Feed

Been reading about how everyone is having trouble with the "Add and Map a Formula Field" module; I am, too. This is frustrating. I've tried creating new Dev playgrounds as well. I've followed every step exactly as instructed. 

not able to give edit permission to created by and last modified by  

Hi everyone, trailblazers! I'm Italian and I'm preparing for the Marketing Cloud Administrator Email Specialist. I hope to take it by next month. Is there anyone who wants to join me?

Hi folks, 

I've finally decided it’s time to move our tiny-but-growing org to package-based development. It’s 2025 already and although I know plenty of orgs still living in the Happy Soup, I still feel late to the party.

The plan is not a big-bang refactor. I’ve been warned off that enough times that it's burned into my brain. I want to start small, but I feel like starting small still needs a clear map, otherwise I'll end up shuffling fields and classes between packages every other week. 

After disappearing for a couple of days, I grouped every piece of metadata I could find, and sketched the diagram that's attached. It's very first-draft, sketched only by me without anyone to provide their thoughts yet, so poke as many holes in it as you like. 

What you'll see in the picture

• Unpackaged layer: stuff that really depends on everything that can't/shouldn't be packaged (layouts, permission sets + groups, lightning record pages, etc.)
• Business layer: four core packages (Sales, Service, Finance, Marketing) plus where a new package, such as a billing-system integration (not created yet), would go.
• Extension layer: add-ons like sbqq-ext, blng-ext, millio-ext, and a sf-test-data-generation package that spits out test/demo data. For instance, sbqq-ext would hold SBQQ.QuoteModel, QuoteCalculatorQueueable, etc. Also, some custom fields...?
• Structural layer: one big common package. It holds the generic SObjectBuilder, base triggers, and anything that should be usable only knowing our objects, not their fields.
• Managed layer: the third-party managed packages we depend on (sbqq, sbaa, blng, millio, pi, cldseq.)
• Utilities layer: FFLib, Trigger Actions, force-di, Nebula Logger, Apex Libra, etc.

A few things I’m wrestling with

1. Does it even make any sense? Is it on the right path, over engineering, both?  
   I'm coming up with these layers and packages with no real prior experience. Do they make sense? What metadata components would you say go or don't go into them? 
    

2. Fields and Trigger Handlers:  
   I envision that we would put fields like SBQQ__Quote__c.TextField__c in sbqq-ext. But it's never so clear cut like that. What if we wanted a field on quote (just an example) looking up to blng__LegalEntity__c? I guess, since that might relate to a requirement for finance dept., that we'd put that in the Finance Core. Is that the correct thinking? Also, sf-test-data-generation is a pretty standalone package, and introduces custom objects and triggers for it. While all our triggers are in Common, does it make sense to have a custom object definition, triggers, etc. (specific to sf-test-data-generation) in that package? 
    

3. Also, where should the sf-test-data-generation package live?  
   It isn't really "business logic", but it also isn't as low-level as
   utilities. I dropped it in the extension layer for now. Does that placement make sense? 
    

4. Business vs. Extension: are these really two separate layers?  
   Should there even be something like an "extension layer"? The names were, again, just invented on the spot. In practice they both hold code that depend on managed packages. I know that for sbqq-ext, we would have CPQ API classes in there, and maybe some other relevant queueables. But would we also store some custom fields there? Similar question to #2.

I've nearly a decade of building on Salesforce on my belt, but this experience has made me realize that although I can ship features fast, I don't really know how to design an application, one that can grow without turning to spaghetti. This packaging exercise is forcing me to think quite a bit deeper, and so I could really use a few extra heads. So if you've walked this road, even part of it, I'd love any feedback, nit-picks, or "you're totally off here" comments. 

Thanks a ton in advance! 

Hi everyone 👋 

 I'm currently exploring how to implement

🔹 1. Different surveys per client and campaign

We work with multiple clients, and each client can have several campaigns. The survey questions must vary depending on the

combination. 

Or is there a way to use a single survey and show/hide questions dynamically based on

🔹 2. Display Case information before survey questions

Before the customer starts answering the survey, we want to display some key information from the

, like a header or pre-filled section. 

For example:

• Case number
• Service ID
• Service date
• Service name
• Campaign
• Advisor or agent who handled the case

Is it possible to show these values at the top of the survey (as non-editable context)? Ideally, these fields should be visible for reference before answering.

🔹 3. Percentage-based survey sending with agent control

Survey sending is

• Some clients require surveys for 100% of closed cases.
• Others only need surveys for 40% or 10% of closed cases.
• Additionally, the service agent must have the ability to manually choose whether to send the survey (via button or action). 
  However, the survey
  must always be linked to the Case, and it must include key case fields when sent.

How would you recommend implementing this logic?

Would a

I'm doing some exploration and learning around this functionality, and I’d really appreciate any ideas or guidance from the community. 🙏 Thank you in advance! 

I'm getting an unhandled exception error. Please reference ID: VXDZSBHI. 

It seems to be a common error people are getting. Has there been answer yet? I haven't found it. 

#Trailhead Challenges

Hey 

I am doing a mass data update to our account teams and am trying to create one report that shows only what was taken from a sales rep and what was moved to a sales rep.  

The original report with no user filters works fine. So when I add the filter of new value $USER, the results are 0. 

What am I doing wrong? Does anyone have other ideas of how to pull that data?