Visma and Gusto integration

What moves between them

The main flow runs from Gusto into Visma. When an employee is created or updated in Gusto, ml-connector reads the employee event (name, job title, location, department) and writes or updates the matching Employee record in Visma. The sync is event-driven via Gusto webhooks, so changes reach Visma within seconds of taking effect in Gusto. Gusto has no native GL accounts, invoices, or accounting dimensions, so the flow is unidirectional; ml-connector does not write accounting data back to Gusto.

How ml-connector handles it

ml-connector registers a Gusto webhook endpoint to receive employee lifecycle events (created, updated, terminated, onboarded). When an event arrives, it verifies the HMAC-SHA256 signature using the Gusto subscription verification token, then looks up the employee details via the Gusto API using the company's OAuth2 token (which ml-connector refreshes automatically when the 2-hour access token expires). Each employee record includes name, job title, location, and department assignments. ml-connector maps the Gusto location or department to the corresponding Visma Dimension or employee classification, then creates or updates the Employee record in Visma using OAuth2 client credentials and the ipp-company-id header. Gusto enforces 200 requests per minute per OAuth grant on a rolling window, so ml-connector queues and batches calls to stay within this limit. If the Gusto API returns a 409 Conflict, it re-fetches the current employee version and retries with the updated version field, respecting Gusto's object-versioning idempotency model. Webhook delivery retries continue for up to 3 days if the ml-connector endpoint is unavailable, and every transaction is logged for full audit and replay on failure.

A real-world example

A mid-market consulting firm uses Visma.net ERP for project accounting and timekeeping across five Scandinavian offices, and Gusto for payroll and benefits administration for all 150 employees globally. Before the integration, the accounting team received a monthly employee roster export from Gusto and manually updated employee names, titles, and department assignments in Visma to ensure timecards posted to the correct cost center. When employees changed roles or transferred between offices, the HR team and finance team had to coordinate the update to avoid mismatched GL postings. With Gusto and Visma connected, each employee change flows into Visma automatically, departments stay synchronized across both systems, and finance can run accurate project cost reports without chasing HR for manual reconciliation.

What you can do

• •Sync employee names, job titles, and department assignments from Gusto to Visma whenever an employee is created or updated.
• •Receive Gusto webhook events for employee lifecycle changes (onboarding, role updates, terminations) and apply them to Visma in near real-time.
• •Verify Gusto webhook signatures using HMAC-SHA256 and the subscription verification token for security.
• •Handle Gusto's object-versioning idempotency model and rate limits (200 requests per minute per OAuth grant) transparently.
• •Authenticate both systems using OAuth2 and manage token refresh automatically.

Questions

Does the integration sync payroll data or GL accounts from Gusto to Visma?

No. Gusto has no native GL accounts, invoices, or accounting dimensions. ml-connector syncs employee master data (names, titles, departments, locations) from Gusto into Visma so the two systems stay aligned on headcount and organizational structure. Payroll GL posting would require custom mapping on top of the employee sync; that work would be scoped separately.

How does ml-connector handle Gusto's rate limit of 200 requests per minute?

ml-connector batches and queues employee update calls to stay within the rolling 200-request-per-minute limit per OAuth grant. If a call returns a 429 rate-limit error, ml-connector backs off and retries automatically. For multi-company setups, each company gets its own OAuth token and rate-limit bucket, so a high-volume sync in one company does not throttle another.

What happens if a Gusto webhook fails or ml-connector is temporarily unavailable?

Gusto retries the webhook delivery up to 16 times over 3 days if the endpoint returns non-2xx or does not respond within 10 seconds. ml-connector queues the update for processing once it recovers, and the full transaction log lets you replay any missed updates manually if needed.