Alright, I need some real talk from people who've been in the trenches. We're six weeks into a Salesforce to HubSpot migration for a ~300 person sales team. Our chosen "specialist" vendor just emailed us that they're filing for bankruptcy. Effective immediately. All work has stopped, and we're locked out of their migration environment.
We're left with:
- A half-completed data mapping spreadsheet (mostly source-to-target field lists).
- Partially transformed historical data in their now-inaccessible staging DB.
- Zero documentation on their transformation logic or error handling.
- A hard deadline from Finance to be off Salesforce by end of quarter.
The obvious options seem to be:
1. **Scramble for a new vendor.** Risk repeating discovery, blowing the budget, and missing the deadline.
2. **Bring it in-house with middleware.** I'm leaning towards this, but need to gauge the effort. We have a Workato license, and I've done simpler migrations.
3. **Manual CSV exports/imports.** A non-starter for data integrity and volume, but the "keep it simple" crowd is suggesting it.
My main questions for those who've hit a similar wall:
* Has anyone successfully recovered a stranded migration by rebuilding the pipeline in a tool like Workato or Celigo? How long did it *really* take you, starting from incomplete maps?
* What are the critical data objects you'd prioritize to get a sales team minimally functional in the new CRM? (We're thinking Accounts, Contacts, Opportunities, Activities).
* Any gotchas when trying to reverse-engineer transformation rules from a broken vendor? Our main pain point is legacy custom objects in Salesforce that need to be flattened into HubSpot's schema.
I'm pushing for the middleware path. Point-to-point via CSV is how you get 3am phone calls about duplicate records for the next two years.
```json
// Example of the mapping debris we have:
{
"source_object": "Salesforce.Opportunity",
"target_object": "HubSpot.Deal",
"field_maps": [
{
"source": "Name",
"target": "DealName",
"status": "mapped"
},
{
"source": "Amount",
"target": "Amount",
"status": "mapped"
},
{
"source": "Custom_Pipeline_Stage__c",
"target": null,
"status": "pending",
"notes": "Needs transformation to HubSpot stage code"
}
]
}
```
What would you do next?
Integration is not a project, it's a lifestyle.
Scramble for a vendor and you'll pay a 100% panic tax. They'll smell your deadline.
With Workato and a half-done map, bringing it in-house is your only budget-aware move. You've already done the hardest part: agreeing on what fields map where. The vendor's "transformation logic" was probably a dozen simple formulas. Rebuild it.
Don't touch CSV imports. The time you'd spend cleaning errors will sink you. Treat this like a corrupted backup: your source-of-truth is still Salesforce. Start the extraction again from there, using your mapping spreadsheet as the spec.
You need to parallelize. One person builds the Workato pipelines. Another starts testing with a subset of data. A third lines up the business users for UAT the second you have a clean batch. This is a three-week sprint, not six.
cost per transaction is the only metric