Conversation Designer
Conversation Designer is the visual bot-building UI inside Salesforce Einstein Bots, the previous-generation conversational AI product that pre-dates Agentforce.
Definition
Conversation Designer is the visual bot-building UI inside Salesforce Einstein Bots, the previous-generation conversational AI product that pre-dates Agentforce. It lets admins lay out intents, entities, dialogs, and flow-like branching logic on a canvas, attach them to channels (Messaging, Web Chat, WhatsApp), and configure handoff rules to live agents. The output is a deployable Einstein Bot that classifies user messages, extracts named entities, and walks the conversation through structured dialog steps.
Conversation Designer is mostly a legacy surface as of 2026. Salesforce continues to support it, and many orgs still run Einstein Bots in production, but new conversational AI work has moved to Agent Builder and the Agentforce architecture. The two paradigms differ fundamentally: Designer is intent-and-dialog-driven with deterministic branching, while Agentforce is topic-and-action-driven with LLM-based reasoning. Migration is not automatic and most orgs run both side by side during transition.
Why Conversation Designer still ships even as Agentforce takes over
Where Conversation Designer fits in the bot stack
Conversation Designer is the editor surface; Einstein Bots is the runtime; the underlying NLP model is the Einstein Bots intent classifier. Open Setup, Einstein Bots, pick or create a bot, and Designer opens as the canvas. The left rail lists dialogs (logical conversation steps), entities (named values to extract), and variables (state held across turns). Connectors at the bottom expose the bot to channels and integrate with Flow for record actions. The architecture is recognizable to anyone who built bots in 2019 to 2022; it has aged into a stable plateau rather than evolving rapidly.
Intents and the training-phrase model
Intents are the classification primitive. Each intent has a name, a description, and 20 to 100 training phrases that the model uses to recognize messages of that type. When a user message arrives, the bot computes intent scores against all defined intents and picks the highest. The dialog tied to that intent runs next. Training phrases need to be varied (different lengths, different vocabulary, different sentence structures) for the model to generalize. The biggest difference from Agentforce topics: intents require literal training phrases, while Agentforce topic classification descriptions are natural-language descriptions of when to use the topic.
Dialogs and the deterministic branching model
Dialogs are the conversation steps. Each dialog has a series of actions: ask a question, validate input, set a variable, invoke a Flow, call Apex, send a response, transition to another dialog. The transitions are deterministic: dialog A always goes to dialog B if condition X is true. There is no LLM-driven planning in Conversation Designer. The output is fully predictable given the same input, which is the appeal (and the limitation) of the model. Predictability makes regulated conversations easier to ship and certify; the limitation is that the conversation cannot adapt to phrasing the dialog author did not anticipate.
Entities and slot-filling
Entities are typed values the bot extracts from user messages. Standard entities (DateTime, Number, Email, Currency) handle common values. System entities pull from Salesforce data (a specific Account name, a known Case Number format). Custom entities are admin-defined lists or regex patterns. A dialog can require an entity before proceeding ("Slot filling: ask the user for their order number, validate against the OrderNumber entity, store in variable orderNumber"). The model is sturdy and explicit, but every entity needs to be defined and trained, while Agentforce Atlas extracts parameters from natural-language descriptions without explicit entity types.
Channels, deployments, and live agent handoff
Bots deploy to Messaging, Web Chat, SMS, WhatsApp, Facebook Messenger, and Slack through Connectors. Each Connector handles channel-specific formatting (rich text, buttons, quick replies). The handoff dialog is a special dialog type that ends the bot session and routes the conversation to a live agent queue. The transcript carries to the agent so they pick up mid-conversation. Handoff rules can be condition-based (escalate when the user says "agent" three times) or signal-based (escalate when sentiment turns negative, using the Sentiment Analysis service).
Migration to Agentforce: what carries and what does not
There is no one-click migration from Conversation Designer to Agentforce. The conceptual mapping is: intents become topic classification descriptions, dialogs become topic instructions plus actions, entity extraction becomes parameter descriptions, Flow invocations become Agent Actions. The hard part is reframing deterministic dialog branches as LLM-driven planning. Some Designer logic does not translate cleanly (multi-step slot filling with custom validation) and is better redesigned in Agentforce rather than ported. Most teams keep Einstein Bots running for stable, well-understood conversations and build new conversations directly in Agentforce.
Where Designer still wins over Agentforce
Designer is not obsolete. Three places it still wins: deterministic conversations that regulators require to follow a strict script (financial services disclosures, healthcare informed-consent flows), simple bots where LLM cost would dominate the value, and orgs with deep investment in Einstein Bots training data that would be expensive to replicate. The honest call for most other use cases is to build new in Agentforce. Salesforce has signaled that Conversation Designer will continue to be supported but will not get major new capability investment.
How to decide whether to build a new bot in Designer or in Agentforce
The decision is rarely about technical capability. It is about regulatory predictability, cost sensitivity, and team familiarity. Conversation Designer is the right call when conversations must follow a scripted path; Agentforce is the right call when conversations need to adapt to unanticipated user phrasing. Most orgs land on a mix.
- List the conversations the bot must handle
Write each use case as a one-line description. Tag each with regulatory requirements, expected volume, and adaptation needs. The tagging drives the build decision.
- Sort use cases by determinism requirement
Use cases that must follow an exact script (compliance disclosures, regulated questions) go to Designer. Use cases that benefit from adapting to user phrasing (general FAQs, exploratory questions) go to Agentforce.
- Estimate volume and per-conversation cost for each candidate
Agentforce conversations cost roughly two dollars each in 2026. Einstein Bots run on the org subscription with no per-conversation cost. High-volume, low-value conversations may be cheaper in Designer.
- Build a pilot in the chosen surface
For Designer, create a bot in Setup, Einstein Bots, design intents and dialogs in the canvas, deploy to a single channel. For Agentforce, follow the standard agent-build path. Pilot one use case for two to four weeks.
- Compare actual user outcomes across the surfaces
If you pilot the same use case in both, compare resolution rate, deflection, customer satisfaction. The data resolves debates that the technical comparison alone cannot.
- Document the surface decision per use case for the long term
Once you decide, write down why each use case lives where it lives. The decision rationale matters more than the decision itself because conditions change.
- Plan migration for use cases that drift to the other surface
Conditions change. A regulatory requirement loosens, a cost ratio shifts, a team gains LLM expertise. Revisit the surface decisions annually and migrate use cases that no longer fit where they were built.
Classification primitives in Designer; each holds 20 to 100 training phrases that train the intent model.
Conversation steps with deterministic branching; the building block of Designer bot logic.
Typed value extractors (Standard, System, Custom). The Designer equivalent of Agentforce action parameters.
Channel adapters (Messaging, Web Chat, WhatsApp, Slack). Drive how the bot is exposed to users.
Special dialogs that end the bot session and route to a live agent queue with the transcript.
- Training phrases must be varied for intents to generalize. Twenty near-identical phrases produce a model that only matches near-identical messages in production.
- Dialog branching is deterministic. Designer bots cannot adapt to phrasing the author did not anticipate, which is a feature for compliance and a limitation for general use.
- Migration to Agentforce is not automatic. Plan to re-design rather than port; the architectural shift is large enough that direct conversion produces brittle agents.
- Designer is in a stable plateau. New capability investment is going to Agentforce. Long-term roadmap planning should account for the direction even if the immediate decision is Designer.
- Per-conversation cost math can favor Designer at high volume for simple bots. Run the numbers before defaulting to Agentforce on a high-volume FAQ use case.
Trust & references
Cross-checked against the following references.
- Einstein Bots referenceSalesforce
- Agentforce as Designer successorSalesforce
Straight from the source - Salesforce's reference material on Conversation Designer.
- Einstein Bots OverviewSalesforce Help
- Conversation Designer CanvasSalesforce Help
- Intents and Training PhrasesSalesforce Help
About the Author
Dipojjal Chakrabarti is a B2C Solution Architect with 29 Salesforce certifications and over 13 years in the Salesforce ecosystem. He runs salesforcedictionary.com to help admins, developers, architects, and cert/interview candidates sharpen their fundamentals. More about Dipojjal.
Test your knowledge
Q1. What is Conversation Designer used for?
Q2. How does Conversation Designer relate to Bot Builder?
Q3. Who benefits most from Conversation Designer?
Discussion
Loading discussion…