Docket’s Salesforce integration syncs AI-driven prospect conversations into Salesforce to create Leads, and Events, as well as update existing Contacts, Accounts, and Opportunities.
With the agent-level CRM write-back capabilities, you have granular control over exactly what data gets written back and where.
Pre-requisites
Admin seat in Docket — required to manage integrations
Salesforce edition with API access (Enterprise, Unlimited, Performance, Developer, Professional) — required
OAuth credentials with API Enabled — required for authentication
System Administrator or Integration User — recommended for full access
Read access to Account, Contact, Lead, Opportunity objects
Write permission to Account, Contact, Opportunity, Lead, and Events
Integration Process
Step 1: Connect Your Salesforce Account
Navigate to the Settings section within your Docket platform and select the Integrations tab.
From the list of Apps and Integrations, locate Salesforce CRM and click Connect.
Select Production or Sandbox.
Sign in with Salesforce OAuth, approve the API access scope, and click Allow.
Step 2: Configure Read Access (Import to Docket)
Docket separates Read and Write access. Read access syncs CRM data into Docket to enrich live agent conversations via the CRM data.
Click the gear icon on the Salesforce integration card and choose Settings.
Toggle the Read switches for Account, Opportunity, Leads, and Contacts to allow Docket to retrieve contextual data.
After tailoring the integration to your needs, click "Save" to activate your settings.
Step 3: Enable Agent-Level Write-Back
Write-back is configured individually for each agent, giving you granular control over where conversation data is logged.
Navigate to your Marketing Agent and select the Analysis and Actions tab.
Scroll down to Post Call Actions.
Toggle on Sync to Salesforce to automatically write conversation data to Salesforce when an engagement ends.
Step 5: Configure Object and Field Mapping
Once Sync to Salesforce is enabled, expand the configuration to map fields for Lead, Contact, Account, Opportunity, and Activity Events.
Write operations are non-blocking—if an object is unavailable or permissions are missing, Docket simply skips it and continues.
How Docket Keeps Salesforce Updated:
Leads: If the person isn’t found in Salesforce, Docket creates a new Lead. Click Add New Field to map custom Docket fields (like UTM parameters, Click IDs, and your newly created Post Call Analysis fields) directly to your Salesforce Lead properties.
Contacts & Accounts: Docket doesn’t create Contacts or Accounts. When a matching record exists, Docket updates the mapped fields.
Opportunities: Docket doesn’t create Opportunities. When an existing Opportunity is associated with the Contact/Account, Docket updates the mapped fields.
Events: After every conversation, Docket logs an Activity Event. Event linking starts with Leads, but if a Contact is discovered, the Event is linked to the Contact, and subsequently to the associated Opportunity (if read access exists). The event includes the agent's name, interaction count, page URL, and UTM parameters.
Advanced Tracking & Attribution
Docket now supports writing rich attribution and interaction data directly to Salesforce standard or custom fields:
UTM Parameters & Click IDs: Map fields like
Docket UTM Source,Docket Page URL, andDocket Click IDto custom Salesforce fields (e.g.,Docket_UTM_Source__c) to capture exactly which campaigns drive meaningful conversations. This enables exporting click IDs back to ad platforms for offline conversion tracking.Interaction Lifecycle: Track
Docket First Interaction Date,Docket Latest Interaction Date, andDocket Interaction Count. Use these fields to trigger Salesforce alerts when a cold lead returns to chat, or to build lists of high-intent visitors.
Leveraging Post-Call Analysis Field Mapping
Beyond standard properties, you can define custom Post Call Analysis instructions to automatically extract insights from every conversation (e.g., Pain Points, Budget, Interested Products) and these extracted fields will now be available to map directly into your Salesforce properties.
Navigate to your Marketing Agent and select the Analysis and Actions tab.
Under Post Call Analysis, click Add to create a new extraction prompt.
Select the Data Type (e.g., Text), give it a Name, and provide clear instructions for the LLM to evaluate the call. You can optionally define permitted responses (like "Qualified" or "Unqualified").
Mapping Post-Call Analysis fields unlocks powerful CRM automations:
Targeted Segmentation: Map captured pain points or qualification statuses to build targeted lists. For example, enroll everyone who mentioned "integration challenges" into a specific nurture track.
Rep Readiness: Discovery questions, pain points, and next steps are written to the Timeline Event. A simple workflow can send this summary to the Account Executive before a booked meeting, ensuring they walk in fully prepared.
Pipeline Attribution: Group deals by the mapped
Docket Page URLor agent ID to see exactly which website pages and agent configurations are driving actual pipeline, not just conversation volume.
Other notes
Docket syncs data from Salesforce every 24 hours
Docket pushes updates to Salesforce almost in real time
Salesforce to Docket reads refresh every 24 hours
Disconnecting and holding data pauses all sync operations instantly.
Permanent delete removes ingested data and configuration.
Docket respects field and object permissions. If write access is not granted for a specific object, the workflow skips it and continues without blocking.
Rate limits may delay sync; automatic retries occur.
Your team can now leverage the combined power of Docket and Salesforce to streamline workflows and enhance customer relationships.
For additional assistance or troubleshooting, contact [email protected].