Migration steps are not cumulative. Depending on your initial version, you will have to apply all upgrade tasks listed on the main Build Item (eg: Build 3.58.x), all the way from your current version up to the version you are upgrading to.
Build 3.58.7 (04tRl000001GPzZIAW)
Apply the package
If you are upgrading from a version below 3.58, you must run the migration script
Duplicates: customer-specific display issue in Duplicate Data Steward View (Pairs not displayed)
Build 3.73.x
Build 3.58.8 (04tRl000001GhDJIA0)
Apply the package
If you are upgrading from a version below 3.58, you must run the migration script
Build 3.58.6 (04tRl000001FVsXIAW)
Apply the package
If you are upgrading from a version below 3.58, you must run the migration script
Install Delpha Connector for LinkedIn Enrichment
Install the Delpha Connector browser extension to link your LinkedIn account with Delpha and unlock powerful enrichment features. Compatible with Chrome and Edge for seamless integration.
Install the Delpha Connector for LinkedIn Enrichment
To enable advanced LinkedIn enrichment features in Delpha:
Install the Delpha Connector browser extension from the Chrome Web Store
The extension is compatible with both Google Chrome and Microsoft Edge
Once installed, it seamlessly connects your LinkedIn account to Delpha, allowing for smoother and more accurate data enrichment
This integration helps automate and enhance LinkedIn data collection within your Delpha workflows.
Build 3.60.4 (04tRl000001IyUzIAK)
Apply the package
Build 3.58.5 (04tRl000001FAmTIAW)
Apply the package.
If you are upgrading from a version below 3.58, you must run the migration script
Real time duplicate detection not working when an exact includes a boolean (asynchronous detection is ok)
Build 3.72.x
Build 3.60.1 (04tRl000001IJTRIA4)
Apply the package
Build 3.79.x
Build 3.58.7
This version provides a fix on the mass process from the Duplicate Data Steward view: some duplicate records pairs could be deleted before or after being processed
This issue is only impacting the display of the duplicate pairs. No record is wrongly deleted because of it.
Build 3.77.x
Duplicate
Explore all the latest updates to the Delpha Duplicate use case here.
Build 3.74.x
Delpha Setup
Explore all the latest updates to the Delpha Setup app here.
Build 3.76.x
Build 3.60.7
Score Meter
Fix tooltips not displaying for bullets
Yellow bullet for Status = Unknown is now a triangle
Display LinkedIn title instead of LinkedIn name
Build 3.70.0
Negative view for tiles
Add "Last Modified Date" column to the steward
Implement filtering for data steward views (Data Quality and Duplicate)
Build 3.56
New Score Meter: Review & Fix Everything in One Place
Before: You had to switch between the Score Meter and the Bot to review and apply corrections.
Now: Everything happens directly in the Score Meter — review, select, and apply corrections all in one place.
New interface: Cleaner, clearer, and optimized for batch actions. You can now apply changes to 1 / some / or all suggestions with just a few clicks.
No more back and forth. Just smarter data review in a single streamlined experience.
Try it out now and speed up your remediation process!
Beta Features
Build 3.70.0
First draft of the new create account / create contact use cases
Build 3.60.6
Fix some display issued on Cross Merge Field Rules.
Build 3.74.1
Data Quality for Lead
Build 3.58.6
This build introduces some enhancement in the display of the Duplicate Data Steward pairs. Loading time is divided by 2.
Data Steward Views
Delpha’s updated Data Steward View offers a fully customizable, Salesforce-native experience. Easily filter records, choose the fields you want to see, and create or share your own views.
Build 3.60.6
Bug Fix: incomplete Legal ID
Legal address was missing the street part. This is fixed.
Build 3.60.7
Add missing permission set for duplicate merge editor.
Build 3.60.4
Fix some limitations related to the Auto Number fields when used in Expression or Filtering rules.
Build 3.58.8
Bug Fix: Filtering with Select All
Previously, when applying a text filter in the Data Quality or Duplicate Data Steward views, the Select All action was incorrectly applied to all records instead of only the filtered results. This issue has been resolved in this release.
Delpha Setup
Setting Up Delpha in 3 Easy Steps
The Delpha Setup menu provides quick access to key configuration areas:
These tasks help ensure optimal performance and unlock advanced enrichment features.
Welcome to Delpha Documentation
Explore Delpha’s documentation for setup, integration, deduplication, and LinkedIn enrichment. Includes FAQs, job tracking, and support options.
This Help Center is organized into structured categories to help you quickly find the guidance you need, whether you're setting up Delpha for the first time or fine-tuning deduplication rules.
DELPHA DOCUMENTATION
Start here to understand how Delpha works and how to get set up.
Welcome to Delpha: Introduction and overview
Delpha Setup:
Delpha Upgrade:
Delpha Apps:
Use Cases Setup:
Delpha Campaigns:
HOW TO – FAQ
Common tasks, step-by-step guides, and practical how-to articles.
Delpha Integration:
Delpha Job Tracking:
Delpha LinkedIn Connector:
TROUBLESHOOT
For when you need assistance or something isn’t working as expected.
Install Delpha package from the Salesforce App Exchange
Quickly enable Delpha for your Salesforce org with minimal setup.
How to Install Delpha from the Salesforce AppExchange
Log in to Salesforce using an account with system administrator rights.
Go to the Salesforce AppExchange and search for Delpha.
or use the
On the Delpha listing page, click “Get It Now” to begin.
Choose your environment (Production or Sandbox) based on where you want to install the package.
Log in again if prompted to confirm your identity.
Accept the terms and conditions by checking the agreement box.
Set the security level, selecting who should have access to Delpha (Admins only, All users, or specific profiles), then click Next.
Click “Install” to start the installation process.
Once complete, Salesforce will send you a confirmation email indicating that Delpha was successfully installed.
During the setup, you are asked to Approve Third-Party Access
api.hubapi.com: HubSpot endpoint used by Delpha to synchronize HubSpot and Salesforce records after Delpha merges Salesforce records during the Duplicate process.
This integration is disabled by default and can be enabled only by a Delpha Administrator in the Duplicate Settings.
[Score meter] Fix tooltip not displaying for bullets
[Data steward] Adjust combobox size to fix the content
[Data quality] Yellow bullet for unknown is now a triangle
What's New
This section helps users quickly access the latest improvements across the platform.
The "What's New" section is organized by key feature areas, making it easy to browse recent updates and improvements. Each item links to a dedicated page summarizing the latest changes and enhancements for that component:
Information that the data quality is being assessed is displayed at icon level
Current information is still available while processing, in read only mode.
Build 3.72.1
Duplicate check at creation
is now available from the General Settings bloc in App: Delpha Setup > Duplicate > Detection Settings
New display rules for the Merge Editor
For an easier analysis, Candidates values are displayed in red when they are different from the Master values.
In groupings, Candidates grouped values are displayed in red when they are different from the Master grouped values
Build 3.58.2 (04tRl000001EpLRIA0)
Duplicate Settings - Migration to the new JSON based format
From this version, the Duplicate Settings have been moved from SObejcts into a unique JSON.
This reduces the complexity of Duplicate settings spanned over many objects (mainly DDQ Duplicate Objects, DDQ Duplicate fields...). This change requires a migration step that is made simpler by using a dedicated migration package provided by Delpha.
After the main Delpha package is installed, add the DelphaUpgrade258 package (unmanaged) to your org. It will facilitate the migration process.
This section is essential for Delpha Administrators managing Delpha integration and workflows.
Access to the app is granted only to users with the Delpha Admin Permission Group
Install Delpha
Installing Delpha in Salesforce
To get started with Delpha in Salesforce, follow these quick setup steps:
from the AppExchange
Assign a Delpha licence to the users
Assign Delpha licenses in Salesforce easily via Installed Packages or directly from user profiles. Ensure your users have access to Delpha features with quick license management.
Assigning Delpha Licenses in Salesforce
You can assign Delpha licenses using either of the following methods:
Connect the org to Delpha
Connect your Salesforce org to Delpha by using the Delpha Setup app and configuration tab. Fast integration in just a few clicks to enable full platform functionality.
Connecting Your Salesforce Org to Delpha
To link your Salesforce organization with Delpha's backend services:
Open the Delpha Setup app in Salesforce
Up to Build 3.50.6
For Delpha versions later than 3.50.6,
Build 3.70.x
Upgrade Steps from any version previous 3.63.0
Apply the package
Assign a Delpha Permission Set Group
Assign the right Delpha permissions by selecting a persona-based permission set group—Admin, Data Steward, or User—for each Salesforce user. Simplify access control and role-based setup.
Assigning Delpha Permission Set Groups
To assign the correct permissions in Delpha, follow these steps:
Determine the User’s Persona: Choose the appropriate role —
(Optional) Activate Delpha Licenses purchased via Heroku Marketplace
Follow these steps to activate Delpha licenses purchased through the Heroku Marketplace and set them up on your Heroku account.
If you purchased Delpha through the Heroku Marketplace, follow these steps to activate your licenses. If you’re a direct customer, you can skip this section.
Build 3.56
Single stop for Duplicate setup
All the settings are now available in one single place, in the Duplicates section of the Delpha Setup app.
Build 3.70.0
[Ultimate]
App Delpha Setup > Data Quality > Account
Once the Ultimate toggle is activated, you can manually trigger a reassessment of the Ultimate analysis.
in the App Delpha Setup, the left-hand menu is now grouped by functional categories (e.g., Data Quality, Duplicates, Conversations), making navigation more intuitive and setup tasks easier to locate.
Process Duplicates from the Score Meter & Merge process New UI
For Accounts and Contacts, processing of Duplicates at record level is now part of the Score Meter. It is no longer done in a conversation but in a dedicated layout.
Import Conversations
Import conversations into Delpha to enable advanced data quality and deduplication use cases. Select recommended components for full feature coverage and smarter data management.
Importing Conversations into Delpha
For an initial setup, the conversations are automatically imported by applying the step Delpha Conversations
Form now on, Delpha performs an analysis on your targeted duplicate population and proposes you to discard in one click all values that may trigger too much pairs. Thsi information is displayed in the app Delpha Setup > Duplicate > Object Settings, in the Properties section of each object.
You can
Build 3.58.5
Account - Main Entity Address
When enriching an Account, you retrieve the Main Entity address in the tab Delpha > Address. You can reuse this information in a regular address (billing, shipping...).
Note that this address can be different from the Legal Address.
Build 3.58.2
Legal Id types
2 new Legal Ids types are supported by Delpha : Australia and New Zealand
LinkedIn enrichment cleanup
Build 3.60.1
Manage Duplicate from the Score Meter
The Score Meter now includes the Duplicate information for Accounts and Contacts.
By clicking on the link you start the
Conversation Builder
Build smart, no-code conversational workflows with Delpha’s Conversation Builder in Salesforce. Guide users with interactive Q&A and automate business logic all within a visual decision-tree interface
All the conversations available in your org are listed here.
Build 3.56
New Data Steward View – Improved Usability & Customization
We’ve redesigned the and the interface to align with Salesforce’s native List View experience:
Familiar Navigation: Use the standard cog and filter icons to create
Build 3.58.2
Direct access to filters
Based on the users feedback, we improved the Data Steward view by having the most common filters back on the main data steward screen area.
Build 3.60.1
You can assess selected records directly from the Data quality Steward view.
If you click on the Assess button, all selected records will be Data Quality assessed. This feature is compatible with the Select All, allowing mass reassessment from the Data Quality Steward View.
First Steps
Connect your Salesforce org to Delpha, review configuration health, and optimize batch size settings for the Ultimate Account use case. Ensure smooth setup and maximum performance.
First Steps section
Connect to Delpha
Delpha Data Quality
Use Delpha’s Data Quality and Duplicate Steward views in Salesforce to monitor record quality, apply intelligent recommendations, and resolve duplicates efficiently.
Delpha Data Quality Menu Overview
The Delpha Data Quality section provides dedicated views for data stewards to monitor and improve the quality of records in Salesforce:
Delpha Apps
Access key Delpha apps in Salesforce, including Setup, Data Quality, and Score Meter, to streamline configuration, monitor data health, and optimize data-driven decisions.
Delpha Apps Overview
The Delpha Apps section in Salesforce gives you access to the core components of the Delpha platform:
Delpha Setup –
[Duplicate] Add missing permission set for duplicate merge editor
Option 1: Via Installed Packages
Go to Setup in Salesforce
In the Quick Find box, search for and select Installed Packages
Find Delpha in the list and click Manage Licenses
Click Add Users, select the desired users, and click Add
Option 2: Via User Profile
Go to Setup > Users via the Quick Find box
Click the name of the user you want to license
In the Managed Packages section, find Delpha and click Assign
Click Save to confirm
Both methods will properly activate Delpha access for selected users.
Do not forget to assign a Delpha license to you
This step is useless when installing Delpha on a Sandbox org
Go to the Delpha Configuration tab
In the First Steps section, click Connect
A new window will open—follow the prompts to authorize and complete the connection
Once connected, Delpha can begin syncing data and enabling core features.
The profile and permissions of the user that performed the connection are going to be used by Delpha. Make sure the user has an administrator profile and will keep the Delpha Licence and Delpha Admin permissions.
We recommend using a dedicated 'technical' user with Admin rights for that step.
This user should not be reused when connecting Sandboxes to Delpha.
Using the same user to connect sandboxes will overwrite the current settings. In other words, you will disconnect from Delpha the org (prod or sandbox) where this user has done the initial connection.
Delpha Technical Setup is done!
Admin
,
Data Steward
, or
User
— based on their responsibilities.
Assign the Permission Set Group:
Go to Users in the Salesforce Setup.
Find the user and assign one of the following groups:
Delpha Admin – Full access including Setup and Data Quality tools
Delpha Data Steward – Access to the Data Quality and Duplicate Steward Views to do mass review / mitigation, Score Meter and Bot
Delpha User – Access to Score Meter and Bot only
Save your changes to apply the permissions.
This ensures each user has the appropriate access to Delpha features based on their role.
Do not forget to assign the Delpha Admin permission set group to you
Highlight differences in grouping when merging manually
Fixes
Setup: review of Delpha permission sets for Apex permissions
Score Meter: fix incorrect duplicate count
Steward View:
Cosmetic & Consistency
True / false value properly set when creating a filter involving a boolean
Setup: Add Ultimate tokens to the count of tokens.
Improved “Connect to Delpha” Mechanism
Previously, Delpha asynchronous operations were executed either:
on behalf of the user who connected the org to Delpha, or
via the Salesforce Automated Process user, which introduced several limitations:
No clear audit trail for changes made by Delpha
Inability to assign Permission Set Groups to the Automated Process user
Reduced visibility and control for admins
✅ Now, a dedicated named user is explicitly selected during the setup. All actions performed by Delpha will consistently appear under that user’s name, providing:
Clear traceability of changes
Full permission customization and access control
Alignment with Salesforce best practices for automated systems
When upgrading to version 3.53.x or later from 3.50.x or earlier, you will need to reconnect Delpha as part of the setup process.
Faster Setup
No email activation required: You no longer need to Activate Delpha sending an email to activate core features, simplifying the onboarding process.
Conversations are now automatically imported once Delpha is connected, eliminating the need for manual setup steps.
Locked Records tab
Salesforce Validation Rules set on a record can fire an error and prevent Delpha from updating it. When this happens in a batched process, the whole batch is rejected. The system now more reliably handles these situations to minimize disruption in workflows by catching the error and logging the record ans the error message in the Locked Records table so you can take care of it later. This allows the other members of the batch to be properly updated.
A locked record is ignored by Delpha Data Quality as long as it is listed. You must delete the entry from the Locked Record table to let Delpha process the records.
What's New
Duplicate Settings
You can now provide a description when creating or editing duplicate settings. This description is also used in the Merge Record On Demand feature.
When defining your duplicate population using the Record Filter section, the count of targeted records now appears in the section header.
You can now enable or disable the filtering rules.
If you're a Delpha data steward or admin, you can merge records on demand. To do this:
Open a record.
In the Delpha Score Meter, click the "Duplicate" link.
Click the "Add Duplicate" button.
In the new window, select the second record and choose the duplicate settings.
Confirm the pair and proceed in the Merge Editor.
To select a record, use its ID or name.
Delpha evaluates duplicate settings to see which is suitable for your record, but you can choose any setting even if it doesn't align with your data.
Check Duplicate settings compatibility
In the merge editor, when merging multiple potential pairs, Delpha checks for duplicate settings compatibility. It will alert you of any issues and halt the merge process until resolved.
Fixes
Duplicate Settings
The comment set in an Exact Match rule is now visible in both the Duplicate Data Steward view and the Merge Editor.
Conversations
The conversation list from the Delpha setup has been updated to align with the Score Meter enhancements.
in the
Delpha Integration
section of the
Delpha Setup
app.
Later on, you can reimport or update the conversations following the steps below.
To import conversations for use in Delpha’s data quality workflows:
Open the Delpha Setup app in Salesforce
Navigate to the Import Conversations tab
Select the components you wish to import
Click Import Data to begin the process
For full functionality, it’s recommended to include the following components:
Version 3.56 and onward: the conversation package Data Quality is no longer needed. All the Data Quality topics being handled in the score meter.
Versions 3.60 and onward: most of the data quality and duplicate conversations are managed from the Score meter.
By importing these components you add the Delpha standard conversations to the org. They are all listed in the app Delpha Setup below the Conversation Builder tab.
If you remove the D LinkedIn URL from your record (Account or Contact), all related information is deleted from Delpha fields, now including the logo / picture profile.
Validation Rules
Validation rules can act as a blocker for Data Quality.
When a record is updated by Delpha in the Data Quality process (assessment or recommendation phases) it must follow the validation rules or the update will be blocked.
When it happens in a mass assessment process (Automation, Campaign, Data Steward mass update...) the record is detected and blacklisted in the Delpha Setup > Tab: Locked Records until a Delpha Admin deletes the entry (and takes proper actions to avoid the validation breach again).
When it happens in a single enrichment phase, using the Score Meter, a warning message is displayed to the user so he knows that his record cannot be updated and why.
Establish a secure connection between your Salesforce org and Delpha's backend. This may require authentication and enabling API access. If not connected, follow the prompts to complete the setup.
Review Configuration
Check for any missing or incorrect settings. If issues are detected, Delpha provides step-by-step instructions for remediation to ensure proper functionality.
Ultimate Account Batch Size
Defines how many accounts are processed per batch in the Ultimate Account use case. A green check indicates the correct setting. If not, follow the recommended adjustments to optimize batch processing.
mail.delpha.services: Delpha endpoint used by the Email Insight feature to extract business information from incoming emails.
This feature is off by default and is activated at the Delpha license level (FMA).
Once the upgrade package is installed, open a developer console and run the script below.
If you use a boolean field (e.g., Do Not Assess) to exclude certain records from duplicate detection, make sure to update your detection scope in App Delpha Setup > Duplicate > Detection Settings, under the Record Filter section.
If you skip this configuration, the duplicate analysis will run on allrecords, which may consume more credits than intended.
Set up a Heroku app with the Delpha Data Quality add-on to activate the licences.
Prerequisites
You have access to an Enterprise Account, directly setup with the Heroku Team.
This Enterprise Account is set at Heroku contract signature. The Tenant Owner receives an email from Heroku to activate the account. If you think that you did not receive this email, liaise with your Heroku account manager.
If you are activating Delpha Licenses on a Personal Account, the process is the same. Make sure that billing information is set up (required for paid add-on).
Switch to your Enterprise Account
If the top-left account switcher, displays Personal, click on it and select a Team within your Enterprise Account
If it is the first time you log in your Heroku Enterprise Account, you may have no team created under your Enterprise Account. You should create one for easier access management to the Delpha resources on Heroku.
2 - Create a new app
Go to the tab Apps and click Create new app.
App name: use something meaningful, e.g. your-companyname-delpha-heroku-licence (must be globally unique).
App Owner: select your Team.
Location: choose the closest (e.g., Europe or United States).
Leave “Pipeline” inactive.
Click Create app.
3 - Add the Delpha add-on to the newly created App
Case 1: you are on a Custom Plan
Using Heroku CLI, copy paste the command from your Custom Plan email. It will add Delpha Data Quality add-on to the app you just created and associate it with the Custom Plan.
If Delpha has created a Custom plan for your company, you received an email titled Heroku - You have been invited to install delpha-data-quality:NameOfTheCustomPlan sent by [email protected] with this command line.
If you did not receive this email or if you do not know what to do with it... add Delpha Support with email address [email protected] to your Heroku Team so we can manage the Custom plan activation for you.
To add a Team Member on Heroku, just follow these steps: in your Heroku account: Select your Team / go to the Access tab / click 'Invite User'.
Make sure that the Delpha Support team member has add-on provisioning rights.
Case 2: you are on the Standard plans
Open your new app and go to the Resources tab.
In Add-on Services section, search Delpha Data Quality and select it.
Plan Name: choose Pro.
Click Submit Order Form to provision the add-on.
Final check
You should now see the Service Delpha Data Quality under Add-on Services section with a Plan selected.
Licences are now active for this Heroku app. You can start using Delpha.
Filtering rules are applied from Top to Bottom. Once a rule is applied the filtering processing stops.
You can reorder the rules by drag / drop. We recommend packing the 'Keep' rules at the end of the list, after all the 'Discard' rules.
Direct Match rule Editor (Expression)
With the new visual rule builder, you can:
Define Filtering Rules and Matching Expressions using clear, editable blocks.
Instantly see and adjust logic with intuitive field comparisons (e.g., “if BillingCity is not the same…”).
Remove the risk of syntax errors—no more editing raw JSON!
Enjoy a more organized layout for thresholds, filters, and expressions.
Existing rules have been migrated automatically—just open the new interface to start using it right away.
[LinkedIn]
App Delpha Setup > Data Quality > Contact > LinkedIn
If Active, the system will proceed to replace the existing contact title with the new one from LinkedIn, based on the prioritization logic below.
None, do not update the title: The system fetches the data but does not push any update to the title field.
Only update with the latest Job Title: The system applies the most recent title from the contact's job history.
Only update with the Profile Title: The system applies the title listed on the main LinkedIn profile (often the headline).
Prioritize latest Job Title, else use Profile Title: The system attempts to apply the latest Job Title first. If it's missing or invalid, it defaults to the Profile Title.
Prioritize Profile Title, else use latest Job Title: The system attempts to apply the Profile Title first. If it's missing or invalid, it defaults to the latest Job Title.
[Duplicate]
App Delpha Setup > Data Quality > Duplicate
Duplicate settings have been separated into two distinct categories: Detection Settings and Merge Settings.
Duplicate Setup: New organisation of Duplicate Settings
Duplicate: Update on Auto No Pair Processing
"Auto No pairs" or "Discarded Pairs" can now be managed directly in the merge editor when a custom merge is required. The mass processing feature from the Duplicate Data Steward view remains available.
This feature is exclusively for Delpha Admins and Delpha Data Stewards. It is controlled through a specific permission set. For more information, please contact Delpha.
Fixes
Data Quality: Do not evaluate fields with the status set to "DoNotAssess." The global Do Not Assess feature, which limits the overall record analysis scope, functions properly and remains unaffected.
Data Quality: Fix score computation when applying recommendation from the score meter.
What's New
Duplicate:
New status for a pair that becomes invalid after one of its members is deleted due to another merge. This pair, once removed, is now being monitored.
Status
Processed
Template of Account duplicate settings enhanced.
Enhancing Merge Rule Invalidity Detection
Auto merge job scheduler is now per setting instead of global
Fixes
Duplicate:
Fix pair becoming invalid after merging another pair with related record
Data Steward View
Error when sharing a view being a standard SFDC user profile
Billing & Shipping Address: Mass apply / reject from the data steward view is no longer failing.
Data Quality for Lead: the lead data is an hybrid of contact (Name, Email, LinkedIn profile, phone) and account (Legal Id, LinkedIn, Website, phone) data.
Mass analysis of Leads cannot be done with a Campaign. You must rely on the global settings of Lead Data Quality or on the Steward view 'Analyse' feature
Duplicate
New UI for Merge Field Rules creation
Improve Merge Field Rule issue detection
Enhancement of the merge editor:
better difference display, especially in groupings
displays the same data as the Steward View
Updated Merge Field rules templates
Steward View
Better filtering options for dates
List view menu is now searchable
Data Quality
You can directly select the Object and the field from the main screen. To access the extra filtering capabilities, click on the filter logo.
Duplicate
You can select the Object and the processing state (True or False). To access the extra filtering capabilities, click on the filter logo.
Check my org settings
Ensure your Salesforce org settings are properly set to get the best of Delpha
Starting in early September 2025, Salesforce blocks end users from authorizing or using uninstalled Connected Apps. Only apps that are explicitly installed (or users with special permissions) will work. (Salesforce)
Why Delpha is impacted
Delpha’s Connected App (DelphaAssistantV2) is not installed with the Delpha Assistant package. That makes it an uninstalled application. New users attempting to connect LinkedIn through Delpha will fail until the app is installed (or the user has a bypass permission). ()
Case 1 — The user doing the Delpha connection has the System Administrator profile
In Setup, open Connected Apps OAuth Usage
(Quick Find → “Connected Apps OAuth Usage”). ()
Locate DelphaAssistantV2.
In Actions, click Install (you’ll typically also see Block beside it).
Tip: In this list, an app showing “Install” is not yet installed; one showing “Uninstall” is already installed.
If you are using a Delpha version below 3.62, you will have to apply the same remediation to the connected app DelphaApp
Case 2 — The user doing the Delpha connection isn’t a System Administrator
Create a temporary Permission Set that grants the system permission: Approve Uninstalled Connected Apps. ()
Assign this Permission Set to the user who will perform the install.
Have that user repeat Case 1 to click Install on DelphaAssistantV2.
If you are using a Delpha version below 3.62, you will have to apply the same remediation to the connected app DelphaApp
Notes & validation
Existing users who previously authorized an uninstalled app may continue working in some cases, but new authorizations are blocked unless the app is installed or the user has the new permission.
If “Install” isn’t visible, you may already have the app installed (you’ll see “Uninstall” instead), or you’re not viewing from an account with rights to install.
To double-check, open Connected Apps OAuth Usage, select the app, and review Manage App Policies. ()
Lightning Web Security: activated (Mandatory)
To activate the Lightning Web Security feature in Salesforce:
Access Setup: Click on the gear icon and select "Setup."
Locate Session Settings: Enter "Session Settings" in the Quick Find box.
Enable Lightning Web Security: Scroll to find the "Enable Lightning Web Security" option and check the box.
In newly created Salesforce orgs, the LWS feature is activated by default.
If you do not activate it, the display of the all the LWC components in the org will be too slow to allow a good user experience.
To deactivate the Adopt updated CSP Directives setting in Salesforce:
Access Setup: Click on the gear icon and select "Setup."
Locate Session Settings: Enter "Session Settings" in the Quick Find box.
Adopt updated CSP directives: Scroll to find the "Adopt updated CSP directives" option and uncheck the box.
If this option remain selected, some links provided by Delpha (such as Company logo) will not display properly and will appear as broken links.
Make sure all Delpha users can access User External Credentials Object (Mandatory)
Delpha relies on OAuth for secure authentication that's why, to use Delpha features, the users must be able to access the User External Credentials Object.
To activate this access:
Option 1: For a permission set, under Apps, click Object Settings, and then click User External Credentials.
Option 2: For a profile, scroll to Standard Object Permissions and then select User External Credentials.
()
Accounts as Campaign Members: enabled (Recommended)
Access Setup: Click on the gear icon and select "Setup."
Locate Accounts As Campaign Members: Enter "Accounts As Campaign Members" in the Quick Find box.
Enable Accounts As Campaign Members: Select the option "Enable"
This setting is mandatory if you want to deal with Account Enrichment through Delpha Campaigns.
Contacts to Multiple Accounts: enabled (Recommended)
To activate the feature allowing users to relate a contact to multiple accounts in Salesforce:
Navigate to Setup: Click the gear icon and select "Setup."
Search for Account Settings: Use the Quick Find box to locate "Account Settings."
Enable the Feature: Find the "Contacts to Multiple Accounts" option and check the box to enable it.
This setting is recommended when dealing with contacts belonging to different accounts.
Build 3.56 (04tRl000001D4hxIAC)
From all versions below 3.50.6
Reconnect your Org to Delpha: App Delpha Setup / Section: Delpha Integration / Click the Connect button
Remove Data: Select All conversation package and click on 'Remove Data
Import Data: Based on the use cases you are implementing, select your packages : DuplicateAccountsWD, Duplicate ContactWD, DuplicateLeadWD, DuplicatePersonAccount, LeadVsContact, LinkedInEnrichment, LinkedInEnrichmentAdvanced.
Do not select the package Data Quality as the corresponding .
pairs following these steps
delete current scheduled tasks involving the Apex class: DDQ_DuplicateRecordAutoMerge
from the app Delpha Setup > Duplicate > Initialization, section Auto Merge Scheduler, select the hours to run the auto merge.
This script updates the existing Duplicate Expressions and Filtering rules and ensure their proper display with the .
Open the developer console and apply the following script.
Build 3.70.0
LinkedIn
Since version 3.70.0, the “Connect to LinkedIn” conversation is no longer used to link your LinkedIn account to Delpha. Connection is now handled through a dedicated UI, which automatically collects the required information from the Delpha Connector Chrome extension.
New LinkedIn connection process
Dedicated UI: App Delpha Data Quality > Tab Delpha User Settings
This UI can only be used if the Delpha Connector Chrome extension is installed in your browser.
If the extension is not installed, or if you prefer to manually set a cookie for a user profile, go to Delpha Setup → User Settings. Then edit the user settings you want to update.
Notification on score meter when LinkedIn cookie needs to be set
You can access this page directly, or simply click the alert shown in the Delpha Score component when a configuration issue is detected.
Click the red-dot icon to open the Delpha User Settings page instantly.
Score Meter
Add address recommendations in score meter
When an address recommendation is available, it is displayed to the end user in the Score Meter. The user can accept or reject it directly from the Score Meter, based on the data quality analysis and the recommendation’s confidence score.
Add uniqueness modal to score meter
For any monitored field, if the unicity value is not equal to 1, you can click the message shown in the score meter to display all records involved in the unicity issue.
From that window, you can open and review each record.
Add message/indicator on score meter when record is locked
Ultimate
The comments now explain how the Ultimate Parent was determined and provide the justification for mapping it to an existing account.
If the Ultimate Parent does not already exist in your org, the resolution process will automatically create it (score meter or data steward view).
More info on and
Job Tracker
Configure Delpha’s Job Tracker to monitor job changes and manage contact movements. Match accounts by ID or name, auto-create missing accounts and automate contact updates when job shifts are detected
Job Tracker Configuration in Delpha
The Job Tracker allows Delpha to track job changes for Contacts and enrich your Salesforce data by identifying movements across companies. It includes flexible automation settings for account matching, new account creation, and contact movement management.
Key Configuration Options
Match By
Description: Defines how Delpha matches accounts between Salesforce and LinkedIn
Options:
Key Takeaways for Users:
Use Match By to determine how Salesforce accounts should be matched with LinkedIn data. The recommended option is "By ID then by Name."
Conversations
Import conversation modules in Delpha to enable automated handling of duplicates, enrichment and data quality via Delpha Bot. Easily manage records and boost data health through guided conversations
Importing Conversation Modules for Delpha Bot
To enable conversational features in Delpha, import the relevant modules via the Import Data section and click Import Data. These modules power Delpha Bot’s interactions for handling data quality, duplicates, and enrichment.
Conversation Modules Overview
All imported conversations will appear in the Conversation Builder.
To Remove all conversations from a Conversation Module from your org, just select the module and hit Remove Data
Delpha configuration
Customize Delpha for Salesforce with configuration tools for data quality, deduplication, normalization, and default values—accessible via the Delpha Setup app.
Delpha Configuration Overview
The Delpha Configuration section centralizes all key setup options to tailor the platform to your org’s needs:
First Steps –
Token Usage –
Data Quality - / / – Configure field rules and quality scoring
Duplicates –
Job Tracker –
Default Values –
Duplicates
Configure Delpha to detect and merge duplicate Salesforce records. Learn about Detection Settings (matching logic) and Merge Settings (Golden Record rules).
Delpha Duplicate Management
Overview
Duplicate records in Salesforce fragment your customer view, distort reporting, and confuse AI agents. Delpha's Duplicate Management module allows you to automatically detect, group, and merge redundant records across your Account, Contact, and Lead objects.
As seen in the Delpha Setup menu, this feature is managed through two distinct configuration areas:
Detection Settings: Define how Delpha finds potential duplicates (the matching logic).
Merge Settings: Define how Delpha combines them (the rules for retaining data).
Before you can clean your data, you must define what constitutes a "match."
Logic: Delpha uses advanced AI matching that goes beyond exact text matches. It can identify duplicates despite typos, nicknames (e.g., "Bob" vs. "Robert"), or different formatting.
Scope: You can configure detection rules for specific objects to control sensitivity and performance.
Once duplicates are found, Delpha needs instructions on how to consolidate them into a "Master Record."
Master Record Selection: Rules to determine which record survives (e.g., Oldest Created Date vs. Most Recently Updated vs. Custom Selection rules).
Field Retention: Rules to ensure valuable data (like mobile numbers or titles) is not lost during the merge.
Token Usage
Monitor Delpha token and duplicate credit usage in real time. Track consumption, reset dates, and usage limits to optimize your subscription and avoid overages.
Delpha Token Usage Overview
The Token Usage screen in Delpha gives you full visibility into your current usage of tokens and duplicate credits. It helps you track consumption and manage your subscription resources effectively.
Key Sections:
Usage Reset Date
Shows the next reset date for your token and credit usage (e.g., 2025-02-15), marking the start of a new billing or allocation cycle.
Tokens
Progress Bar: Visual indicator of how many tokens have been used
This page helps you stay in control of your consumption and avoid overages.
Key Takeaways for Users:
Regularly monitor this screen to avoid exceeding your plan’s limits.
Create a “Hello” Conversation on Delpha Home Page
Step-by-step guide to create a “Hello” conversation in Delpha: use Conversation Builder, set display rule, add welcome message, question/answers, and activate.
Step by step procedure
Use Case: Recommend a simple "Hello" conversation on the home page. To build this conversation, follow the below steps.
From the Delpha Setup app, search Conversation Builder and select it.
Click New Conversation on the conversation builder page and enter a name. Append your organization’s name on the trial environment for reference.
To start your conversation, click on the grey ‘ + ’ button. Once you click on the first ‘+’ button, a new node for conversation initialization is created.
Click on Initialization, select previously created display rule (Hello on Home Page) under Display Rule and Save.
Create a Message, add “Hello” as the welcome message and Save.
Create a question, add “How are you doing?” as the display text and Save.
Create two Answers “Good” and “Okay” and Save.
The conversation at this point should look something like this.
Create New nodes under Good and Okay. Add relevant Message to each node
The completed conversation should look like this. Activate the conversation and test the Hello conversation on the Home Page.
You must activate the conversation to make it visible to the Delpha Users. Click on the Activate Button
Delpha Upgrade
Access Delpha’s release notes to upgrade versions using a package ID and virtual install URL. Quickly apply updates in Salesforce by selecting your desired version from the official documentation.
If some scripts must be applied for compatibility topics between versions of Delpha, a side release package and a 'how to' will be provided in the release notes of the target build.
From Delpha Setup app
Open the Delpha Setup app and select the tab Delpha Configuration
In the section First Steps, consider the Review Configuration block
You will find here all the items that needs a manual update with enough information for a quick application.
Once the updates are done, refresh your page. The section should only display green ticks.
Default Values
Set default field values in Delpha for Account, Contact, and Lead creation. Ensure required fields are always filled to prevent Salesforce validation errors and streamline automated record creation.
Default Values Configuration in Delpha
The Default Values screen allows you to predefine field values for Accounts, Contacts, and Leads. This ensures smooth record creation and compliance with Salesforce validation rules by automatically filling in required fields.
Key Configuration Sections
Account Default Values
Action: Use Add New Field to define defaults (e.g., Industry, Account Type, Region)
These settings are critical for the Job History use case as you can be proposed to create a new account or a new contact.
Duplicate
Delpha’s Duplicate menu helps you set up, detect, and clean duplicate Salesforce records with configurable rules and step-by-step remediation workflows.
Duplicate Use Case
The Duplicate section under Use Cases Setup helps you manage the full lifecycle of duplicate detection and remediation:
Create Settings
Enable Long Text fields display for data stewards.
For example, when reviewing a Legal ID, you can now view the current address alongside the Legal ID recommended address for easier decision making.
By ID then by Name(default)
By ID
By Name
Purpose: Ensures accurate enrichment by prioritizing LinkedIn ID with fallback to name matching
Create New Active Accounts
Description: Determines what to do if a contact’s job points to an account not in Salesforce
Options:
Do Nothing
Ask to Create a New Account(default)
Create Automatically New Active Accounts
Create Automatically and Link Contact to Account
Manage Contact Has Moved
Description: Handles how Delpha reacts when a contact has changed companies
Options:
Do Nothing
Ask to Move(default)
Move Automatically When Possible
Move Automatically and Update Title
Ask to Create a New Contact
Create New Contact Automatically When Possible
Configure Create New Active Accounts to define the level of automation for handling new accounts, from manual prompts to full automation.
Adjust Manage Contact Has Moved to decide how contact movements should be handled, with options ranging from no action to automatic updates or contact creation.
Current Usage: Displays used vs. total tokens (e.g., 22,030 / 1,000, or 2,203% usage)
Duplicate Credits
Progress Bar: Visual progress on duplicate credit consumption
Current Usage: Shows used vs. available credits (e.g., 4,528 / 10,000, or 45% usage)
If your token or credit usage is consistently high, consider upgrading your plan using the "Change Plan" button.
Keep track of the usage reset date to plan resource-intensive tasks effectively.
Purpose: Prevent errors during Account creation due to missing required fields
Contact Default Values
Action: Add defaults for fields like Email, Phone, or Job Title
Purpose: Ensure Contact records are created with all mandatory info
Lead Default Values
Action: Set fields such as Lead Source, Status, or Company Name
Purpose: Avoid issues during Lead creation where validation rules apply
Confirm installation. After this, end users can authorize via Delpha normally.
Remove the temporary Permission Set after installation.
Save Your Configuration: Click "Save" to ensure the settings are applied.
Save Your Configuration: Click "Save" to ensure the settings are applied.
Save Your Configuration: Click "Save" to ensure the settings are applied.
List<delpha__DDQ_DuplicateObject__c> dupObjs = [SELECT Id, Name, delpha__Expression__c FROM delpha__DDQ_DuplicateObject__c];
for(delpha__DDQ_DuplicateObject__c dupObj : dupObjs) {
String expRuleStr = dupObj.delpha__Expression__c;
System.debug('Current Rule: ' + expRuleStr);
if(String.isBlank(expRuleStr) || !(expRuleStr.startsWith('[[') && expRuleStr.endsWith(']]'))) { //check here if expRule is in old format (field is not filterable in query)
continue; //skip if already in new format
}
List<Object> expRuleList = (List<Object>) JSON.deserializeUntyped(expRuleStr);
List<Map<String, Object>> newExpRules = new List<Map<String, Object>>();
Integer i = 1;
for(Object rule : expRuleList) {
Map<String,Object> newRule = new Map<String, Object> {
'name' => 'Rule ' + i,
'rules' => rule
};
newExpRules.add(newRule);
i++;
}
System.debug('New Rule: ' + JSON.serialize(newExpRules));
dupObj.delpha__Expression__c = JSON.serialize(newExpRules);
}
Database.update(dupObjs);
Create a New Conversation
Name the conversation
Conversation Initialization
Message
Question
Answer 1
Answer 2
Conversation flow
New Node 1
New Node 2
Final flow
Duplicate Data Steward view
Use Delpha’s Duplicate Data Steward View to detect and merge Salesforce duplicates at scale. Compare fields side by side, trust AI-powered scores, and clean your data with precision and confidence.
Duplicate Data Steward View in Delpha
The Duplicate Data Steward View is designed to help you efficiently identify and merge duplicate records in Salesforce. Using side-by-side comparisons, smart scores, and visual cues, you can quickly take action with confidence.
Main View Overview
The navigation in the Duplicate Steward view is aligned with Salesforce’s native List View experience:
Cog icon: Use the standard cog to manage your views (create, clone, rename, share, select fields to display, delete)
Filter Icon: Create custom views filtering on a Campaign, an Object, a Field and the score.
Pinned and Favorite Lists: Find the best view when you need it
By default, you’ll see the “Recently Viewed” records. To unlock the full power of the Data Steward views, we recommend creating your own custom views—tailored filters, field selection, and sharing options give you complete control over your data review
At the top of the screen, a summary dashboard provides a snapshot of duplicate detection:
Total Potential: Total duplicate groups detected (e.g., 806)
Match Categories:
Exact Match, Likely Match, Possible Match, Unlikely Match
Each row shows:
Account name variations
Field-level identifiers (e.g., Website, LinkedIn)
Match Status (e.g., Potential, Auto Yes)
Side-by-Side Comparison
When you select a row, Delpha displays a detailed comparison view:
Fields from both records are shown side by side
Green dots = selected master value
Red highlights = mismatched fields
This layout helps you quickly decide what data should be kept post-merge.
Available Actions
When rows are selected, the following actions appear at the top:
Accept
Merge using recommended values
Finalize the merge and mark as processed
How to Use
Filter by object (e.g., Account)
Review suggested matches
Click to open detailed comparisons
Merge operation is not applied instantly.
Duplicate that are validated by the Data Steward are processed by batch following a .
Format Delpha Conversation messages
Formatting rules
The Delpha messages can contain more than simple texts. By inserting special tags, you can display dynamic and formatted messages. Tags can be used to modify text style, add icons, add images, add hyperlinks, or insert a variable value. The tags can be used on any message displayed to the user such as:
In the builder, via the message/question/answer display text field.
Formatted message displayed on Delpha
In the apex code, while creating dynamic answers.
Supported Tags
Insert Variables / Record Links
Add Variable
Default or custom variable defined in the builder can be displayed in the bot using the following tags:
Tag: {!VariableName}
Example: {!CurrentRecordId}
Tag: {!VariableName.RelatedObject.FieldName}
Example : {!CurrentRecordId.Contact.Name}
The associated records can be accessed by chaining them as shown above, and you can go on like {!CurrentRecordId.Contact.Account.Name}.
Add Record Link
Tag: [record Id= value]display text[/record]
The value can either be hard-coded or a variable, as shown below:
Example:[record id=0031D00000aIxZtQAK]Andy Young[/record]
Example:[record id={!targetRecordId}]Andy Young[/record]
Tag: [color=value]Colored Text[/color]
The value can either be or hexadecimal code as shown below:
Example: [color=green]Hello World![/color]
Example: [color=#FF0000]Hello World[/color]
Tag: [img]https://img.png[/img][img width=50]
Note: Currently, Salesforce doesn't consider the image height attribute added in the tags. However, the image scaling is respected during rendering.
For security reasons and to avoid vulnerabilities such as XSS, we only support our tags.
Delpha Score Meter
The Delpha Score Meter provides a clear 0–100 quality score based on six dimensions, helping you assess, prioritize, and improve Salesforce data accuracy and reliability at a glance.
What You Can Do in the Score Meter
The Score Meter is your all-in-one command center for data quality. It helps you understand, review, and apply improvements — all in a single, intuitive interface.
1. Understand Your Data Quality at a Glance
See your overall quality score for the selected object (e.g. Accounts, Contacts).
The score is visually represented with a color-coded gauge for quick assessment.
The number of available recommendations is clearly listed.
2. Review Suggestions Inline
View AI-generated recommendations directly from the Score Meter.
Each recommendation includes:
A confidence score based on the data quality dimensions
You access this view clicking on the number of recommendations link or navigating in the Score Meter component using the 'next page' sign (>).
3. Apply Changes — Your Way
Choose how to apply recommendations:
One by one
Bulk select multiple items
What the Score Represents
The score is based on six key data quality dimensions, evaluated field by field:
Dimension
Meaning
Each field within the record is evaluated and color-coded:
Red – Poor
Orange – Warning
Green – Good
Why It Matters
The score helps you:
Quickly assess trustworthiness of a record
Use Cases Setup
Use Delpha’s Use Cases Setup to configure data quality and duplicate detection rules in Salesforce. Tailor each use case to meet your business needs and maintain clean, reliable CRM data.
Use Cases Setup Menu in Delpha
The Use Cases Setup menu allows administrators to configure key operational workflows in Delpha:
Duplicate –
Data Quality – .
These options ensure your Salesforce org is equipped to handle both duplicate management and ongoing data health monitoring.
Data Quality - Account
Customize how Delpha scores and enriches Salesforce Account data. Set automated rules, reassessment schedules, and quality thresholds. Maintain accurate and up-to-date Account records effortlessly.
Data Quality – Account Configuration in Delpha
The Data Quality – Account section lets you define how Delpha evaluates, scores, and improves your Salesforce Account records. It helps ensure data remains complete, consistent, and ready for decision-making.
Data Quality - Contact
Configure how Delpha assesses, cleans, and enriches Salesforce Contact data. Set automatic rules for fields update and track global quality scores to maintain clean, reliable contact records.
Data Quality – Contact Configuration in Delpha
The Data Quality – Contact section in Delpha allows you to configure how contact records are evaluated, cleaned, and scored to maintain high-quality, actionable data.
High-level overview of the conversation builder
Build smart, no-code conversational workflows with Delpha’s Conversation Builder in Salesforce. Guide users with interactive Q&A and automate business logic all within a visual decision-tree interface
The article provides a high-level description of the processes and components involved in building a conversation using the Delpha conversation builder. The Delpha Author can be defined as the user of Salesforce who has access to the Delpha conversation builder.
The author playground mainly comprises:
Delpha Conversation Builder
The conversation builder is a no-code interface used to build new conversations and review existing templates, whereas, the configuration is used to install and uninstall templates.
Data Quality Steward view
The Delpha Data Steward Quality View offers an intuitive interface to monitor, clean, and optimize Salesforce data. Filter by field and object, apply smart recommendations to meet high data quality
Data Steward Quality View in Delpha
The Data Steward Quality View is your command center for reviewing and improving the quality of Account and Contact data in Salesforce. It offers field-level diagnostics and smart recommendations to keep your data clean and campaign-ready.
Total to Be Merged: Duplicates awaiting confirmation
Score (0–100 confidence)
Processed status (True/False)
Reject
Dismiss the suggestion
Keeps records unchanged
Delete
Permanently remove selected records (⚠️ use with caution)
Select records
Click Accept to merge or Reject to skip
scheduled task
Duplicate Data Steward View
Extra information to help the decision
You can expand or collapse details for quick scanning or deep inspection.
Accept all suggestions in one click
Accuracy
Is the value correct (factually or logically)?
Timeliness
Is the information still relevant and up-to-date?
Grey – Not applicable / no data
Decide whether to clean, enrich, or reject the data
Track data quality improvements over time
Completeness
Is the value present or missing?
Validity
Does the data follow the correct format or rules?
Uniqueness
Is this value duplicated elsewhere in the system?
Consistency
Delpha Quality Score
Score Meter - Review and Apply
Is the data coherent across fields or systems?
Key Configuration Areas
Account Assessment Criteria Field
Use a Boolean field to filter which Accounts should be assessed. By default, Delpha uses the Account.D Do Not Assess field, but you can define your own formula-based field for dynamic control.
Account Assessment
Enable this to assess all existing and new Accounts once.
Account Continuous Assessment
Automatically reassess Accounts whenever a monitored field is updated.
Account Reassessment
Schedule recurring quality checks, even if no changes occur.
Options include: Never, Once a Month, Once per Quarter, Twice a Year, Once a Year.
These settings activate Data Quality automation that will consume tokens to ensure optimal data quality in your org.
If you want to control the token consumption we recommend you
to not activate:
Account Continuous Assessment: if you do a mass update impacting a monitored field, the records are supposed to be reassessed.
Account Reassessment: if a record has not been assessed for more than x months, it will be automatically reassessed, according the setting.
Field-Level Automation Settings
Fields monitored for quality scoring:
Website
Billing Address
Shipping Address
Phone
LinkedIn
Legal ID
For each field, you can:
Enable or disable its inclusion in the quality score
Set automatic recommendation rules (trigger + score threshold)
Example – Website
Trigger: When value is Incomplete
Triggering conditions are:
Never
When value is Incomplete
When value is Invalid
When value is inconsistent
Score Threshold: 100
Example – Billing Address
Trigger: Disabled
Score Threshold: (none set)
Ultimate Account Assessment
Evaluates and maintains relationships with ultimate accounts to clarify ownership hierarchies.
Account Global Scoring
Calculates a global quality score across all Accounts daily, giving you a snapshot of overall data health.
Key Takeaways for Users:
Use the Criteria Field to fine-tune which accounts should or should not be assessed.
Use Account Assessment and Account Continuous Assessment to ensure all accounts are reviewed and updated whenever changes occur.
Configure Account Reassessment to address potential data decay by selecting a reassessment schedule.
Select the fields involved in Quality Score definition by toggling them on or off.
Define the automation rules for automatic data quality remediation
For hierarchical relationships, activate Ultimate Account Assessment when needed.
Enable Global Scoring for a daily, holistic view of account quality.
Key Setup Options
Contact Assessment Criteria Field
Filter which Contacts should be assessed using a Boolean field (default: Contact.D Do Not Assess). You can use a formula field for more dynamic control.
Contact Assessment
Enable this to assess all existing and newly created Contacts once.
Contact Continuous Assessment
Automatically reassess Contact records when a monitored fields change.
Contact Reassessment
Schedule regular reviews even if the record hasn't changed.
Options: Never, Once a Month, Once per Quarter, Twice a Year, Once a Year.
These settings activate Data Quality automation that will consume tokens to ensure optimal data quality in your org.
If you want to control the token consumption we recommend you
to not activate:
Contact Continuous Assessment: if you do a mass update impacting a monitored field, the records are supposed to be reassessed.
Contact Reassessment: if a record has not been assessed for more than x months, it will be automatically reassessed, according the setting.
Name Field Normalization
Configure how Delpha standardizes and corrects first and last names:
Apply Recommendation Automatically
Set conditions (e.g., inconsistent value) and minimum score threshold for auto-correction.
First/Last Name Rules
Control letter case, allowed characters, spacing, and removal of unwanted text blocks.
Setting
Description
Letter case
Controls capitalization.
typical: Capitalize All or ALL UPPERCASE
Characters
Specifies accepted alphabets.
Spacing
Handles white spaces.
Text Blocks
Field-Level Automation Settings
Monitored Contact fields:
Name
Email
Phone
Mobile Phone
LinkedIn
Mailing Address
For each field, you can:
Include/exclude it from quality scoring
Define when recommendations are applied
Set a score threshold (default: none)
Examples:
Email
Trigger: When value is Incomplete
Triggering conditions are:
Never
When value is Incomplete
When value is Invalid
When value is inconsistent
Score Threshold: 100
→ Only perfect recommendations are applied automatically
Phone
Trigger: When value is Inconsistent
Normalization enabled (no score threshold needed)
Contact Global Scoring
Automatically computes a daily quality score for all Contact records—giving you a clear view of your data health at a glance.
KeyTakeaways for Users:
Adjust the Criteria Field to include or exclude specific contacts from assessments based on your organizational needs.
Enable Contact Assessment and Continuous Assessment to ensure all contacts are reviewed and updated whenever changes occur.
Configure Contact Reassessment to prevent data decay by setting a reassessment schedule.
Select the fields involved in Quality Score definition by toggling them on or off.
Define the automation rules for automatic data quality remediation
Use Global Scoring to monitor the overall quality of contacts daily.
Salesforce flow module
The flows help in building business automation used in the conversation builder.
The Conversation
The conversation is used to describe a use case. It facilitates interactive communication with Delpha assistant & end-user. Every conversation relies on 3 fundamental steps :
Recommendation
These are based on contextual rules. The author can define when and where a particular conversation (use case) is proposed.
Engagement
Guide the end-user through predefined & Interactive questions & answers.
Execution
Take necessary actions to deal with the corresponding use case efficiently.
Conversation Builder layout
Conversation Builder
The Builder consists of 3 main components:
Toolbox: The toolbox appears on the extreme left and comprises all the author's text and variables while building a conversation. The messages, questions, and answers are filled automatically while variables used in the conversation need to be defined by the author.
Tree Canvas: The canvas appears in the center. It is the working area and gives a high-level description of the conversation or the use case defined in a decision tree. The author can initialize and create nodes here.
Properties: The Properties tab appears on the extreme right and allows the author to define properties, including messages, questions, and answers associated with each component of the decision tree.
Building a Conversation
Initialization
Initialization
In the Initialization window, you can define:
Display Rule: Defines the page(s) on which the author wants the recommendation from the assistant.
Conditions: Define rule-based evaluation criteria on string fields to display the recommendation.
Action Type: Defines evaluation criteria for contextual rules which cannot be defined using conditions. This can be achieved using a Flow, an Apex.
Once the initialization criteria are valid, you can start the conversation The conversation is made up of nodes that are organized as a decision tree.
Node
Conversation Node
A conversation is made up of nodes. It is the key part of the decision tree and can act either as a redistribution point or endpoint of the conversation. Node consists of:
Unique node name
Message(s)
A question and answer(s) to engage the user and take action.
Answer
The Answer is the key component of the node which lets the author define the type of answer and the action to be taken upon the selection of an answer.
Question
The answer can be Static or Dynamic
Answer - Static vs Dynamic
Static answers display fixed answers.
Static Answer
Dynamic answers display a list of dynamic values to select. The values are passed and stored using variables (discussed later).
Dynamic Answer
The Answer can be Single select or Multi-select
Answer - Single or Multiselect
The Selection view enables checkboxes, whereas, comparison view enables radio buttons.
Answer - Single or Multiselect
The Answer can lead to an action or not lead to an action
Action
The toggle switch can activate new action. The new action can be flow, quick action, apex, conversation, or open in a new tab
Variables
Variables are placeholders. They are used for storing and retrieving values in the conversation flow (actions or messages).
Variables in the Toolbox
Variables can be created on the toolbox and can have multiple formats such as string, number, list, object, Id. The action type requires a set of input and output variables. Example: Input variables such as account Id passed to Flow, which processes the information and provides an output(s) used in the conversation. Variables can be classified into two types:
Global variables
Read-only built-in variables to reference common information throughout the conversation. Currently, there are 2 global variables CurrentPage and CurrentRecordId
Conversation variables
User-created variables. They are used to transfer information/data from an action (Flow, Apex) to a conversation or between 2 conversations.
Main Screen Overview
Quality Dashboard with Scores and Recommendations
The navigation in the Data Steward view is aligned with Salesforce’s native List View experience:
Cog icon: Use the standard cog to manage your views (create, clone, rename, share, select fields to display, delete)
Filter Icon: Create custom views filtering on a Campaign, an Object, a Field and the score.
Pinned and Favorite Lists: Find the best view when you need it
By default, you’ll see the “Recently Viewed” records. To unlock the full power of the Data Steward views, we recommend creating your own custom views—tailored filters, field selection, and sharing options give you complete control over your data review
Key Metrics at a Glance
Total records reviewed
Average data quality score
Breakdown of quality indicators: Complete, Valid, Consistent, Accurate
Interactive Tiles
Click once: Filter for records that pass a dimension (tile turns green)
Click again: Show records that need fixing
Click Recommended: Show only records Delpha has suggested improvements for
Row-by-Row Insights
Each record shows:
Current values
Field-level quality indicators
Recommended corrections
Score & status:
Auto Yes – Your value already matches the recommendation
Duplicate Settings now support multiple configurations.
Each configuration runs its own full duplicate assessment, and the results appear separately in the Duplicate Data Steward View.
To create a new Duplicate Setting, click Add New, choose the target object, and decide whether to start from the Delpha default template for detection and merge settings.
You can also duplicate an existing configuration using the Clone
Data Quality - Lead
Customize how Delpha scores, enriches, and validates Salesforce Lead data. Leads act as a hybrid between a Contact and an Account and Delpha monitors both sides using a specialized set of rules.
Data Quality – Lead Configuration in Delpha
The Data Quality – Lead section in the Delpha Setup App allows you to configure how Lead records are evaluated, cleaned, and scored. This ensures that your Sales Development Reps (SDRs) only focus on conversion-ready prospects.
Key Configuration Areas
Lead Assessment Criteria Field: Filter which Leads should be assessed using a Boolean field (default: Lead.Delpha__Do_Not_Assess__c). You can replace this with a custom formula field for dynamic control (e.g., only assess Leads created this year).
Lead Assessment: Enable this to assess all existing and newly created Leads once.
Lead Continuous Assessment: Automatically reassess Lead records whenever a monitored field changes.
Lead Reassessment: Schedule regular reviews even if the record hasn't changed to catch data decay.
Options:Never, Once a Month, Once per Quarter, Twice a Year, Once a Year.
Important Note on Token Consumption: These settings activate automation that consumes tokens. To optimize usage:
Do not activate both Continuous Assessment AND aggressive Reassessment schedules simultaneously unless necessary.
If you perform mass updates on Leads, consider temporarily disabling Continuous Assessment.
Monitored Fields
Delpha monitors specific fields that are critical for Lead Conversion. For each field, you can enable/disable scoring and set automatic recommendation rules.
Profile & Identity (Contact-side)
Name: Validates capitalization and removes suspicious characters.
LinkedIn Contact: Ensures the profile URL is valid.
Contactability (Reachability)
Email: Checks syntax, domain validity, and deliverability.
Phone & Mobile Phone: Standardizes to E.164 format and validates line type.
Address: Standardizes street, city, postal code, and country.
Company Data (Account-side)
Website: Validates domain resolution.
LinkedIn Account: Links the Lead to a corporate entity.
Automatic Recommendation Rules
For each monitored field, you can define Triggering Conditions to automatically apply Delpha's recommendations:
When value is Incomplete: (e.g., fill in a missing Country based on Phone Number).
When value is Invalid: (e.g., fix a typo in an email domain).
When value is Inconsistent: (e.g., match the Company Name to the Website domain).
Lead Global Scoring & Metrics
Delpha calculates a Global Quality Score and engagement metrics daily. These are crucial for sorting Leads in list views.
D Score (0-100): The composite quality score.
Engagement Counters:
D Email Count: Volume of email interactions.
D Call Count: Number of logged calls.
D Event Count: Number of meetings/demos.
Key Takeaways for Users
Hybrid Validation: Delpha checks both the person (Email/Phone) and the company (Website) simultaneously.
Visual Feedback: Use the Score Meter component on the Lead layout to see quality at a glance.
nsures that your Sales Development Reps (SDRs) only focus on conversion-ready prospects.
Key Configuration Areas
Lead Assessment Criteria Field: Filter which Leads should be assessed using a Boolean field (default: Lead.Delpha__Do_Not_Assess__c). You can replace this with a custom formula field for dynamic control (e.g., only assess Leads created this year).
Lead Assessment: Enable this to assess all existing and newly created Leads once.
Lead Continuous Assessment: Automatically reassess Lead records whenever a monitored field changes.
Lead Reassessment: Schedule regular reviews even if the record hasn't changed to catch data decay.
Options:Never, Once a Month, Once per Quarter, Twice a Year, Once a Year.
Important Note on Token Consumption: These settings activate automation that consumes tokens. To optimize usage:
Do not activate both Continuous Assessment AND aggressive Reassessment schedules simultaneously unless necessary.
If you perform mass updates on Leads, consider temporarily disabling Continuous Assessment.
Monitored Fields
Delpha monitors specific fields that are critical for Lead Conversion. For each field, you can enable/disable scoring and set automatic recommendation rules.
Profile & Identity (Contact-side)
Name: Validates capitalization and removes suspicious characters.
LinkedIn Contact: Ensures the profile URL is valid.
Contactability (Reachability)
Email: Checks syntax, domain validity, and deliverability.
Phone & Mobile Phone: Standardizes to E.164 format and validates line type.
Address: Standardizes street, city, postal code, and country.
Company Data (Account-side)
Website: Validates domain resolution.
LinkedIn Account: Links the Lead to a corporate entity.
Automatic Recommendation Rules
For each monitored field, you can define Triggering Conditions to automatically apply Delpha's recommendations:
When value is Incomplete: (e.g., fill in a missing Country based on Phone Number).
When value is Invalid: (e.g., fix a typo in an email domain).
When value is Inconsistent: (e.g., match the Company Name to the Website domain).
Lead Global Scoring & Metrics
Delpha calculates a Global Quality Score and engagement metrics daily. These are crucial for sorting Leads in list views.
D Score (0-100): The composite quality score.
Engagement Counters:
D Email Count: Volume of email interactions.
D Call Count: Number of logged calls.
D Event Count: Number of meetings/demos.
Key Takeaways for Users
Hybrid Validation: Delpha checks both the person (Email/Phone) and the company (Website) simultaneously.
Visual Feedback: Use the D Score Meter component on the Lead layout to see quality at a glance.
Prioritization: Sort your Lead Lists by D Score (Descending) to focus on the highest-quality prospects first.
Automation: Configure Lead Continuous Assessment to catch invalid data immediately upon entry (e.g., from web forms).
Confidence that your data is clean and campaign-ready
Is it aligned across systems or related records?
Accuracy
Is it factually correct?
Timeliness
Is it up-to-date?
option.
Detection Settings
General Settings
Setting
Status/Value
Explanation
Activate Duplicate Detection
Active (Toggle On)
This is the master switch. Duplicate detection is currently enabled and running based on the configured rules.
Run Evaluation
Evaluate (Button)
This is the action button used to manually trigger the duplicate detection process. When clicked, the system scans the records and flags potential duplicates according to the rules and threshold.
Record Filter
The goal of the Record Filter is to limit the scope of the duplicate scan, ensuring the system only checks records that meet specific criteria you define.
1. Filter Logic
Filter Logic: This area allows you to combine multiple individual conditions using boolean operators (AND, OR, NOT) to create complex filtering rules.
e.g., 1 AND (2 OR 3): This example illustrates how the logic works. It means:
Condition 1 must be true, AND
Either Condition 2 OR Condition 3 must be true.
The numbers (1, 2, 3) refer to the individual conditions defined below.
2. Adding Conditions
Search Input Field: You use this to find and select a field name from your database (e.g., Status, Creation Date, Region).
Add Condition Button: Once you select a field, you click this button to add a new condition row.
Condition Rows (implied below the image): Each row will typically let you:
Select a Field (e.g., Record Status).
Select an Operator (e.g., Equals, Is Not Null, Is Greater Than).
Example Use Case:
You might set up a filter to only check records where:
Status Equals Customer
AND Last Modified Date Is Greater Than 90 days ago
This filter ensures the duplicate detection process only runs on active customer records that have been modified recently, saving processing time.
If you use a boolean field (e.g., Do Not Assess) to exclude certain records from duplicate detection, make sure to update your detection scope in App Delpha Setup > Duplicate > Detection Settings, under the Record Filter section.
If you skip this configuration, the duplicate analysis will run on all records, which may consume more credits than intended.
You will have to set up a filter to only check records where:
Do Not Assess Equals False
Filtering Rules
The purpose is to apply specific rules before potential duplicates are reviewed and merged.
The rules are divided into different Rule Types that trigger specific actions:
Rule Type
General Purpose
Action Taken
EXACT MATCH
Used to automatically flag a pair as a certain duplicate if they have an exact match on high-certainty identifiers.
If the specified conditions are met, the system automatically considers them as perfect duplicates. The score is set to the maximum (100), the Status is set to Auto Yes and a specified comment is added. This can prepare them for automatic or priority merging.
DISCARD
Used to prevent a pair of records from being considered duplicates if they match a specific rule.
If the values in the specified field(s) match, the system discards the pair. The Status is set to Auto No. The pair is not proposed to the Delpha User and can only be processed (accepted or rejected) by the Data Steward.
Matching Fields
The Matching Fields configuration, is the core mechanism used by the application to calculate the numerical Duplicate Score for any pair of records. This section defines what fields matter and how much they matter when determining if two records are duplicates.
This screen shows the Advanced Settings for duplicate detection. This section provides granular control over how the system initially screens for duplicates, cleans data for scoring, and manages duplicate creation in real-time.
1. Screening Fields
Purpose is to quickly filter the large pool of records down to a smaller, more relevant subset before the full, weighted matching score calculation (from the "Matching Fields" section) is performed. This significantly improves performance.
The system performs a fast initial match (relying on Name). If the name match is inconclusive, it uses these Screening Fields (e.g., ZIP Code, First Letter of Company Name) to narrow down which records should proceed to the full scoring calculation.
2. Discard Placeholder Values
Purpose is to prevent "dummy" or default values in your records from artificially inflating or skewing the duplicate match score.
You can enter common placeholders that should be ignored into the input field or directly select them from the proposed values.
3. Duplicates at Creation
Define the system's immediate response when a user attempts to create a new record that matches an existing record. This is a crucial real-time defense against data decay.
You choose an option from the dropdown menu to determine what action the system takes:
Option
System Action
Outcome
Allow (Async Detection)
(Default/Least Restrictive) The record is created immediately.
The duplicate check runs after creation, and the record is flagged for later review.
Block (Prevent Creation)
(Most Restrictive) The system performs a real-time check. If an is found, it stops the creation process and forces the user to resolve the conflict (e.g., update the existing record).
Creation is prevented, ensuring no new duplicates enter the system.
This sets the minimum match score for a pair of records to be flagged as potential duplicates. Pairs must have a score $\ge 50$ (on a likely 0-100 scale) to be surfaced for review. A lower number flags more potential duplicates.
KEEP
Used to automatically flag a a pair by adding a comment if they match a specific rule.
When a pair of records meets the criteria of a KEEP rule, the system retains the pair in the set of potential duplicates and add a specific comment.
