Message Throughput —TPS, transactions per second— defines how many messages can be processed per second within a delivery workflow. TPS should be understood as a dynamic capacity, ultimately determined by provider limitations and the specific channel being used.

TPS Management in the Platform

Messangi controls outbound traffic using the Throttle setting, available when sending an SMS or WhatsApp campaign from the Summary section.

This option defines the maximum number of messages per second that the platform will attempt to send through the selected delivery channel (SMS or WhatsApp).

Throttle acts as an internal ceiling, not as a guarantee of final throughput.

Actual delivery capacity may be lower due to:

Limitations imposed by providers and operators.
Channel-level or network conditions outside the platform’s control.

The platform ensures it will not exceed the configured limit, but it cannot increase TPS if the final provider applies stricter restrictions.

API-Level (Rate Limiter)

In addition to the visible Throttle configuration in the platform, Messangi applies an API-level rate limit to control the volume of requests entering the system. Both mechanisms follow the same principle: a maximum number of allowed operations per second.

The difference lies solely in where each limit is applied:

Throttle (UI):
Used when sending campaigns through the platform. It controls how many messages per second the system will attempt to process toward SMS or WhatsApp channels.
Rate limit (API):
Applied when customers integrate directly with Messangi’s APIs. It limits how many requests per second the system can accept from an external integration.

In other words:

Throttle and API rate limiting operate under the same logic.
Throttle governs message throughput when sending from the platform UI, while the API rate limit governs how many incoming requests are accepted when sending via API.

This API limit is internal and cannot be configured through the request body. Its purpose is to maintain platform stability and prevent overload, especially under high-volume traffic.

TPS in SMS

Internal Control

For SMS, Throttle regulates the rate at which the platform will attempt to send messages for delivery. This value defines an internal limit, but does not represent the actual throughput permitted by mobile operators.

Even when selecting full capacity, the final TPS depends on providers and the routing assigned to each country.

Provider Restrictions

Once messages are sent, the effective TPS may vary due to:

Throughput limits are defined by providers per country or route.
The maximum message-per-second rate allowed by each mobile operator.

Real example: Some operators in Colombia enforce a 20 TPS limit. Even if the platform is configured to send at a higher rate, the operator will only accept up to 20 TPS.

TPS in WhatsApp

Internal Control

In WhatsApp, Throttle also regulates the sending rate to Meta. This value represents an internal limit, but it does not determine the final throughput observed by the customer.

Capacity Defined by Meta

The effective TPS is determined by Meta:

80 messages per second (mps) per phone number
Possible automatic increase up to 1,000 mps (this is an estimate and may vary)

Meta applies this limit to all events associated with the number, including:

Outbound messages
Inbound messages
Delivery receipts
Read receipts
Other WhatsApp-related events

This means that the 80 mps limit does not apply exclusively to outbound messages, but to the complete activity of the phone number.

Understanding TPS as a Relative Capacity

The final TPS depends on the most restrictive component in the entire delivery chain. Even if a customer selects full capacity or configures a high Throttle value, actual throughput may be affected by:

Network traffic conditions
Channel-specific rules (SMS or WhatsApp)
Regional throughput restrictions

For these reasons, TPS should always be interpreted as a relative and variable capacity, dependent on the specific conditions of each channel and provider.

For optimal and stable performance, we recommend operating at up to 50 TPS, especially during continuous or high-volume messaging workflows. This helps ensure consistent delivery even when providers or networks apply temporary throughput variations.