A login-code spike can exceed a provider’s per-second cap even when your app is nowhere near its daily quota. The result is a delayed or rejected message, not necessarily a problem with your daily-volume estimate.

This guide explains transactional email rate limits and shows how to estimate peak capacity, handle throttling, and understand the limits that apply when sending through Volanea.

Transactional email rate limits are provider-enforced ceilings on recipient volume or sending throughput. A quota caps recipients over a time window; a rate cap limits recipients per second. The applicable values depend on provider, account status, region, and configuration. Check both, then compare them with peak demand and your queue’s ability to pace work.

What are transactional email rate limits?

A sending quota limits the total number of recipients a provider will accept during a time window. A sending rate limits how quickly it will accept them. These are separate constraints: an app can remain below its daily quota and still exceed its per-second rate during a burst of login codes or receipts.

Amazon Simple Email Service (Amazon SES) makes the distinction explicit. In the SES sandbox, an account can send 200 emails per 24-hour period and 1 email per second. SES counts recipients, not messages: an email sent to 10 recipients uses 10 units of quota. Its 24-hour quota is rolling, so SES considers the previous 24 hours when a send is attempted. (SES quotas; how SES calculates sending quotas)

Those sandbox figures are not universal limits. SES quotas are separate for each AWS Region, and sandbox status and limits can differ by Region. For accounts outside the sandbox, AWS says the 24-hour quota and sending rate vary based on the specific use case. Other providers set their own rules, which can also vary by account status, region, and configuration. (SES Regions)

For an application, this means checking only a daily total is not enough. Estimate the peak rate for each important flow, such as authentication or payment notifications, as well as the total recipients sent during the provider’s quota window. Treat them as separate capacity checks.

Rate limit, message size, and bulk threshold: what’s the difference?

A rate limit measures sending volume over time. A message-size limit measures the encoded size of an individual email, including attachments. A bulk threshold would be a recipient count at which a send is classified as bulk, but there is no universal technical threshold that applies across providers and jurisdictions.

ConstraintWhat it measuresAmazon SES example
Sending quotaRecipients sent during a time windowSandbox: 200 emails per 24-hour period; counts recipients, not messages.
Sending rateRecipient throughputSandbox: 1 email per second.
Message sizeEncoded size of one message, including attachmentsSES v1 API: 10 MB per message after base64 encoding. SES v2 API and SMTP: 40 MB per message after base64 encoding.
Bulk classificationPurpose or scale of a sendThere is no universal recipient-count threshold that turns an email into bulk mail.

The SES message-size limits are not interchangeable with its sending quota. Messages larger than 10 MB are also subject to bandwidth throttling, a separate constraint from the number of recipients accepted per second. (SES service quotas)

Classification depends on why a person receives a message, not only how many people receive it. A password reset triggered by an individual action is usually transactional; a newsletter or product announcement sent to a list is marketing or bulk email. GDPR, CCPA, CAN-SPAM, and CASL obligations depend on the message and applicable jurisdiction; a low recipient count does not by itself make a promotional email transactional or exempt it from marketing rules.

How to find your provider’s sending limits

Check the limits for the account and region your application actually uses. A general product page may not show account-specific quotas, so use provider documentation to understand the terms and the account dashboard to find the values that apply to your sending setup.

Use this checklist:

  1. Find the quota window. Identify whether the ceiling is daily, rolling, monthly, or based on another period. Confirm whether the provider counts messages or recipients.
  2. Find the throughput cap. Record the rate and its units, and determine whether it applies per account, region, or sending identity.
  3. Look up throttling behavior. Find the response returned when a quota or rate is exceeded, and whether the provider retries, rejects, or drops the send.
  4. Check account status and scope. Confirm the active region, sandbox or production status, and any account-specific restrictions.
  5. Keep the values together. Record the account, region, quota window, rate, and date you reviewed the dashboard so your capacity estimate refers to the right limits.

SES is a useful example of why scope matters. Its sending quotas are regional, and production quota and rate vary by account and use case rather than following one fixed production value. AWS documents that an SES account must leave the sandbox to send to unverified recipients; moving out of the sandbox involves requesting production access and can include a request for higher sending quotas. (SES quota documentation; SES production access and quota management)

The official AWS quota pages do not state a publication or update date for these figures. For an SES account, use its current regional dashboard values rather than treating an example or a value from another account as your own limit.

What SES limits mean when you send through Volanea

Volanea is built on Amazon SES, and every email sent through Volanea goes out over SES. That does not mean a Volanea customer has to request SES production access or manage the underlying SES account’s sandbox quota. Volanea’s sending account does not require an SES production-access request, but you must verify your sending domain before sending to real recipients. (Volanea)

For Volanea users, plan around the Volanea plan’s monthly email volume and your application’s peak traffic rather than applying the SES sandbox figures to your own account. Volanea plans differ only by monthly volume and domain count; every feature is available on every plan. If the monthly limit is reached, sending pauses. There is no overage billing.

That pause is different from a per-second provider throttle. A monthly limit concerns how much you can send over the plan period; throughput concerns how quickly sends can be accepted during a burst. The product facts here do not specify a numerical per-second Volanea throughput allowance, so do not infer one from the SES sandbox example or assume that the monthly allotment guarantees a particular burst rate. Estimate your peaks and contact the provider when you need to confirm a specific throughput requirement.

Volanea provides both an SMTP relay and a REST API with a documented POST /v1/send endpoint. The sending pipeline includes suppression checks, contact upserts, template rendering, tracking, and delivery. These product capabilities do not remove the need to handle queueing and peak demand in your application. (Volanea API and SMTP details)

How to estimate the sending capacity your app needs

Estimate capacity from peak demand, not just daily volume. Identify the highest sustained send rate you expect and the largest burst your system must clear within a chosen time window. For each flow, estimate recipients triggered during that window, then divide by the time available to send them.

For example, suppose a login surge creates 240 login-code emails in 60 seconds. That requires an average of 240 ÷ 60, or 4 sends per second while the burst lasts. If receipts add another 120 emails during the same minute, combined demand becomes (240 + 120) ÷ 60, or 6 per second.

A daily average can hide that peak. A total of 8,640 emails spread evenly across a day averages 0.1 per second, but that average says little about a one-minute surge. For a quota based on recipients, also count all recipient addresses, including any To, CC, or BCC recipients—not just the number of API calls or message objects.

Now add a campaign spike. If a campaign is scheduled to send 1,800 emails over five minutes, its planned rate is 1,800 ÷ 300, or 6 per second. If it overlaps with the login-and-receipt burst, the combined estimate is 12 per second. That is a planning estimate, not a safe target at the exact provider ceiling: leave room for uneven arrivals, retries, and other application events.

Before launch, write down:

  • Expected peak sends per second for each flow.
  • The largest burst and the period in which it should clear.
  • The number of recipients, not only messages, for quota calculations.
  • Which messages are time-sensitive and should take priority if capacity is constrained.
  • Campaign timing that could overlap with authentication, receipts, or other operational mail.

This makes the trade-off visible. If a campaign can be paced over an hour but login codes must be sent promptly, reserve capacity for the codes and schedule the campaign around them. For Volanea, compare the total monthly volume you expect with the plan’s monthly limit as well as planning for peak throughput; those are different questions.

What happens when you hit a rate limit?

The result depends on the provider and the response handling in your application. Amazon SES drops a message that would exceed the daily sending quota or maximum sending rate; it does not attempt redelivery. For API or AWS SDK sends, SES returns a ThrottlingException with either Daily message quota exceeded or Maximum sending rate exceeded. AWS advises waiting an interval of up to 10 minutes before retrying. (SES throttling errors)

Do not make a user-facing request depend on an immediate provider send succeeding. Put outbound messages in a durable queue, record the provider’s response, and let a worker pace delivery against the applicable limit. A queue lets your application retain a receipt or login-code task after a temporary rejection rather than silently losing it when the request handler ends.

Retry only errors that may be transient, and use a controlled schedule instead of an immediate loop. Repeatedly resubmitting a throttled message can create more rejected requests while competing with newer, time-sensitive messages. Keep retry state and attempt counts, make retries idempotent to avoid duplicate receipts or codes, and route messages that exhaust the retry policy for investigation.

For SES SMTP sends, the client might display 454 Throttling failure: Maximum sending rate exceeded or 454 Throttling failure: Daily message quota exceeded. SMTP clients handle those errors differently, so confirm whether yours exposes a temporary failure to your queue or treats it as final. Do not interpret a throttle response as confirmation that the recipient received the email.

How to scale transactional email without hurting delivery

Scaling means managing both the amount of mail and its effect on delivery. Increase production volume gradually and monitor delivery outcomes as volume changes. A growing sending rate is not a reason to ignore bounces or complaints; those results can indicate that the recipients or message source need attention before more traffic is added.

AWS says it may automatically increase SES sending quotas for out-of-sandbox accounts sending high-quality production email. Its listed conditions include sending wanted, actual production content; regularly approaching but not exceeding the daily maximum; and keeping bounce and complaint rates low. These are AWS’s criteria for its own quota increases, not universal bounce-rate benchmarks. (AWS quota increase guidance)

Monitor bounces and delivery outcomes by message type where reporting allows. An overall account metric can obscure a problem in one campaign segment or in a particular source of signups. If bounce rates rise or delivery outcomes worsen after a volume increase, pause further growth and investigate recipient quality and the affected source before sending more.

Campaigns and application-triggered messages can have different timing requirements. A campaign burst can compete with login codes or receipts for shared sending capacity, so use separate queues or pacing policies where possible. Volanea combines transactional sending, campaigns, and automations on one API, which makes it practical to plan these message types together; it does not make their traffic patterns identical. Volanea also has a reputation guard that slows sending when a list may harm the sender domain. Account for that possible slowdown when planning a campaign rather than expecting all queued sends to proceed at the same pace.

Volanea sends from shared Amazon SES infrastructure and does not offer dedicated IPs. That makes list quality and paced sending relevant to the shared sending environment; it also means a dedicated-IP strategy is not an option within Volanea. For more sending and delivery guidance, see our email deliverability and sending guides.

SMTP or API: does the sending method change the limit?

No. A provider can enforce quota and throughput limits on sends submitted through SMTP or an email API. The protocol changes how your application connects and receives errors; it does not automatically provide more sending capacity.

Amazon SES applies sending quota and rate controls to email sent through either path. An API or AWS SDK send that exceeds a limit returns a ThrottlingException; an SMTP client might display a 454 Throttling failure response. The practical distinction is how your application surfaces and handles the error. Connect either path to a queue so a temporary rejection does not silently discard a time-sensitive message. (SES throttling guidance)

An API may fit an application that handles structured responses directly. SMTP may fit an existing mail library or framework, but check how that client reports temporary failures and whether it retries automatically. In either case, avoid overlapping library retries with your own worker retries unless you can account for both; otherwise, a single message can generate more attempts than your queue policy suggests.

Message size is a separate concern. SES supports up to 40 MB per message after base64 encoding through SMTP and the SES v2 API, compared with 10 MB through the SES v1 API. Those message-size limits do not bypass sending quotas. (SES service quotas)

When to request a limit increase or change providers

First decide whether the constraint is temporary or recurring. A campaign that can be paced without missing its delivery window may need scheduling, not a higher quota. Recurring peaks that delay login codes or receipts beyond your product’s latency target are stronger evidence that you need more capacity or a different sending setup.

For SES, request a sending-rate or quota increase when measured production demand regularly approaches the current account limit and your bounce and complaint rates are low. AWS routes increase requests through the Service Quotas console for the relevant Region and states an SLA of up to 24 hours for sending quota or rate increase requests. Production limits vary by account and use case, so build a request around your observed peak and expected growth rather than assuming a standard SES production quota. (AWS quota increase process)

Consider another provider when you cannot obtain the capacity required for your response-time targets, when the cost at your expected volume no longer fits, or when you cannot get enough visibility into limits and delivery outcomes. Compare stated capacity, escalation process, delivery reporting, migration effort, and cost at the same volume. A higher ceiling alone will not fix poor recipient quality or a queue that retries too aggressively.

For a concrete Volanea plan reference, Scale costs $149/month for 200,000 emails. That figure gives a stated price and volume for comparison; it is not an apples-to-apples cost comparison with running SES directly, which depends on the services and engineering work included in your own setup. Volanea’s plans differ by monthly volume and domain count, and features are available on every plan. (Volanea)

FAQ

Is there a fixed bulk email threshold?

No single recipient count defines bulk email across providers and laws. A one-recipient password reset is usually transactional because it responds to an individual action; a product announcement sent to a list is promotional even if the list is small. Apply GDPR, CCPA, CAN-SPAM, or CASL requirements according to the message and relevant jurisdiction, not an assumed universal cutoff.

What happens when I reach Volanea’s monthly sending limit?

Sending pauses when the plan’s monthly email limit is reached; Volanea does not charge overages. Every feature is available on every plan, and plans differ by monthly volume and domain count. If a workload is approaching its limit, measure expected transactional and campaign activity and choose a plan whose volume fits that forecast.

When should I switch providers instead of requesting more capacity?

Consider switching when your provider cannot support recurring peak demand within your latency target, or when its limits, escalation process, reporting, or cost do not fit your operation. Compare providers at the same projected volume and include migration work for sending domains, templates, and suppression data. A new provider will not correct a high bounce rate caused by poor recipient quality.

Start free with 1,000 emails a month