# Service level agreements

Google enforces several Service Level Agreements (SLAs) to ensure a reliable and
responsive user experience. Examples of common Google Standard Payments SLAs and
how they are measured are provided here.

> [!WARNING]
> **Warning:** The terms on this page are not legally-binding. You must refer to the specific terms in your Google Standard Payments contract or Payment Processing Services Agreement for the complete set of SLAs your integration is expected to adhere to.

### Example SLAs

#### System availability

##### Example terms

The Payment Integrator shall make their Systems available to
Google for 99.9% of requests over any 24 hour period (per API method).

The System shall be considered unavailable for a given request if System fails
to respond to a valid API request with a valid API response within 7 seconds as
measured by Google after a well-formed transactional packet has been sent to
the Payment Integrator.

The measurement period will be defined as one hour if more than 1000 API request
are sent to a given API method over the course of an hour. If fewer than 1000
API requests are sent to a given API method per hour, the measurement period
will be defined as the 24 hour period from midnight to midnight (UTC). Any calls
sent during a scheduled maintenance period will be ignored when calculating the
System availability SLA.

##### Example evaluation

System availability is then calculated as follows (for each API method):
$$ Availability = \\frac{CountOfSuccessfulCalls}{CountOfTotalCalls} $$

#### Latency

##### Example terms

The System responds to 95% of API requests within 5 seconds, and 90% of API
requests within 3 seconds over any one hour period (per API method).

The measurement period will be defined as one hour if more than 1000 API request
are sent to a given API method over the course of an hour. If fewer than 1000
API requests are sent to a given API method per hour, the measurement period
will be defined as the 24 hour period from midnight to midnight (UTC). Any calls
sent during a scheduled maintenance period will be ignored when calculating the
Latency SLA.

##### Example evaluation

Latency is measuring the time it takes for the Payment Integrator's System to
respond to a Google API request sent while the system is available. Therefore,
latency is calculated as follows:
$$ Latency = DatetimeResponseReceivedByGoogle - DatetimeRequestSentByGoogle $$