Errors & limits

Error responses

Errors use conventional HTTP status codes with a JSON body describing the problem. Alongside the human message, most carry a stable code — branch on code, not on the message text:

A 400 from a bad values map tells you exactly which key to fix — the message quotes the human label the template shows, and details gives the key your payload uses:

{
  "statusCode": 400,
  "error": "Bad Request",
  "message": "\"Client Name\" is required",
  "code": "invalid_values",
  "details": [{ "key": "client_name", "label": "Client Name", "reason": "is required" }]
}

Rate limit

Two limits apply, and you hit whichever is tighter:

• 120 requests per minute per IP address — shared by everything calling from that address, including other traffic from your network.
• 600 requests per minute per API key — your integration's own budget.

Both report through x-ratelimit-limit, x-ratelimit-remaining and x-ratelimit-reset (seconds until the window resets); a 429 also carries retry-after. If you're polling many signature requests, spread the polls out rather than bursting — or list them in one call with GET /esign/submissions.

Monthly signature-request allowance

Sending a signature request — from the app or through the API — counts against your workspace's monthly allowance, which resets at the start of each calendar month (UTC):

When the allowance is exhausted, POST /esign/submissions returns 403 with code: limit_reached (and the message limit_reached:signatureRequests), carrying the limit and how much was used. Listing templates, polling status, reading usage, and downloading signed documents are never metered, and a retried request with an already-used idempotencyKey is not charged again.

Rather than discovering the ceiling as a failed send, read GET /esign/usage before a batch.