Salesforce API Testing
Salesforce API testing validates the requests and responses that move data in and out of your org through the REST, SOAP and Bulk APIs — checking payload structure, status codes, field-level correctness, error handling and partial-failure behavior at the contract level, below the user interface. It is how integrations are proven to behave before a real system relies on them.
SyntraFlow is designed to exercise Salesforce API calls directly and assert on what comes back, so integration defects surface at the interface rather than in production.
Why the API is where integrations break
Every external system that talks to Salesforce — an ERP, a middleware layer, a customer portal, a data warehouse — does so through an API, not the browser. When those calls go wrong, the failure is invisible to anyone clicking through the UI: a field silently truncated, a null where a value was expected, a status code that says success while the record never persisted. The interface looks fine while the integration quietly corrupts data.
API testing targets the contract itself. It is distinct from UI testing, which drives the Lightning experience a human sees. The most common Salesforce API failures it catches include:
- •
Payload shape drift — a field is renamed, removed or retyped in the org, and the caller's JSON or XML no longer maps, so records fail to insert or map to the wrong field.
- •
Wrong or misleading status codes — a call returns 200 or 201 but a downstream trigger, validation rule or Flow silently rejected the write.
- •
Unhandled error responses — a 400, 401 or 500 comes back with a Salesforce error code the integration never anticipated, and the caller does not retry or roll back.
- •
Silent partial failures — a bulk job of thousands of records reports success overall while individual rows failed, so counts diverge and no one notices until reconciliation.
These are contract defects, and they belong to the Salesforce integration testing discipline. This page owns the API layer specifically; for the systems those APIs connect — middleware and back-office endpoints — see the parent pillar.
A Salesforce API test, step by step
SyntraFlow can be configured to drive Salesforce API calls as part of a scenario and assert on every layer of the response, not just the surface result.
Authenticate and establish the session
A connected app or OAuth flow issues a token, and the test confirms the integration user has exactly the access the call requires before any data moves.
Send a known request
A REST, SOAP or Bulk request is issued with a defined payload — a create, update, upsert, query or composite operation — using controlled test data.
Assert on the status code
The HTTP or SOAP status is checked against what the operation should return — success, created, bad request, unauthorized — so a wrong code is caught immediately.
Validate the response body field by field
Returned IDs, field values, types and structure are asserted against expectations, so a truncated, null or mistyped value fails the test rather than slipping through.
Confirm the record actually persisted
A follow-up query verifies the write survived triggers, validation rules and Flows — proving the API call changed the org, not just returned a hopeful status.
Exercise the error and partial-failure paths
Deliberately invalid payloads and oversized bulk jobs confirm the correct error codes come back and that per-row results are read, not assumed.
What to validate by API type
Salesforce exposes several APIs, each suited to a different job and each with its own failure modes. SyntraFlow is designed to test all three, and the deeper siblings below go further on each one.
| API type | Typical use | What to validate | Key failure modes |
|---|---|---|---|
| REST API | Record CRUD, queries, composite and sub-request calls from apps and services. | JSON payload shape, HTTP status codes, field values, composite ordering, error bodies. | Field mapping drift, misleading 2xx codes, unhandled 4xx/5xx responses. |
| SOAP API | Enterprise and partner WSDL integrations, strongly-typed legacy callers. | XML structure against the WSDL, fault elements, typed fields, session handling. | Schema mismatch after metadata change, unhandled SOAP faults, type coercion errors. |
| Bulk API | High-volume load, extract and sync jobs of thousands to millions of records. | Job and batch status, success and error counts, per-row results, throughput. | Silent partial failure, count divergence, batch timeouts, unread error rows. |
Go deeper on each in the dedicated REST API testing, SOAP API testing and Bulk API testing pages.
API testing versus UI testing
Both matter, but they answer different questions. Confusing the two leaves gaps: a green UI suite can sit on top of an integration that is quietly losing data.
API testing asks: is the contract correct?
It sends requests directly to REST, SOAP or Bulk endpoints and asserts on payloads, status codes and persisted state. It is fast, deterministic and blind to layout, so it catches data-level defects a screen would never reveal — and it is exactly how machine-to-machine integrations actually communicate.
UI testing asks: does the experience work?
It drives the Lightning interface a person uses — page layouts, buttons, Flows, validation messages. It is essential for human-facing journeys but slower, more brittle, and unable to see the raw payloads an integration exchanges. It complements API testing rather than replacing it.
What SyntraFlow is designed to do at the API layer
These capabilities are available for demonstration and proof-of-concept validation against your org's APIs.
Direct request execution
REST, SOAP and Bulk calls can be issued with defined payloads and authenticated sessions as steps inside a larger test scenario.
Field-level response assertions
Returned values, types and structure can be asserted individually, so a single truncated or null field fails the test rather than passing quietly.
Status-code verification
Expected HTTP and SOAP statuses are checked for success, error and edge cases, catching misleading codes that mask a rejected write.
Composite and multi-step calls
Composite REST operations and chained requests can be validated as a unit, confirming ordering and that dependent sub-requests resolve correctly.
Bulk result reconciliation
Job and batch results can be read row by row, so success and error counts are reconciled and partial failures are reported, not swallowed.
Persistence check-back
A follow-up query can confirm the record truly landed after triggers, validation rules and Flows — proving the call changed the org.
The same call has two ends
A Salesforce API call rarely exists for its own sake — it is one side of a conversation with an ERP, HCM or finance system. SyntraFlow is Oracle-native and expanding into Salesforce, so it is built to test both ends of that conversation: assert on the Salesforce response, then confirm the data arrived correctly in Oracle, SAP, Workday or NetSuite. That whole-transaction view is where a payload-level check becomes real integration confidence.
See the back-office depth behind this in the Oracle ERP testing tool, or run these API checks inside a broader suite with Salesforce test automation.
What API testing makes possible
Defects caught at the contract
Payload and status problems are found where machines talk to machines, before a downstream system consumes bad data.
Confidence in high-volume loads
Bulk jobs can be trusted because per-row results are reconciled, so a partial failure is a reported event rather than a silent data gap.
Faster, steadier feedback
API checks run quickly and deterministically, giving integration teams a stable signal that does not break on a layout change.
Salesforce API testing FAQs
What is Salesforce API testing?
Salesforce API testing validates the requests and responses that move data through the REST, SOAP and Bulk APIs, below the user interface. It checks payload structure, HTTP or SOAP status codes, field-level values, error responses and, for bulk jobs, per-row results. The goal is to prove that machine-to-machine integrations behave correctly before a real external system depends on them.
How is API testing different from UI testing?
UI testing drives the Lightning experience a person uses and checks layouts, buttons and on-screen behavior. API testing sends requests directly to endpoints and asserts on raw payloads, status codes and persisted state. A UI suite can pass while an integration quietly loses or corrupts data, because the screen never shows the payloads, so the two disciplines are complementary rather than interchangeable.
Which Salesforce APIs should be tested?
The three most common are the REST API for record CRUD, queries and composite calls, the SOAP API for strongly-typed enterprise and partner integrations, and the Bulk API for high-volume load and extract jobs. Each has distinct failure modes, so each is worth testing where your integrations use it. Dedicated REST, SOAP and Bulk API testing pages cover the specifics of each.
Why do partial failures in bulk jobs matter so much?
A Bulk API job can report overall success while individual rows within its batches failed on a validation rule, trigger or type error. If nobody reads the per-row results, record counts diverge silently and the gap only surfaces later during reconciliation. SyntraFlow is designed to read batch results row by row so success and error counts are reconciled and partial failures are reported.
Can a call return success but still not persist the record?
Yes. A status code reflects how the API request was handled, but a downstream trigger, validation rule or Flow can still reject or alter the write. That is why a thorough API test issues a follow-up query to confirm the record actually landed with the expected values, rather than trusting the status code alone.
Does SyntraFlow replace tools like Postman or Apex tests?
No. SyntraFlow complements them. Manual tools and unit-level Apex tests remain valuable during development, while SyntraFlow is designed to run API checks as repeatable steps inside broader end-to-end scenarios and, uniquely, to confirm the same transaction on the connected system's side. Its API testing capabilities are available for demonstration and proof-of-concept validation against your org.
Prove your Salesforce APIs before production does
See how SyntraFlow is designed to exercise REST, SOAP and Bulk calls, assert field by field, and confirm the data landed on both sides.