Insert, upsert, update, and delete records in Salesforce through the REST API, in throttled single-record or batched write modes, with per-record success and failure output on every run.
Salesforce is the system of record for sales, service, and customer data at most B2B companies. Sales operations teams push lead and opportunity updates into it. Customer success teams sync account health scores. Marketing operations teams write campaign membership and engagement data back into it so reps can see what prospects have done before a call.
The problem shows up the moment those writes need to be fast, small, or frequent. Salesforce's Bulk API is built for large batch loads on a schedule, not for the steady trickle of smaller updates many teams actually need, like a lead score changing, a support ticket closing, or a campaign response arriving in near real time. Bulk jobs also make it hard to know exactly which rows succeeded and which failed without building custom monitoring on top of them.
That's the problem the new Integrate.io Salesforce REST API Destination connector solves.
What the connector does
The Integrate.io Salesforce REST API Destination connector writes records directly to Salesforce through its REST API, supporting insert, upsert, update, and delete operations, in either throttled single-record mode or batched write mode. The full Integrate.io platform sits in between, which means:
-
Scheduled or event-driven writes: Push records to Salesforce on any cadence, from near real-time triggers to hourly, daily, or custom schedules, without waiting for a batch window.
-
Transformation built in: Map source fields to Salesforce object schemas, flatten nested structures, apply dedupe logic, and handle any custom transformation in low-code or Python before a record ever reaches Salesforce.
-
Any source: Pull from warehouses, SaaS APIs, databases, or files, and write the transformed result straight into Salesforce standard or custom objects.
-
Production-grade operations: Per-record success and failure CSV output on every run, standard and customer-managed OAuth support, retry logic, full run history, and alerting, the same infrastructure running millions of business-critical pipelines for our customers today.
It's not a glorified data loader. It's a real pipeline, built for teams who need Salesforce writes to be as fast, granular, and observable as any other production data destination.
How the connector works
The connector writes to any standard or custom Salesforce object, like Account, Contact, or a custom object such as Invoice, and supports four operations: insert, update, upsert, and delete. Update and delete match on the Salesforce record ID, while upsert matches on an external ID field you define, so you can write records without looking up Salesforce IDs first.
Write mode controls how records reach Salesforce. Single per record mode sends one HTTP request per record, with a configurable request sleep interval from 0 to 60,000 milliseconds, so you can throttle request volume to stay inside your org's API limits. Batch mode groups records into a single request using Salesforce's composite sObjects Collections API, up to 200 records per call, the maximum Salesforce allows.
Error handling is built into the destination itself. You can set a maximum failed records threshold so a job fails once cumulative failures cross it, or leave it open to push through every record regardless of individual failures. Every run can output a CSV of failed records with the Salesforce error message attached, and a separate CSV of successful records including the Salesforce ID returned for each insert or upsert, to Amazon S3, Google Cloud Storage, SFTP, FTPS, Azure Blob Storage, or Google Drive.
Beyond the basics: use cases for any Salesforce customer
Here are several patterns Salesforce customers are increasingly asking about:
-
Centralize account health scores in Salesforce: Warehouse teams computing churn risk, usage trends, or NPS-derived health scores can write them straight into Salesforce accounts and opportunities, eliminating one-off scripts or manual CSV imports run by an analyst each week.
-
Trigger downstream workflows on record updates: Writing updated fields into Salesforce through upsert mode can kick off existing Salesforce workflows, like a case escalation or a task assignment, without changing anything inside Salesforce itself.
-
Feed AI and analytics models back into the CRM: Propensity-to-buy scores, next-best-action recommendations, or churn probabilities generated by external models can be written back into Salesforce fields reps already use. Any data Integrate.io ingests can be transformed and prepared for downstream AI workflows before it's written back.
-
Power customer-facing or partner-facing dashboards: Data enriched outside Salesforce, like usage metrics or support sentiment, can be written into Salesforce objects that feed dashboards built for partners or customer-facing teams.
-
Replace fragile manual exports: Many teams still have someone exporting a CSV and importing it into Salesforce on a recurring basis. The connector replaces that with a pipeline that runs continuously, with full observability and no babysitting.
Why fixed-fee matters for this use case
Most data integration tools charge per row, per connector, or per monthly active row, and Salesforce writes are exactly the kind of workload that punishes that model. Frequent, smaller writes, like the throttled single-record mode this connector supports, can rack up costs quickly on a per-row pricing plan, discouraging teams from writing data back as often as the business actually needs.
Integrate.io charges a fixed fee regardless of row volume or write frequency. A team pushing tens of thousands of small, near real-time updates into Salesforce every day pays the same as a team running a single large nightly batch. This matters most for revenue operations teams, agencies managing Salesforce for multiple clients, and any company where Salesforce write volume is expected to grow.
Getting started
The Salesforce REST API Destination connector is available now to all Integrate.io customers.
The fastest way to get started is to book a call. A solutions engineer will walk through your Salesforce objects, write modes, and ingestion cadence, and help you map out the pipeline before you build anything.
Every new connector setup includes a two-week free pilot, where our team builds the initial pipeline alongside you. No DIY trial, no figuring it out alone, just a working solution by the end of the pilot.
To get started, you'll need:
- A Salesforce account with either standard or customer-managed OAuth credentials configured
- The Salesforce objects and fields you need to write to, along with your preferred write mode (insert, upsert, update, or delete)
- A clear picture of the source system your Salesforce data should come from
Have a Salesforce use case you're trying to unblock? Schedule a time to speak with us so that we can show you how we can help.