At 100,000 emails a month, Mailgun Scale is $90/month after its first free month, while Postmark Basic calculates to $177/month at the same volume. At 20,000 emails a month, Mailgun Basic and Postmark Basic both calculate to $33/month. If you plan to migrate from Mailgun to Postmark, those figures make plan selection and workload fit more useful than comparing entry prices alone.

To move safely, inventory each sending workload, compare costs using your expected monthly volume, and configure Postmark’s DNS records before changing application credentials. Then map each Mailgun send path to Postmark’s REST API or SMTP, rebuild the pieces that are not portable, and test message rendering, delivery events, and recipient rules before routing production traffic. Keep the old configuration available during the cutover.

What migrating from Mailgun to Postmark costs

The cost difference depends on volume and plan, not just the starting monthly fee. Mailgun Basic starts at $15/month for 10,000 emails, with additional messages from $1.80 per 1,000. Postmark Basic is also $15/month for 10,000 emails, with extra emails at $1.80 per 1,000. Both pricing pages list those rates: Mailgun and Postmark.

At 20,000 emails/month, the listed Basic rates produce the same cost: Mailgun Basic vs Postmark Basic at 20,000 emails/month: 15 + (20,000 − 10,000) ÷ 1,000 × 1.80 = $33/month. A move at this volume is a workflow decision rather than a Basic-tier saving.

At higher volumes, compare the actual tiers. Mailgun Scale is free for one month, then $90/month for 100,000 emails and includes 5,000 email validations. Postmark Basic at 100,000 emails/month: 15.00 + (100,000 − 10,000) ÷ 1,000 × 1.80 = $177/month. That is a substantial difference between Mailgun Scale and Postmark Basic at this volume, so compare the capabilities you need as well as the totals.

Postmark Pro starts at $16.50/month for 10,000 emails, with extra emails at $1.30 per 1,000. Platform starts at $18/month for 10,000, with extra emails at $1.20 per 1,000. Both include unlimited emails per day; Pro and Platform also list inbound email processing and customizable data retention up to 365 days, with 45 days as the default. Postmark’s paid plan charges for extra emails at the end of the billing cycle and includes them with the next month’s plan cost.

Add-ons can change the comparison. Postmark DMARC monitoring starts at $14/month per domain. Its dedicated IP add-on starts at $50/month per IP, for customers sending at least 300,000 emails per month on Pro or higher. Mailgun lists additional dedicated IPs at $59/IP/month. Mailgun validation is also separately priced: Basic and Foundation validations start at $1.20 per 100, while Scale includes the first 5,000 and then starts at $0.80 per 100. Include these only if they match services you actually use.

Inventory workloads before changing providers

We start migrations with a workload inventory, not a credential edit. A SaaS product may use Mailgun for account verification, password resets, receipts, product updates, and inbound support messages. Those flows can live in different services and deployment environments, so a single successful test send does not demonstrate that the full application has moved.

Record each sending domain, application, environment, message category, and send mechanism. Then note the templates, recipient rules, event consumers, and operational owner associated with each one. Mailgun supports both an HTTP API and an SMTP endpoint, so include both paths in the inventory if different parts of your system use them.

A concise migration register might look like this:

WorkloadCurrent pathDestination pathTest to pass
Password resetMailgun API from account servicePostmark REST APILink, sender, and event handling work
Invoice receiptMailgun SMTP from billing workerPostmark SMTP or REST APICorrect recipient and attachment behavior
Product updateMailgun campaign or application senderChosen Postmark sending pathRecipient rules and message rendering match
Support replyMailgun inbound routeRecreated inbound processing flowParser and downstream ticket action work

The table is a planning aid, not an assertion that every Mailgun feature has a direct equivalent. Mailgun documents APIs for managing domains, keys, and templates, but those capabilities do not provide a complete transfer package for another provider. We treat templates, suppression decisions, webhook behavior, and historical records as migration work to inventory and recreate deliberately.

The Email Provider Migration Checklist for Mailboxes and App Email is useful when a project includes both application email and hosted mailboxes. They are separate changes: moving application-generated messages does not migrate employee mailboxes.

Prepare DNS for Postmark

Add the sending domain in Postmark and use the hostname and value displayed in its DNS settings. Postmark’s domain verification uses a DKIM TXT record and a custom Return-Path CNAME. Its default Return-Path hostname is pm_bounces, and the value is pm.mtasv.net; use the values assigned in your account when configuring a real domain.

DNS purposePostmark configurationMigration action
DKIM signingTXT record with hostname and value from domain settingsAdd the record, then use Postmark’s Verify action
Return-PathCNAME; default hostname pm_bounces, value pm.mtasv.netAdd the assigned CNAME and confirm its status
DMARCPolicy remains configured at your domainReview alignment and reports after the new sender is active

Postmark says DKIM verification normally appears within 48 hours, and its Verify button can start a manual check. After verification, Postmark signs messages sent through that domain; any address on the verified domain can be used as a sender without adding it individually as a Sender Signature. Postmark uses 1024-bit DKIM keys, a detail to include in your security review.

Postmark says its custom Return-Path helps SPF alignment and that a separate custom SPF record is not required for that setup. Its documentation says SPF-based DMARC alignment requires relaxed alignment (aspf=r); the custom Return-Path cannot meet strict SPF alignment (aspf=s). Review the current DMARC policy and reporting with the domain owner before making changes.

Mailgun provides DNS-record access in its Control Panel under Sending > Domains > select domain > Domain Settings > DNS Records. Its domain verification checks A, CNAME, SPF, DKIM, and MX records. Use that view to understand the existing setup, but add Postmark’s required records separately and preserve DNS entries that other services still need.

Map API and SMTP configuration

Postmark supports sending through its REST API and/or SMTP, as does Mailgun. The interfaces are not interchangeable just because both providers offer those protocols. We recommend isolating provider-specific details behind one application boundary so teams can change credentials or request construction without editing every message-producing service.

A deployment configuration can make that boundary explicit. The names below are illustrative application variables, not provider-defined environment variable names:

EMAIL_PROVIDER=postmark
EMAIL_API_KEY=<secret stored by deployment platform>
EMAIL_FROM=receipts@example.com

In code, route application calls through a function such as sendTransactionalEmail(message). Keep provider authentication, endpoint selection, request mapping, and response handling inside that adapter. This gives you one place to compare the fields your Mailgun implementation currently sends with the fields required by the Postmark API, while the rest of the application continues to express business intent: recipient, subject, content, and message category.

Build a mapping table from real requests in your codebase. For each Mailgun request, record the sender, recipients, subject, text and HTML content, attachments, metadata, and error behavior. Then implement and test the equivalent Postmark request rather than assuming that similarly named fields have identical semantics. Avoid copying production credentials into test logs or source control.

For SMTP, change the endpoint and credentials in a controlled deployment, then test the mail library’s authentication and error handling. For REST, update the provider-specific client and secret. In either case, search environment files, serverless settings, background workers, scheduled jobs, and CI deployment configuration; stale configuration in one worker can keep sending through Mailgun after the main application has moved.

If your sender runs on AWS Lambda, inspect each function’s environment and deployment secrets as well as shared code. Our AWS Lambda email guide covers the deployment context for API-based sending, which is useful when a migration touches multiple functions or stages.

Rebuild templates and recipient rules

Treat template content as an application asset, not a provider setting that will move automatically. Mailgun’s API supports listing templates and retrieving their content and versions, and each template can store up to 40 versions. Retrieve the versions your application uses, identify their variables and conditional behavior, and recreate the required messages in the destination system.

Compare rendered output with representative data, including missing optional values, long names, non-ASCII characters, links, and empty fields. A visually correct preview with one sample payload is not enough if production requests use different variable names or omit fields. Keep a small fixture set for each important message type so that future template edits can be checked consistently.

Make the same deliberate pass over suppression and unsubscribe behavior. Identify why an address is excluded from sending, which system currently owns that decision, and how new bounce or complaint events update it. Preserve the business rule, not just a file of addresses: an opt-out, a hard bounce, and a temporary delivery problem should not automatically be treated as the same state.

For inbound processing, check whether the application uses Mailgun routes or inbound messages. Postmark’s Pro and Platform plans list inbound email processing; Basic does not list it. If inbound mail is part of the workflow, include the required plan and receiving path in the design, then test parsing, attachment handling, and the application action triggered by a received message.

Rebuild event handling and observability

Both providers list webhooks, but your application should not depend on provider-specific event payloads being identical. Inventory the Mailgun events that trigger application actions, then map each action to the corresponding Postmark event behavior using the actual integration and documentation. Update parsing, validation, database writes, and retry handling as one change.

For example, if a bounce event changes a recipient’s eligibility, make that update idempotent. A repeated event should not trigger duplicate customer notifications or reverse a later user action. Test delivery, bounce, complaint, and unsubscribe-related paths where those events are part of your application, and capture the provider message identifier alongside your own message or job identifier.

Keep enough structured logging to trace a send from application request to provider response and event callback. Useful fields include message category, environment, provider, internal job ID, and provider message ID. Avoid logging full message bodies or unnecessary personal data simply to make a migration easier to debug.

Postmark Basic lists 45-day data retention. Pro and Platform list customizable retention up to 365 days, with 45 days as the default. Decide whether that history supports your support and debugging needs, and retain application-side records separately if your process requires them for longer.

Run a staged cutover with a rollback path

We use a controlled sequence so that configuration, application behavior, and event handling are tested independently before a broad switch. Postmark’s migration guide describes sending and domain setup, but it does not prescribe a universal traffic cutover schedule; choose steps that match your release process and the consequences of a failed message.

  1. Configure the Postmark domain and verify DKIM and Return-Path records.
  2. Deploy the API or SMTP adapter to a non-production environment and send representative messages.
  3. Test templates, recipient rules, event callbacks, and inbound processing if the application uses it.
  4. Move one low-risk production workload, then inspect application logs and provider activity.
  5. Move the remaining message categories in planned releases, with an owner watching each change.
  6. Keep the Mailgun path available during the rollback window, but prevent both providers from sending the same logical message.

A worked example: a SaaS team sends receipts from a billing worker, password resets from its account service, and a monthly product update from a campaign process. The team can migrate and validate each path separately, rather than changing shared credentials and discovering later that a low-frequency worker still uses the old provider. They should also test the event handlers that update recipient state before relying on the new path for all three workloads.

Do not delete old credentials or DNS records immediately after the first successful Postmark send. First check production services, low-frequency jobs, staging environments, and webhook consumers. Remove old configuration only after the team has confirmed that no active workload depends on it.

When Volanea may fit the same evaluation

A Mailgun-to-Postmark move is not the only point at which to review email infrastructure. If your team wants transactional email, campaigns, and automated workflows through one API, we built Volanea on Amazon SES for that combined use case. We store contacts for $0 per contact, and every feature ships on every plan; plans differ by monthly volume and domain count.

Our free plan includes 1,000 emails per month. When a project reaches its monthly limit, sending pauses rather than generating overage billing. Volanea Scale is $149/month for 200,000 emails. We send from shared Amazon SES infrastructure and do not offer dedicated IPs, so teams that require a dedicated IP should weigh that constraint before considering us.

That makes the comparison about operating model as well as provider price: whether one API for transactional sends, campaigns, and workflows fits your architecture, and whether the monthly volume and domain allowance fit your needs. For teams whose current Mailgun and Postmark costs are driven by different feature requirements, compare the actual workload rather than treating a per-message rate as the whole decision.

FAQ

Is Mailgun or Postmark cheaper at 100,000 emails a month?

Mailgun Scale is listed at $90/month after its first free month for 100,000 emails, while the calculated Postmark Basic cost at that volume is $177/month. Postmark also has Pro and Platform tiers with different starting prices and lower additional-message rates than Basic, so compare the plan features and applicable overage rate you would use.

Can I use Postmark’s free plan for production email?

Postmark’s free plan is $0/month for 100 emails per month, has no overages, and does not expire. It can support a small test or a very low-volume application, but the allowance is much lower than the 10,000-email starting quota on its paid plans. Mailgun Free instead lists 100 emails per day, so compare the time basis as well as the monthly price.

Does Postmark include email validation?

The Postmark plan features listed on its pricing page do not include an email validation service. Mailgun lists validation as a separately priced capability on Basic and Foundation, while Scale includes the first 5,000 validations. If validation is part of your current flow, account for that dependency and its cost before retiring the Mailgun implementation.

Should I switch from Mailgun if my monthly volume is below 10,000?

Compare the free allowances and the work your application needs before switching. Mailgun Free includes 100 emails per day, while Postmark’s free plan includes 100 emails per month with no overages. A low-volume project may still move for workflow reasons, but first test the complete message and event path and confirm that the destination plan supports the features you use.