Agent supports customer service teams by managing early case intake, answering common questions, completing simple steps, and gathering details before a human agent joins the conversation. It can read case fields, follow your support policies, and update the case automatically. When a situation needs a human touch, it hands off the conversation with all the gathered information already added to Salesforce.
Both agents operate as their own Salesforce users, follow permissions you define, and stay inside your CRM processes. They reduce manual work, increase accuracy, and help teams respond faster.
How does Agentforce Marketing improve campaigns and customer journeys?
Agentforce marketing gives marketers a way to build, personalise, and optimise campaigns without long production cycles. Inside Marketing Cloud Next, marketers can describe a campaign idea in the Agentforce chat window, and the agentforce platform creates a full campaign brief with audience details, key messages, and goals. It then drafts emails, landing pages, SMS messages, and content sections based on that brief.
Because it uses Data 360, Agentforce understands customer profiles in real time and adjusts journeys based on behaviour and intent. It can support two-way email, SMS, WhatsApp, and adaptive websites that react to customer actions.
It also monitors performance while campaigns are running and surfaces insights in Slack so teams can adjust spend, pause weak ads, or shift focus to stronger segments.
What are the core Salesforce Agentforce features I should know about?
The most important Salesforce Agentforce features include:
- Agents built on topics, actions, and prompt templates;
- Data grounding through Data Cloud;
- Support for sales, service, and marketing inside one system;
- Real-time guidance and record updates;
- Two-way email, SMS, and web chat responses;
- Sales coaching with role-play and pitch feedback;
- Guided case plans for service teams;
- Campaign creation powered by AI in Marketing Cloud Next;
- Optimisation tools that analyse performance and suggest changes.
Together, they create a single environment where teams can work faster with fewer mistakes and better access to customer data.
What makes the Agentforce platform different from other AI tools?
The Agentforce platform works inside Salesforce rather than outside it. This means every Agentforce agent uses the same security settings, field-level permissions, record access, and data your team already has. It also means you control how the agent behaves. You decide which tasks it can perform, which data it can use, how it should respond, and when it should hand work to a human.
Because Agentforce uses your CRM records as the source of truth, the results are more accurate and grounded. You can also customize every part of it, from coaching style to case handling rules. This level of control is why sales, service, and marketing teams trust the salesforce agentforce platform to handle real work.
.webp)
together