Einstein Case Classification
also Einstein Case Wrap-Up; AI that predicts and fills case fields from past closed cases
Builds a model from an org’s closed cases and predicts values for case fields — when a case is created (Case Classification) and when a chat ends (Case Wrap-Up); service reps see the recommendations in the Service Console, and with the add-on licence it can fill fields automatically and route cases. Salesforce: “To set up one or both apps, create a model that tells Einstein what closed cases to learn from.” Salesforce: “The add-on license also lets Einstein Case Classification automate field completion and optionally route cases.”
Einstein Work Summaries
AI summaries of service conversations and cases: summary, issue and resolution
At the end of a conversation between a service rep and a customer, or with one click on a case, generative AI writes a summary, the issue and the resolution, on messaging, voice, chat and email channels; reps review, edit and save them, or admins can have them saved automatically without review. Salesforce: “At the end of a conversation between a service rep and a customer, or with a single click from the Case page, Einstein predicts and fills a summary, issue, and resolution.” Salesforce: “Admins can turn on automatic saving to save summaries automatically without service rep review.”
Enhanced Summaries
AI case summaries tailored to the reader’s role
Generates a case summary, issue and resolution with one click on the case page, tailored to the role of the service rep or supervisor reading it; users review, edit and save them and can post them to the case feed. Salesforce: “Boost service team efficiency and save time with AI-generated case summaries tailored for each role and context.”
Einstein Article Recommendations
AI that suggests knowledge articles for each new case
A model built from an org’s case history that examines new cases as they come in and suggests the knowledge articles most likely to solve them, in the Service Console; a version for voice and messaging conversations suggests articles from what is said. It was turned on for Lightning Knowledge with a prebuilt model from Summer ’21. Salesforce: “Einstein Article Recommendations is being turned on for Lightning Knowledge starting Summer ’21.”
Service Replies for Email
also called Einstein Service Replies for Email; generative AI that drafts service email replies from knowledge articles
Drafts an email reply to a customer from a recommended knowledge article, or from up to three relevant articles; service reps review, edit and send it. Salesforce: “When service reps select a recommended article on a case, Einstein analyzes it and then drafts an email response based on the article.”
Einstein Service Replies for Chat
generative AI that drafts replies to customers during live chat and messaging
Analyses a live chat or messaging session as it unfolds and drafts replies for the service rep, who can post them with one click or edit them first. Salesforce: “When a service rep begins a customer Live Chat or Enhanced Messaging session, Einstein analyzes your chat as it unfolds and then drafts replies from context or based on an assigned Agentforce Data Library.”
Einstein Reply Recommendations
AI that recommends approved replies during chat and messaging
A model built from an org’s closed chat transcripts that recommends replies to service reps during chat and messaging sessions; it recommends only replies the org has reviewed and published. Salesforce: “Replies are based on your Salesforce org’s closed chat transcripts, and Einstein recommends replies only if you’ve reviewed and published them.”
Einstein Conversation Mining
AI that finds why customers contact you, from conversation records
Reviews conversation data and extracts each customer’s contact reason, then groups conversations by reason in reports, to find candidates for bots, process changes and knowledge articles. Salesforce: “Einstein Conversation Mining uses advanced AI to discover insights from your conversational data.” Salesforce: “Einstein reviews conversation data and extracts the contact reason.”
Einstein Knowledge Creation
AI that drafts knowledge articles
Drafts knowledge articles, filling each Knowledge field of an article type according to the response type an admin maps to it. Salesforce: “Learn how field mapping and response types affect your article drafts.”
Agentforce Service Agent
AI agent that answers customers and resolves their cases on its own
AI agents that connect to customer channels such as messaging, process incoming cases and resolve common requests on their own, escalating complex or sensitive ones to service reps. Salesforce: “Agentforce Service agents intelligently support your customers by processing incoming cases and autonomously resolving common inquiries.”
Einstein Lead Scoring
AI score for every lead, predicting which will convert
Scores leads so sales teams can prioritise them, using a model built from the org’s past converted leads — or, where there is too little data, a global model built from anonymous data from many Salesforce customers. Salesforce: “Give your sales team access to scores that help them prioritize leads.” Salesforce: “The global model uses anonymous data from many Salesforce customers.”
Einstein Opportunity Scoring
AI score for every deal, predicting which will close
Scores opportunities so sales teams can focus on the deals most likely to close, using a model built from the org’s past opportunities or, with too little data, a global model from anonymous data from many Salesforce customers; it can be on by default. Salesforce: “If Einstein Opportunity Scoring is on by default, make sure that scores appear where you want, such as your customized opportunity page layouts and public list views.”
Einstein Forecasting
AI sales forecasts that also track how sales teams are doing
Gives forecast managers AI predictions of forecast results, built from the org’s opportunity data, and tracks how sales teams are performing. Salesforce: “Provide your forecast managers with AI-powered intelligence that improves forecasting accuracy, predicts results, and tracks how sales teams are doing.”
Sales Emails
also called Einstein Sales Emails; generative AI that drafts personalised sales emails to contacts and leads
Generates personalised sales emails to contacts and leads — check-ins, introductions, follow-up nudges, meeting invites or custom types — grounded in the record’s data. Salesforce: “Einstein generative AI generates personalized emails for contacts and leads, and uses Einstein to personalize and contextualize the emails, a method known as grounding.”
Generative ECI
call summaries and generative conversation insights in Einstein Conversation Insights: AI summaries and AI-detected insights for recorded sales calls
Automatically summarises recorded voice and video sales calls — summary, customer impression and next steps — and uses a large language model to detect insights the org defines in natural language; once on, it runs on every call Einstein Conversation Insights processes. Salesforce: “Once the feature is on, a call summary and any generative insights configured will automatically be processed for all ECI calls.”
Einstein Coach
AI feedback on sales reps’ recorded practice pitches
Analyses the transcript of a sales rep’s recorded practice pitch and gives personal feedback on strengths and areas to improve, in Enablement programmes or on demand. Salesforce: “Einstein Coach provides reps with personalized and actionable feedback on how to improve their sales pitch.” Salesforce: “Einstein Coach analyzes your call’s transcription and provides personal and actionable feedback, highlighting your key strengths and growth opportunities.”
Agentforce Engagement
formerly Agentforce Lead Nurturing; AI agent that emails prospects, answers them and hands them to sales reps
An autonomous AI agent that sends customised outreach and follow-up emails to prospects, answers their questions and hands interested prospects to sales reps to book meetings; it acts as a user in the org with its own user record. Salesforce: “Built on the Salesforce trusted AI platform, Agentforce Engagement is an autonomous agent that helps you scale your top of funnel efforts, streamline prospect qualification, and generate pipeline around the clock.”
Agentforce Employee agents
AI agents that help employees find information and complete tasks
AI agents for an organisation’s own employees that help them find information, complete tasks and get support, in Salesforce, chat and Slack; each can be assigned to specific profiles or users. Salesforce: “Agentforce Employee agents help employees find information, complete tasks, and access personalized support across channels.”
Agentforce
AI agents an organisation builds or sets up
Goal-oriented AI agents that carry out sequences of tasks, hold conversations and answer from business data; some assist a Salesforce user, others act on behalf of a user or customer within the use cases and guardrails an admin sets. Salesforce: “They can initiate and complete a sequence of tasks, handle natural language conversations, and securely provide relevant answers drawn from business data.” Salesforce: “Other types can also act on behalf of a user or customer, based on the use cases and guardrails an admin specifies.”
Prompt Builder
tool for building generative AI prompts that run on Salesforce records
Lets admins create prompt templates that send CRM data to large language models, for example to draft sales emails or fill record fields with an AI-written summary or description. Salesforce: “Prompt Builder helps you to make effective prompts that safely connect you and your data with LLMs.”
Einstein Prediction Builder
custom AI predictions that score Salesforce records
Lets an org build its own predictions that score its Salesforce records; a prediction scores records only while it is enabled. Salesforce: “Enable your prediction to put it to work. Until a prediction is enabled, it’s not scoring records.”
Einstein Search
personalization: AI that reorders each user’s search results by what they work with
Reorders a user’s search results based on the records they work with most, looking at the last 200 records they interacted with; on by default. Salesforce: “Personalization is an AI feature that reorders search results based on the objects and records you work with most.”
AI-Generated Search Answers
generative AI answers to search questions from knowledge articles or Data Cloud
Generates answers to users’ search questions from knowledge articles and other data sources; since Spring ’25 only the version that answers from up to five Data Cloud sources can be newly turned on. Salesforce: “AI-generated Search Answers uses knowledge articles and other sources to generate tailored responses to user questions.” Salesforce: “Starting in March 2025 (Spring ’25), you can’t enable AI-Generated Search Answers.”
