Dreamforce 2026 Highlights: Biggest Salesforce Announcements & Key Takeaways

Publish date:

• 21st September, 2026

Dreamforce 2026 marked a major shift in Salesforce’s strategy as the company moved beyond adding AI features inside CRM and focused on making Salesforce data, workflows, business logic, and AI agents accessible across the tools people already use.

The biggest announcement was AIforce, Salesforce’s new live interface layer designed to bring trusted Salesforce context and actions into AI experiences such as Claude, Slack, and Agentforce Coworker. Salesforce also expanded Claudeforce, introduced new Slackforce experiences, unveiled its CRM-focused reasoning model Koa, and announced new Agentforce capabilities designed to handle increasingly complex business workflows.

For Salesforce customers, these announcements point to a different way of using CRM: instead of employees constantly opening Salesforce and manually navigating records, AI agents and intelligent interfaces can increasingly retrieve information, reason across business context, and take approved actions from wherever teams already work.

This Dreamforce 2026 recap covers the most important announcements, how they work, and what businesses should consider before adopting them.

Dreamforce 2026 in 60 Seconds

Dreamforce 2026 focused on turning Salesforce from a traditional CRM interface into an AI-powered enterprise platform that can operate across multiple work environments.

The key announcements included:

• AIforce — brings Salesforce data, workflows, permissions, business logic, and agents into external AI interfaces.

• Claudeforce — connects Salesforce with Claude and includes 37 prebuilt sales skills for activities such as prospecting and pipeline management.

• Slackforce — brings Salesforce context and actions directly into Slack conversations and workflows.

• Koa — Salesforce’s CRM-focused reasoning model designed for enterprise tasks across areas such as sales, service, and marketing.

• Agentforce expansion — introduces more job-specific AI agents and capabilities for handling longer and more complex business processes.

• Enterprise AI governance — Salesforce continued emphasizing permissions, security, governance, and zero-data-retention principles as AI becomes more deeply embedded in business operations.

The larger message from Dreamforce 2026 was clear: Salesforce wants CRM data and business processes to become available wherever employees and AI agents work, rather than remaining limited to the traditional Salesforce interface.

The Big Theme of Dreamforce 2026: AI Becomes the Interface

The central idea behind Dreamforce 2026 was a significant change in how people interact with Salesforce.

For years, using CRM meant opening Salesforce, navigating dashboards, searching records, updating fields, and moving between different applications. Salesforce is now pushing toward a model where users can simply ask for information or request an action from the AI environment they are already working in.

This direction was summed up at Dreamforce: AI is becoming the new interface for enterprise software.

From CRM Screens to AI-Powered Work

Consider a sales representative preparing for a customer meeting.

Traditionally, they might need to open Salesforce, review the account, check opportunities, read previous activities, look at support history, and then manually compile the information they need.

The emerging model is different.

A user could ask an AI assistant to:

• summarize an account before a meeting;

• identify open opportunities and potential risks;

• retrieve relevant customer history;

• recommend appropriate next actions; and

• perform permitted Salesforce actions without requiring the user to navigate multiple CRM screens.

This is where AIforce becomes important.

Rather than functioning as another standalone AI assistant, AIforce is designed as a live interface layer that makes Salesforce data, metadata, business logic, permissions, workflows, and agents available to supported AI experiences.

That means Salesforce's value could increasingly extend beyond the application itself.

Why This Matters for Salesforce Customers

For businesses already using Salesforce, the shift is bigger than simply adding another AI feature.

It could change how organizations approach:

• User adoption: Employees may be able to interact with CRM information through more natural conversational experiences.

• Productivity: Routine research, record retrieval, summaries, and approved actions can potentially require fewer manual steps.

• Data quality: AI agents depend heavily on accurate, structured, and accessible CRM data. Poor data foundations can limit the value businesses receive from these capabilities.

• Governance: As AI gains access to business data and actions, permissions, security policies, auditability, and AI governance become increasingly important.

• Salesforce architecture: Organizations will need to think beyond individual clouds and consider how their data, integrations, automation, and AI agents work together.

For Salesforce leaders, the important question after Dreamforce 2026 is therefore not simply “Which new AI feature should we enable?”

It is:

“Is our Salesforce environment ready for AI agents to understand our data and safely participate in real business processes?”

That question underpins nearly every major announcement from Dreamforce 2026.

AIforce: The Biggest Announcement at Dreamforce 2026

AIforce is Salesforce’s new live interface layer that connects Salesforce data, metadata, business logic, permissions, workflows, and AI agents with the AI tools employees use to get work done.

Instead of requiring users to constantly switch back to Salesforce to find information or complete routine tasks, AIforce makes relevant Salesforce context and approved actions available directly within AI-powered work environments.

This was one of the most significant announcements at Dreamforce 2026 because it changes where users can interact with Salesforce.

How Does AIforce Work?

AIforce sits between Salesforce and supported AI experiences, providing the business context an AI assistant needs to understand a request and determine what it can do.

In practical terms, this can bring together:

• Salesforce data — customer, account, opportunity, service, and other CRM information.

• Metadata and business logic — the structure and rules that explain how an organization's Salesforce environment works.

• Permissions and governance — controls designed to ensure users and AI systems only access information and actions they are authorized to use.

• Workflows and actions — existing business processes that can be surfaced or executed through AI-powered experiences.

• AI agents — specialized agents capable of assisting with specific business tasks.

The result is an experience where users can increasingly work through natural-language requests instead of manually navigating multiple Salesforce screens.

What Could AIforce Look Like in a Real Business?

Imagine a sales manager asks:

“Which enterprise opportunities are at risk this quarter, and what should my team focus on this week?”

Instead of manually reviewing opportunities, activities, account history, and pipeline reports, an AI experience connected through AIforce could use authorized Salesforce context to help surface relevant opportunities and explain the factors requiring attention.

A service team could similarly ask for a summary of a customer's recent cases before responding to an escalation.

The key distinction is that the AI experience doesn't work from a generic prompt alone. Its usefulness depends on the trusted business context, permissions, workflows, and data available within the organization's Salesforce environment.

AIforce vs. Agentforce: What Is the Difference?

The names are similar, but their roles are different.

AIforce	Agentforce
Provides an interface layer between Salesforce and AI experiences	Provides AI agents designed to perform business tasks
Makes Salesforce context available to supported AI interfaces	Uses business context to reason and take permitted actions
Focuses on how users and AI environments access Salesforce capabilities	Focuses on what specialized AI agents can accomplish
Can support experiences involving tools such as Claude, Slack, and Agentforce Coworker	Supports autonomous and assistive agent-based workflows

A simple way to understand the relationship is:

AIforce helps make Salesforce accessible to AI experiences, while Agentforce provides agents that can perform specific work using enterprise context and approved actions.

They are better understood as complementary parts of Salesforce's broader agentic AI strategy than as competing products.

Why AIforce Matters for Existing Salesforce Customers

AIforce could make existing Salesforce investments more accessible, but organizations will not automatically be ready simply because the technology becomes available.

Businesses considering these capabilities should first evaluate:

• CRM data quality and completeness

• user roles and Salesforce permissions

• integration architecture

• existing Flow and automation

• duplicate and fragmented customer data

• governance requirements

• processes suitable for AI-assisted or agent-driven execution

This creates an important implementation consideration.

The quality of an AI experience will depend heavily on the quality of the Salesforce environment behind it.

Organizations with fragmented data, unnecessary customizations, inconsistent permissions, or poorly designed workflows may need to strengthen their Salesforce foundation before introducing AI agents into critical business processes.

For companies planning their next phase of Salesforce AI adoption, Dreamforce 2026 therefore shifts the conversation from simply experimenting with generative AI to preparing CRM architecture for secure, context-aware, agent-driven work.

Claudeforce: Bringing Salesforce Context Directly Into Claude

Another major Dreamforce 2026 announcement was Claudeforce, the deeper integration between Salesforce and Anthropic's Claude.

Claudeforce lets employees work with Salesforce information and business processes directly through Claude. Instead of moving repeatedly between an AI assistant and CRM screens, users can bring relevant Salesforce context into a conversational workflow and complete supported tasks from there.

Salesforce announced 37 prebuilt sales skills for Claude, covering common activities across prospecting, account research, pipeline management, and sales execution.

What Can Claude in Salesforce Help Sales Teams Do?

The practical value becomes clearer when looking at everyday sales workflows.

A sales representative could use Claude with Salesforce context to:

• research an account before outreach;

• summarize previous customer interactions;

• prepare for an upcoming sales meeting;

• identify relevant opportunities and pipeline changes;

• draft personalized follow-up communication;

• analyze account or opportunity information; and

• determine appropriate next steps based on available CRM context.

Instead of manually collecting information from several Salesforce records before asking an AI assistant for help, the goal is to make that enterprise context available within the AI experience itself.

A Simple Claudeforce Example

Imagine an account executive has a customer meeting in 20 minutes.

They could ask:

“Prepare me for my meeting with this account. Summarize recent interactions, active opportunities, key stakeholders, outstanding issues, and the next actions I should consider.”

Claude could then use the Salesforce context available to that user, rather than relying only on information manually copied into a prompt.

This turns AI from a separate productivity tool into a more context-aware assistant connected to actual CRM workflows.

Why Claudeforce Matters

The bigger change is not simply that Salesforce can connect with another AI model.

It represents Salesforce's move toward a multi-model enterprise AI ecosystem, where businesses can use different AI experiences while maintaining access to governed Salesforce context.

For organizations already exploring Claude for enterprise work, this could reduce one of the biggest barriers to practical AI adoption: the gap between an AI model and the business data employees need to do their jobs.

However, the same readiness requirements still apply. Reliable CRM data, appropriate permissions, clearly defined processes, and strong governance become even more important when enterprise AI can interact with customer and pipeline information.

Claudeforce therefore highlights one of the broader lessons from Dreamforce 2026:

The future of Salesforce may be less about asking employees to work inside one interface and more about securely bringing Salesforce intelligence into the tools where work already happens.

Slackforce: Bringing Salesforce Work Into Slack

Dreamforce 2026 also reinforced Salesforce's vision of making Slack a conversational workspace for CRM-powered work.

With Slackforce, Salesforce is bringing more customer context, AI assistance, and business actions into the environment where teams already communicate. The idea is straightforward: employees shouldn't always have to leave a conversation, open Salesforce, locate a record, and then return to Slack to keep working.

Instead, relevant Salesforce context can become part of the conversation itself.

What Could Teams Do With Slackforce?

Depending on the connected Salesforce environment and user permissions, teams could use Slack-based AI experiences to help:

• surface relevant account and opportunity information;

• summarize customer or deal activity;

• prepare teams for upcoming customer meetings;

• collaborate around sales and service issues;

• identify tasks or follow-up actions;

• access AI agents within existing workflows; and

• take supported Salesforce actions without repeatedly switching applications.

What Does This Look Like in Practice?

Consider a sales team discussing a large opportunity in a Slack channel.

Instead of someone manually opening Salesforce and reporting the latest information, the team could request relevant CRM context directly within the conversation.

For example:

“Summarize the latest activity on this opportunity, highlight open issues, and show the next scheduled customer interaction.”

This can turn Slack conversations into more context-rich workflows, keeping collaboration and CRM information closer together.

The same approach could extend beyond sales.

A customer service team investigating an escalation could access relevant case context, while managers could use AI assistance to understand important changes without manually reviewing multiple records and dashboards.

Why Slackforce Matters for Businesses

The potential benefit is less about adding another Salesforce feature and more about reducing context switching.

When employees constantly move between CRM records, communication tools, dashboards, documents, and AI assistants, even simple processes can involve unnecessary steps.

Bringing Salesforce context closer to Slack could help organizations create workflows where conversation, enterprise data, AI assistance, and business actions happen within a more connected experience.

However, organizations will still need to define carefully what information users and AI agents can access and which actions they can perform.

As Salesforce becomes more deeply embedded in conversational work, permission design, data governance, workflow architecture, and AI security become part of the implementation strategy—not an afterthought.

Meet Koa: Salesforce’s New CRM Reasoning Model

One of the most technically significant announcements from Dreamforce 2026 was Koa, Salesforce’s new reasoning model built for CRM and enterprise workflows.

General-purpose AI models are designed to handle a broad range of questions and tasks. Koa takes a more specialized approach, focusing on the types of reasoning required across customer relationships, sales processes, service interactions, marketing activities, and other enterprise workflows.

The goal is not simply to generate better text. It is to help AI understand business context and reason through CRM-related tasks more effectively.

What Makes Koa Different?

Salesforce environments contain more than customer names and contact details. They include relationships between accounts, opportunities, activities, cases, products, workflows, permissions, and business rules.

An AI system operating in this environment needs to understand how those elements relate.

For example, a sales manager may ask:

“Which opportunities need attention before the end of the quarter, and why?”

Answering that effectively may require reasoning across deal stages, recent activities, customer engagement, pipeline history, next steps, and other relevant CRM signals rather than simply retrieving a single field.

This type of enterprise reasoning is the problem Koa is intended to address.

Where Could Koa Be Used?

Potential applications span different Salesforce functions.

• Sales teams could use reasoning capabilities to analyze opportunities, understand account context, identify risks, and determine appropriate next actions.

• Service teams could use customer and case context to help understand complex support situations and determine possible resolutions.

• Marketing teams could potentially use richer customer context when analyzing engagement and coordinating journeys.

• Business leaders could interact with CRM information through natural-language questions rather than relying exclusively on predefined reports and dashboards.

Why Koa Matters for Salesforce's AI Strategy

Koa also reveals something important about Salesforce's broader direction.

AIforce provides a way for AI experiences to access Salesforce context. Agentforce provides agents that can carry out business tasks. Koa adds a reasoning layer designed around CRM and enterprise work.

Together, these technologies point toward Salesforce environments where AI can increasingly:

understand business context → reason about what is happening → recommend an appropriate response → execute an approved action.

For businesses, however, stronger reasoning models do not eliminate the need for a strong Salesforce foundation.

AI still depends on the quality of the information and processes it has access to. Duplicate records, incomplete customer data, inconsistent fields, poorly designed automation, or overly complex customizations can make it harder for AI systems to produce useful outcomes.

That makes data readiness, Salesforce architecture, governance, and process design increasingly important as organizations move from basic generative AI experiments toward more sophisticated agent-driven CRM workflows.

Agentforce Gets More Specialized: From AI Assistants to Digital Workers

Dreamforce 2026 showed the next stage of Salesforce’s Agentforce strategy: AI agents are moving beyond answering questions and assisting with isolated tasks toward handling more specialized, multi-step business processes.

The direction is toward agents that can understand enterprise context, reason through a task, use approved tools and data, and take actions within defined business boundaries.

This makes Agentforce increasingly relevant to organizations looking to automate work across sales, customer service, marketing, commerce, and other Salesforce-powered operations.

What Changed With Agentforce at Dreamforce 2026?

The emphasis is shifting from simply creating an AI agent to giving agents the context, skills, tools, and governance required to perform useful business work.

That includes agents designed around specific roles and workflows rather than one generic assistant attempting to handle every request.

In practice, an Agentforce-powered workflow could involve an agent that:

• understands a customer request;

• retrieves relevant CRM context;

• evaluates the next appropriate step;

• works across connected business processes;

• performs authorized actions; and

• escalates situations that require human judgment.

The result is a model where AI can participate in a workflow rather than only provide information about it.

What Could This Look Like in a Real Salesforce Environment?

Consider a customer service request involving an order issue.

A basic AI assistant might summarize the customer's message or suggest a response.

A more capable agent-driven workflow could potentially identify the customer, retrieve relevant account and case information, check connected order data, determine the appropriate process, perform permitted actions, update Salesforce, and involve a service representative when necessary.

The distinction is important:

AI assistance helps an employee complete a task. Agentic AI can potentially complete defined parts of the workflow on the employee's behalf.

Human Oversight Still Matters

More autonomy also introduces more responsibility.

Businesses should decide which decisions an AI agent can make independently, which actions require approval, and when to transfer a workflow to a human.

For example:

Workflow	Possible Agent Role	Human Involvement
Lead qualification	Analyze and prioritize incoming leads	Review strategic or high-value opportunities
Meeting preparation	Compile account and opportunity context	Sales rep determines meeting strategy
Customer service	Handle defined routine requests	Escalate sensitive or complex cases
CRM updates	Complete approved record actions	Review exceptions or high-impact changes
Pipeline monitoring	Surface changes and potential risks	Manager makes commercial decisions

The strongest Agentforce implementations are therefore unlikely to be those that automate everything.

They will clearly define where agents add value, what data they can access, which actions they can take, and where people remain responsible for decisions.

What Businesses Should Prepare Before Expanding Agentforce

Before deploying AI agents across critical workflows, organizations should evaluate their Salesforce environment across four areas:

• Data: Is CRM information accurate, complete, and accessible?

• Processes: Are the workflows being automated clearly defined?

• Permissions: Does each agent have only the access required to perform its role?

• Governance: Are approvals, monitoring, escalation paths, and human intervention clearly established?

Dreamforce 2026 makes the direction increasingly clear: Salesforce is building toward an environment where people and AI agents work together across CRM processes.

For businesses, the challenge is no longer simply deciding whether to experiment with AI. It is deciding which workflows are ready for agentic automation and how those agents should operate safely within the organization.

Enterprise AI Harness: Governing AI at Enterprise Scale

As AI agents gain access to customer data and business workflows, one question becomes increasingly important:

How can enterprises give AI more capability without losing control over data, permissions, security, and business processes?

Dreamforce 2026 addressed this challenge through Salesforce’s broader Enterprise AI Harness approach—providing the controls and enterprise context needed to manage AI systems as they interact with Salesforce data and workflows.

This matters because deploying AI in an enterprise environment differs greatly from using a standalone chatbot.

An AI agent may potentially interact with customer records, opportunities, service cases, internal knowledge, workflows, and connected systems. Organizations therefore need clear controls around what an agent can see, what it can do, and when human approval is required.

What Does Enterprise AI Governance Need to Cover?

For businesses adopting AIforce, Agentforce, or other connected AI experiences, governance should address several areas:

• Identity and permissions: AI interactions should respect access controls for users, agents, and business roles.

• Data access: Organizations need clear policies governing which customer and enterprise information AI systems can use.

• Approved actions: Agents should operate within defined boundaries rather than having unrestricted authority across Salesforce.

• Monitoring and auditability: Businesses need visibility into how AI-supported workflows operate and what actions are performed.

• Human escalation: Sensitive, unusual, or high-impact decisions should have defined paths for human review.

• AI model governance: Enterprises using multiple models need policies for deciding which models and AI services can access specific business contexts.

Why Governance Becomes More Important With Agentic AI

Traditional generative AI primarily produces an output for a person to review.

Agentic AI can go further by participating in business processes and, where configured and permitted, taking actions.

That increases the importance of controls.

For example, there is a significant difference between an AI assistant that recommends updating an opportunity and an agent authorized to modify the Salesforce record.

The more operational responsibility an organization gives to AI agents, the more carefully it needs to define permissions, approval rules, monitoring, and exception handling.

AI Readiness Is Becoming Part of Salesforce Architecture

For Salesforce customers, AI governance should therefore be considered during implementation—not after agents have already been deployed.

Before scaling agentic workflows, businesses should be able to answer questions such as:

• What Salesforce data can each agent access?

• Which actions can an agent perform independently?

• Which actions require human approval?

• What happens when an agent encounters an exception?

• How are agent activities monitored?

• Which business processes should never be fully automated?

• Who owns the governance of AI agents after deployment?

This is one of the less flashy but more important lessons from Dreamforce 2026.

Enterprise AI success will depend not only on how intelligent an agent is, but also on how securely and responsibly it connects that intelligence to business data and processes.

Dreamforce 2026 Announcements at a Glance

Dreamforce 2026 introduced several connected technologies rather than one standalone AI product. Here is a quick comparison of the major announcements and where they fit within Salesforce’s evolving AI ecosystem.

Announcement	What It Is	Primary Purpose	What It Means for Businesses
AIforce	A live interface layer connecting Salesforce context with AI experiences	Make Salesforce data, logic, permissions, workflows, and agents accessible through AI interfaces	Employees can potentially interact with Salesforce without relying exclusively on traditional CRM screens
Claudeforce	Deeper integration between Salesforce and Anthropic’s Claude	Bring Salesforce context and business capabilities into Claude workflows	Sales and other teams can work with relevant CRM context while using Claude
Slackforce	Salesforce-powered work experiences within Slack	Bring CRM context, collaboration, AI assistance, and supported actions closer together	Teams can reduce application switching and work with Salesforce information inside collaborative workflows
Koa	Salesforce’s CRM-focused reasoning model	Reason across enterprise and customer context	AI systems can become better suited to understanding complex CRM-related tasks
Agentforce	Salesforce’s platform for deploying AI agents across business workflows	Allow specialized agents to assist with or execute defined tasks	Businesses can automate parts of sales, service, marketing, and other processes while maintaining defined controls
Enterprise AI Harness	Salesforce’s approach to connecting and governing enterprise AI	Provide security, permissions, governance, and enterprise context for AI systems	Organizations can establish stronger controls as AI agents gain access to business data and actions

How Do These Dreamforce 2026 Announcements Fit Together?

The easiest way to understand Salesforce’s direction is to look at these technologies as different parts of the same enterprise AI architecture:

AIforce provides access to Salesforce context → Koa helps reason over CRM-related information → Agentforce provides specialized agents → Claudeforce and Slackforce bring those capabilities closer to the interfaces where people work → enterprise governance helps control how AI accesses data and performs actions.

Not every Salesforce customer will need every capability.

A sales organization may prioritize Claude-powered workflows and Agentforce, while a service-heavy business may focus on agents connected to customer cases and knowledge. A company with complex Salesforce architecture may first need to improve data quality, permissions, integrations, and automation before expanding AI adoption.

The practical takeaway from Dreamforce 2026 is therefore not to adopt every new AI capability at once.

Businesses should identify high-value workflows first, assess whether their Salesforce environment is ready, and then introduce AI where it can deliver measurable operational benefit.

What Dreamforce 2026 Means for Existing Salesforce Customers

The announcements from Dreamforce 2026 create significant opportunities, but adopting every new AI capability immediately is unlikely to be the right strategy for every Salesforce organization.

The better starting point is to ask:

Where can AI remove measurable friction from our current Salesforce processes?

The answer will depend heavily on the maturity of your existing CRM environment.

If Your Salesforce Org Is Already Well Optimized

Organizations with clean CRM data, clearly defined processes, strong integrations, and well-managed permissions are likely to be better positioned to explore AI-driven workflows.

Potential next steps include identifying use cases where Agentforce or connected AI experiences could reduce repetitive work, speed up customer response times, improve account preparation, or help employees with complex workflows.

Focus on measurable business outcomes rather than deploying AI simply because the capability exists.

If Your Salesforce Data Is Fragmented or Unreliable

AI does not automatically fix poor CRM foundations.

Duplicate accounts, missing fields, outdated contacts, inconsistent data models, and disconnected systems can reduce the quality of the context available to AI agents.

Before scaling Salesforce AI, businesses in this situation should prioritize:

• CRM data cleanup and deduplication;

• data model assessment;

• integration review;

• permission auditing; and

• identification of authoritative data sources.

Improving the underlying Salesforce environment can often be a more valuable first step than immediately deploying another AI capability.

If Your Salesforce Org Has Heavy Customization

Organizations that have used Salesforce for several years may have accumulated custom objects, Apex, Flows, integrations, validation rules, and legacy automation.

Before introducing agentic workflows, assess which customizations remain necessary and how existing automation could interact with AI-driven actions.

This can reduce the risk of adding AI on top of unnecessary technical complexity.

If You Are Considering Agentforce

Do not begin with the question:

“Where can we deploy an AI agent?”

Start with:

“Which business process is expensive, repetitive, slow, or difficult to scale?”

Then evaluate whether an agent is appropriate for that workflow.

A practical Agentforce readiness assessment should examine:

Area	Key Question
Business process	Is the workflow clearly defined and worth improving?
Data	Does Salesforce contain reliable information for the agent to use?
Integration	Does the workflow depend on systems outside Salesforce?
Permissions	What information and actions should the agent be allowed to access?
Human oversight	When should the agent stop and involve an employee?
Success metrics	How will the organization measure business impact?

If You Are Unsure Where to Start

For many organizations, the immediate opportunity after Dreamforce 2026 is not a full AI transformation.

It is creating a Salesforce AI roadmap.

This can involve reviewing the current Salesforce architecture, identifying high-value AI use cases, assessing Agentforce readiness, strengthening data and integrations, and prioritizing initiatives according to expected business value and implementation complexity.

Working with an experienced Salesforce consulting partner can also help organizations determine where technologies such as Agentforce and emerging AI capabilities fit within their existing Salesforce roadmap—particularly when the environment involves multiple clouds, integrations, custom automation, or complex governance requirements.

The objective should remain simple:

Use AI where it solves a real business problem, prove the value, and then scale what works.

5 Practical Salesforce AI Use Cases Inspired by Dreamforce 2026

The technologies introduced at Dreamforce 2026 become more meaningful when connected to real business processes. For Salesforce customers, the strongest opportunities are likely to be workflows where employees spend significant time finding information, completing repetitive tasks, or moving between systems.

Here are five areas businesses can evaluate first.

1. AI-Powered Sales Meeting Preparation

• The problem: Sales representatives often spend time reviewing accounts, opportunities, previous activities, emails, and customer history before an important meeting.

• Potential AI workflow: An AI assistant could use authorized Salesforce context to prepare an account brief covering recent interactions, active opportunities, key stakeholders, open issues, and relevant next steps.

• Business outcome to measure: Preparation time per meeting, seller productivity, and CRM usage.

2. Agent-Assisted Customer Service

• The problem: Service representatives may need to search cases, knowledge articles, customer history, and connected systems before responding to a customer.

• Potential AI workflow: Agentforce could help interpret the request, retrieve relevant customer context, recommend or execute permitted actions, and escalate exceptions to a service representative.

• Business outcome to measure: Average handle time, resolution time, escalation rate, and case deflection.

3. Pipeline and Opportunity Risk Analysis

• The problem: Sales managers often rely on dashboards and manual reviews to understand which opportunities need attention.

• Potential AI workflow: CRM-focused reasoning could analyze available opportunity context and surface deals that warrant review, along with the signals that drove the recommendation.

A manager could ask:

“Which opportunities require attention this week, and what changed since our last pipeline review?”

Business outcome to measure: Time spent on pipeline reviews, forecast process efficiency, and follow-up completion.

4. Conversational CRM Work Inside Slack

The problem: Employees constantly switch between Slack conversations and Salesforce records to find information or update colleagues.

Potential AI workflow: Salesforce context surfaced within Slack could help teams retrieve account information, summarize deal activity, collaborate around service issues, and initiate supported actions without repeatedly changing applications.

Business outcome to measure: Time spent switching between applications, response times, and workflow completion time.

5. Personalized Marketing and Customer Engagement

• The problem: Marketing teams often have customer information spread across CRM, marketing platforms, service interactions, commerce systems, and other data sources.

• Potential AI workflow: With a well-connected Salesforce environment, AI can help teams interpret customer context, support segmentation and campaign preparation, and assist with more relevant customer interactions while operating within defined governance controls.

• Business outcome to measure: Campaign preparation time, engagement, conversion, and operational efficiency.

Start With One Workflow, Not Every AI Feature

The common thread across these use cases is that the technology should follow the business problem.

Instead of implementing AIforce, Agentforce, and every new Salesforce AI capability at once, organizations can identify one high-value workflow, establish a baseline, implement the right solution, and measure the change.

For example:

Business problem → Current baseline → AI use case → Controlled implementation → Measure results → Scale if successful

This approach can make a Salesforce AI implementation easier to govern while giving leadership clearer evidence of whether the investment is delivering meaningful business value.

What Should Businesses Do in the Next 90 Days After Dreamforce 2026?

Dreamforce 2026 introduced several new possibilities for Salesforce customers, but organizations do not need to adopt everything at once.

A practical approach is to spend the next 90 days determining where AI can create measurable value and whether the existing Salesforce environment is ready to support it.

Days 1–30: Assess Your Salesforce AI Readiness

Start with the foundation.

Review your existing Salesforce environment to understand whether your data, architecture, integrations, permissions, and business processes are ready for AI-assisted workflows.

Focus on:

• CRM data quality and duplicate records

• existing Salesforce automation and Flows

• critical integrations and connected systems

• roles, profiles, and permission structures

• high-volume manual processes

• current AI or Agentforce initiatives

• security and governance requirements

At this stage, the goal is not to choose an AI product.

It is to identify where your current Salesforce environment could limit or enable AI adoption.

Days 31–60: Prioritize High-Value AI Use Cases

Once you understand the foundation, identify a small number of workflows where AI could have a measurable impact.

A simple prioritization framework can help:

Evaluate	Ask
Business value	Does this workflow affect revenue, customer experience, cost, or productivity?
Repetition	Is significant employee time spent completing the same steps?
Data readiness	Is the information required for the workflow available and reliable?
Process clarity	Can the workflow and its exceptions be clearly defined?
Risk	What happens if the AI produces an incorrect recommendation or action?
Measurement	Can improvement be demonstrated using existing KPIs?

Prioritize workflows with high business value, sufficient data quality, clear processes, and manageable risk.

Days 61–90: Build a Controlled AI Pilot

After selecting a use case, move toward a limited implementation rather than an organization-wide rollout.

Define:

• Scope: What exactly will the AI assistant or agent do?

• Access: Which Salesforce records and connected systems can it use?

• Actions: What can it perform automatically?

• Approvals: Which decisions require human authorization?

• Escalation: When should the workflow move to an employee?

• Success metrics: What needs to improve for the pilot to be considered successful?

For an Agentforce project, for example, the first deployment could focus on one clearly defined sales or service workflow rather than attempting to automate an entire department.

What Happens After the First 90 Days?

If the pilot produces measurable improvements, businesses can evaluate additional workflows and gradually expand their Salesforce AI roadmap.

The progression should look something like:

Assess → Prioritize → Pilot → Measure → Optimize → Scale

Organizations with complex Salesforce environments may also benefit from a Salesforce consulting partner that can assess architecture, data readiness, integrations, Agentforce use cases, and governance before implementation.

The objective after Dreamforce 2026 should not be to become the company using the most AI features.

It should be to identify where Salesforce AI can create measurable business value and build the technical foundation required to scale it responsibly.

Ready to Turn Dreamforce 2026 Announcements Into a Salesforce AI Roadmap?

Not sure whether your Salesforce environment is ready for Agentforce and the next generation of AI-powered workflows?

Codleo Consulting can help you assess your Salesforce architecture, identify practical AI use cases, and build a roadmap aligned with your business goals.

Talk to a Salesforce AI Consultant →

Codleo’s Perspective: What Dreamforce 2026 Changes for Salesforce Leaders

From an implementation perspective, the biggest takeaway from Dreamforce 2026 is not any single AI announcement.

It is the growing connection between CRM data, AI reasoning, agents, workflows, and the applications employees use every day.

For Salesforce leaders, this changes the AI conversation from:

“Which AI feature should we buy?”

to:

“Which business processes are ready to become AI-assisted or agent-driven?”

Based on our experience working across Salesforce environments, Codleo believes organizations should focus on four areas before scaling their next phase of AI adoption.

1. Fix the CRM Foundation Before Adding More Intelligence

AI can expose weaknesses that already exist within a Salesforce environment.

If customer information is duplicated, integrations are unreliable, permissions are inconsistent, or important processes still happen outside the CRM, adding an AI agent does not automatically resolve those issues.

The foundation still matters.

Before moving toward advanced agentic workflows, businesses should understand whether Salesforce contains the data, process logic, integrations, and governance required to support them.

2. Choose Business Problems Before Choosing AI Features

Dreamforce introduced a growing portfolio of AI capabilities, but not every organization needs to implement all of them.

A service organization handling large volumes of repetitive cases will have different priorities than a B2B sales organization trying to improve account research and pipeline management.

The technology decision should come after the business problem is clearly defined.

3. Design Human and AI Workflows Together

The strongest enterprise use cases are unlikely to remove people from every process.

Instead, organizations should determine:

• what AI can safely handle;

• where an agent should make a recommendation;

• which actions require approval;

• when an exception should reach an employee; and

• who remains accountable for the final business decision.

This creates a more practical model in which AI handles appropriate work while employees retain control over decisions requiring judgment, context, or accountability.

4. Treat AI Adoption as a Roadmap, Not a One-Time Implementation

Salesforce AI adoption will continue evolving as platforms, models, agents, and business requirements change.

Organizations should therefore build an AI roadmap that can progress from:

readiness assessment → focused use case → controlled pilot → measurable results → optimization → broader deployment

This approach also makes it easier to evaluate whether each new use case is actually creating business value.

The Question Salesforce Leaders Should Ask Now

Dreamforce 2026 demonstrated where Salesforce is heading.

The more important question for individual organizations is:

“What needs to change in our Salesforce environment today so that our data, people, processes, and AI agents can work together effectively tomorrow?”

Answering that question requires more than enabling a new feature. It requires understanding the existing Salesforce architecture, identifying the right business use cases, and building an implementation roadmap around measurable outcomes.

That is where experienced Salesforce consulting and AI implementation expertise can add real value—especially for organizations managing complex data, integrations, multiple Salesforce clouds, or enterprise governance requirements.

Final Thoughts: Dreamforce 2026 Is About What Businesses Do Next

Dreamforce 2026 signals a broader change in how Salesforce sees the future of CRM. AI is moving beyond content generation and isolated assistance toward systems that can understand enterprise context, support decisions, participate in workflows, and take approved actions.

For Salesforce customers, however, the opportunity is not simply to adopt every new announcement.

The priority should be identifying where AI can solve a measurable business problem and determining whether the existing Salesforce environment—including data, integrations, automation, permissions, and governance—is ready to support it.

Organizations that get this foundation right will be better positioned to evaluate technologies such as AIforce and Agentforce as part of a practical, scalable AI roadmap.

Turn Dreamforce 2026 Ideas Into an Actionable Salesforce AI Roadmap

Not sure which Dreamforce announcements are relevant to your Salesforce environment?

Codleo Consulting can help you assess your existing Salesforce architecture, identify high-value AI and Agentforce use cases, and plan an implementation roadmap aligned with your business priorities.

Whether you are exploring Agentforce, modernizing an existing Salesforce org, improving integrations, or preparing your CRM data for AI-driven workflows, start with the business problem—not the technology.

Talk to a Salesforce Consultant →