To use the Ask Elements and Recommendation engine for user stories, you first need to index your Org's metadata in our semantic database.
Article outline
Prerequisites
The ElementsGPT capability must be enabled for your space, as part of a trial or as a paid add-on. Contact your Success or Account Representative for more information.
Your Space must have an Enterprise (ENT) or Consulting plan license
A Salesforce org connected and Org Model sync completed
ElementsGPT users must have Org Model manager permission
Considerations for indexing
Which Orgs should we use for ElementsGPT?
You should focus on one Org to be used for analysis purposes. This can be a Production Org, a Full-Sandbox Org (UAT) or any Sandbox.
If you choose a standard Sandbox, refresh it regularly to ensure the metadata configuration is as up-to-date as possible.
Whenever you refresh the Sandbox, remember to refresh the connection in Elements as well.
Indexing the Org metadata
Open your Org model.
Click on the 'Ask Elements' button. You will be prompted to create an index in our semantic database. The process will take 1-2 minutes.
That's it! The index will automatically refresh every week on its own.
What is indexed?
Currently, we are indexing the following metadata types with their names, API names and descriptions:
Apex classes
Apex triggers
Flows
Objects
Record types
Fields
Profiles
Permission sets
Permission set groups
Validation rules
We plan to index more metadata and provide information about them in the future iterations.
How to re-index?
Once the metadata index is generated in the semantic database, it is then scheduled to refresh once per week. This ensures that ElementsGPT has up-to-date knowledge of your Org configuration and documentation.
If you want to re-index more often than that, then open Ask Elements and trigger manual re-indexing via the Gear icon, as shown below.
