Hi everyone, 

I'm Shraddha Nakra, the PM for Salesforce Connect at Salesforce, based in Portland, Oregon. I moved over from the Metadata platform team earlier this year, and I've spent the last few months talking to customers, digging into support cases, and figuring out where Connect is earning its keep and where it's falling short.

Quick note: I'll be at Dreamforce (Sep 14–17) presenting a session on building custom Salesforce Connect adapters using Claude Code and Salesforce's Headless MCP platform. If you're attending, I'd love to meet in person.  Session Link

.

The reason I'm posting: I want to hear from this group before I finalize roadmap priorities for next year.

Two things I'm genuinely trying to understand: 

1. Where are the gaps in adapter coverage? 

What system have you tried to connect to where none of our built-in adapters like OData, Snowflake, DynamoDB, GraphQL, CrossOrg fit? Are you writing custom Apex adapters today? If so, what was the hardest part? Is it the setup, the maintenance or something else?

Is it reliability like better error visibility, clearer limits, more predictable behavior under load? Or is it the developer experience with easier setup, faster path from "I have an API" to "it's live in Salesforce"?

I read every reply. If you want to go deeper, DMs are open. 

Thanks, 

Shraddha