Skip to main content

#Sales Cloud Einstein27 personnes en discutent

14 réponses
  1. Hier, à 16:49

    I used my personal email and used personal email as username. This worked. 

0/9000

Hello,

I am currently testing Einstein Activity Capture and have run into two issues.

To provide some context: I synchronized my Outlook mailbox with my Salesforce account (using the same email address) and created an event in Outlook that successfully synced to Salesforce. I subsequently deleted the event from Outlook, but it still appears in Salesforce. How can I ensure it is also deleted from Salesforce? 

 

Secondly, all my Outlook events are syncing to my Salesforce calendar. I only want to sync events where the recipient's address matches an address in Salesforce (Lead, Contact, or Account). How do I configure this setting? 

 

Thanks! 

@* Salesforce Einstein * 

1 réponse
0/9000

We are planing to migrate from Lightning Sync to Einstein Activity Capture.  Is it possible to have some pilot users on EAC but everyone else still omn Lightning Sync ? If yes, how to do it ?  The goal is to test EAC for small group of users before rolling out EAC to everyone.   

1 réponse
  1. Hier, à 09:11

    Hi @Simeon Tzanev  Yes, this is supported. You can enable Einstein Activity Capture only for a pilot group by assigning them the appropriate EAC permission set and adding them to the EAC configuration, while the remaining users continue using Lightning Sync. This lets you validate the setup with a small group before migrating everyone. Just avoid having the same users enabled for both Lightning Sync and EAC at the same time, as they shouldn't run concurrently for a single user. 

0/9000

Hi all!    We've been using the Einstein data-driven attribution model to capture campaign influence from our activities. This works great for us usually because it associates a nurtured journey to second named contacts at the same account who actually have the opportunity attached.    Today I noticed that a campaign with one form and no completions had received campaign influence because of a form view. We already exclude email opens by way of attaching these to an excluded campaign, but ideally we don't want to do the same for forms.    Am I missing something? Is there a way to exclude particular activity types from creating campaign influence records? I'm already filtering the auto-association settings by saying the member status must be Responded and the first associated date cannot be blank, but records are still showing in our reporting.    I've looked at filtering this on the reporting level too, but since Campaign Influence to Campaign to Campaign Member is a many to one to many relationship, this hasn't been done. It seems as though the next option is to develop a custom Ajax procedure to create the fields to exclude this.    Has anyone found a simpler way? Is there something I'm missing?    Cheers all,  Alex   

1 réponse
0/9000

Hello,

Following a recent reconfiguration of Einstein Activity Capture synchronization, one of our users has been experiencing a continuous and abnormal increase in the number of contacts stored in Outlook.

The Outlook account now contains more than 40,000 contacts, and the number continues to grow. This issue also affects the user's mobile device because Outlook contacts are synchronized to the phone, making contact management and Outlook usage very difficult.

Observations:

  • The user is assigned the Einstein Activity Capture Permission Set.
  • The contacts created in Outlook appear with the "Sync Salesforce" category.
  • The number of contacts keeps increasing, even though no mass contact creation is expected.
  • Cleanup is extremely difficult due to the volume of contacts.
  • Exporting contacts from Outlook only retrieves around 10,000 contacts, while the total exceeds 40,000.

Has anyone encountered a similar issue with Einstein Activity Capture or Outlook contact synchronization? If so, what was the root cause and how did you resolve it?

Any advice or recommendations would be greatly appreciated. Thank you!    

1 réponse
0/9000

Hi Everyone,

I’m looking for some guidance from anyone who has implemented a similar product recommendation use case.

Here’s the scenario:

  • We have historical Sales Orders containing Products + Value Added Products (VAPs).
  • Currently, this history resides in Snowflake, although I’m also considering a scenario where the complete sales interaction history resides in Salesforce.
  • The goal is to provide intelligent product recommendations during the sales process (for example, complementary products, commonly purchased together, next best product, etc.).

I’m intentionally keeping the question open because I’m interested in understanding how others have approached this problem rather than focusing on a specific Salesforce feature or product.

Some questions I have:

  • If your sales history is already in Salesforce, how did you model the recommendation solution?
  • Did you use Salesforce-native capabilities, Data Cloud, custom Apex/LWC logic, external ML services, or another approach?
  • If your interaction history resides outside Salesforce (Snowflake, data warehouse, etc.), did you build recommendations there or bring the data into Salesforce?
  • What kind of interaction data proved most valuable (orders, quotes, opportunities, views, carts, etc.)?
  • Were there any lessons learned around data preparation, model quality, or keeping recommendations relevant?

I’d really appreciate hearing from anyone who has implemented a similar B2B product recommendation solution, regardless of the technology stack used.

Thanks in advance!

 

@* Salesforce Commerce for B2B & D2C * 

2 réponses
0/9000

Hi Everyone,

I’m currently evaluating Einstein Recommendation Builder and have been trying to understand how the recommendation model is actually built and evaluated behind the scenes.My understanding is that Recommendation Builder takes:

  • Recommended Item Object
  • Interaction Object
  • Interaction Type (View, Purchase, Rating, etc.)

However, I’m struggling to understand what happens after that.

  1. How is the model evaluated? --- for accuracy and precision
  2. What is Salesforce’s recommendation for validating recommendation quality before moving into production? -- is there any smart way to do it.

Recommendation Builder doesn’t appear to expose any similar metrics. 

 

1 réponse
  1. 30 juin, 23:59

    @Mariappan Perumal

     

    Recommendation Builder does not expose the classic precision, recall, and F1 metrics you'd get from a hand-rolled recommender or even from Prediction Builder's scorecard. That's not you missing a screen. The model evaluation in Recommendation Builder is deliberately abstracted, and the single metric Salesforce surfaces is predicted lift, not accuracy or precision. So the honest framing for your evaluation is: you validate Recommendation Builder differently than you'd validate a model where you control the train/test split. 

     

    Here's what's actually happening behind the scenes and how to work with it.

    After you select your Recommended Item object, Interaction object, and Interaction types, the build runs two things you can act on. First the Data Checker, then the scorecard. The Data Checker validates whether each object has enough data to build a meaningful recommendation before it lets you proceed, and the scorecard then gives you the model's quality, top predictive factors, and predicted lift. 

     

    Predicted lift is the metric to anchor on, and it's more useful than it first appears. Lift tells you how much better your recommendations are than simply suggesting the most popular items, and the recommended workflow is to use the scorecard to measure that, and if you're not happy, fine-tune the segments and settings and rebuild until performance is acceptable. So lift is your relative quality signal: a model that can't beat a popularity baseline by a meaningful margin isn't worth deploying.

     

    On your two specific questions.

    For "how is the model evaluated for accuracy and precision," the answer is that Salesforce evaluates it internally against held-out interaction data but does not expose precision/recall to you.  

     

    For "what's the smart way to validate quality before production," since the tool won't give you offline precision, build your own holdout validation around it. This is the approach I'd take and the one worth bringing to a client:

    Reserve a time-based holdout of interaction data that you exclude from the build window. Generate recommendations from the model, then check how many of those recommended items the same users actually interacted with in your held-out period. That gives you a real precision and hit-rate measurement that the tool doesn't surface, computed against ground truth you controlled. It's the closest thing to the precision/recall you're asking for, and it's defensible. 

     

    Run a champion/challenger in production rather than a hard cutover. Deploy the recommendation inside a Next Best Action strategy and A/B test it against either the existing logic or a popularity baseline, measuring actual click-through or conversion lift on live traffic. Recommendations deploy by adding business rules in an NBA strategy, which is also where you control and gate them, so the strategy layer is your natural place to split traffic and measure real-world precision through behavior rather than an offline score. 

    - SP 

0/9000

We are having to adopt EAC but the default email linking is proving a risk for our confidential information. We don't want to block email domains as some emails will be ok to link but some won't.   Does anyone have any experience managing email privacy and security (as EAC is blind to Outlook security settings)    

1 réponse
  1. Forum Ambassador Vuk Stajic (MVRK Inc.)
    30 juin, 20:54

    I think you may have hit a limitation here. 

     

    Why are you "having to adopt"? 

     

    What is the actual business objective :)

0/9000

Hello Community,

We have enabled Einstein Activity Capture in our Salesforce org, and the Microsoft 365 connection appears to be working correctly. Users have successfully logged in with their email accounts, and the synchronization is active.

Currently, when a user sends an email from Salesforce, it is sent successfully and appears in the corresponding mailbox. However, we are experiencing an issue with calendar event synchronization.

Even though event synchronization is enabled in the Einstein Activity Capture settings, Outlook/Microsoft Teams events are not appearing in the Salesforce calendar for users.

Besides verifying that event synchronization is enabled, what other settings, permissions, or configurations should we review? Are there any known limitations or common issues that could prevent Outlook/Teams events from being displayed in Salesforce?

Any guidance would be greatly appreciated.

Thank you! 

Not view events@* Known Issues * @* Customer Success * 

1 réponse
  1. 12 juin, 01:04
    Hello Juan! This may sound silly but my users once couldn't saw events because the calendar did not have checked the proper checkbox and they only saw the events they created and not the ones from Google calendar (we use Google instead of Microsoft)
0/9000

Hi Everyone,   

We are using Einstein Activity Capture (EAC) in Salesforce,  users authenticate their Outlook accounts  from Salesforce.   

I’m trying to identify which users have successfully connected/authenticated their Outlook. The goal is to build a report for users who have their Outlook connected.   

I’ve explored a few options (including standard objects and APIs), but haven’t been able to find a reliable way to retrieve user-level connection status.   

Has anyone implemented a solution for this?

  •  Is there any SOQL query or API that exposes this information? 
  •  If not, what’s the best alternative approach to track/report on this? 

Any guidance or best practices would be really helpful.   

Thanks in advance!    

2 réponses
  1. 9 juin, 20:39

    We just started piloting EAC and need to know which Users had established a connection to Office 365.  We found the following query that tells us if a User has established a connection, needs to reconnect, or has no connection which is blank.  Users will need to be assigned a connection in order to appear. 

     

    SELECT Id, ConnectivityStatus, EmailAddress FROM ActivityUsrConnectionStatus 

0/9000