Migrate from Mailgun to a sender-reputation-first European alternative
In December 2025, Mailgun doubled its Flex pay-as-you-go rate from $1 per 1,000 emails to $2 per 1,000 — overnight, the cheap-when-you-start path that had defined Mailgun's developer appeal since 2010 became roughly twice as expensive. Combined with the Sinch acquisition (closed 2021) that broader strategic shifts in pricing and product direction, plus the long-standing EU-residency trap where region must be set at account creation and migrating between regions causes service disruption, many teams that landed on Mailgun for the API quality and SMTP relay strength are now looking at alternatives. For teams sending 50K-2M emails/month with EU customers or EU operations who want predictable EUR pricing, dedicated IPs from the entry plan, and EU jurisdictional routing without the account-creation-time region lock-in, here's the honest migration path to Authorize Hosting.
Three structural shifts at Mailgun post-Sinch acquisition, and what they mean for teams evaluating alternatives
Sinch acquired Mailgun in 2021. For the first few years after the acquisition, Mailgun continued to operate as the developer-favorite transactional email API it had been since 2010 — the same RESTful API, the same SMTP relay, the same focus on inbound routing as the differentiating feature against SendGrid. The platform powers transactional email for GitHub, Slack, Lyft, Reddit, and a long tail of mid-market SaaS. Pricing was unusually stable; SaaSPricePulse tracked 139 pricing snapshots since 2018 without detecting a single price increase through 2024.
That stability ended in December 2025. Mailgun doubled its Flex pay-as-you-go rate from $1 per 1,000 emails to $2 per 1,000 emails. For teams on Flex who landed there because the pay-as-you-go model fit their unpredictable volume, the bill roughly doubled overnight. The Flex plan, which used to be the cheap on-ramp for early-stage projects, is now impractical above ~17,000 emails/month — at that volume the Foundation fixed plan at $35/month is cheaper. Several Mailgun customers we've spoken with described the doubling as the moment that surfaced longer-running concerns about the post-acquisition direction: it wasn't just about $1 versus $2, it was about pricing volatility that they hadn't experienced before.
The second structural concern is EU data residency. Mailgun does offer EU-hosted infrastructure as a differentiating feature against SendGrid (which lacks it for transactional). The catch: EU residency has to be configured at account creation, and migrating an existing US-region account to EU causes service disruption. Teams that signed up casually for a US account in 2020-2022 and later realized they needed EU routing for GDPR or for customer trust reasons cannot simply flip a region setting; they have to provision a new account in the EU region, re-set up domains, re-warm IPs, and migrate traffic. The architectural decision made at signup propagates forward as a multi-week migration project.
The third shift is around dedicated IP economics and deliverability tool unbundling. Mailgun's dedicated IPs cost $59/month each on plans that support them, and the Optimize suite (which includes the deliverability tools many teams actually need — inbox placement testing, complaint feedback loops, suppression management upgrades) is sold separately starting at $49/month. The Foundation plan at $35/month doesn't include dedicated IPs at all; you need to be on Scale ($90/month minimum) to add them, then $59 each beyond the first. Subaccount management for multi-tenant SaaS also requires Scale minimum. For mid-market teams that need 3-5 dedicated IPs, deliverability tools, and subaccount support, the actual monthly cost lands at $250-400 before any sending volume considerations.
Teams moving off Mailgun in 2025-2026 typically cite some combination of: the Flex doubling that broke their cost predictability, the EU residency lock-in that limits architectural flexibility, the unbundled deliverability tooling that pushes the effective price up, and the strategic uncertainty about Mailgun's direction within the broader Sinch portfolio (which also includes Mailjet, Inteliquent, and several other communications properties). The Mailgun API itself remains technically excellent — the inbound routing is a genuine strength that most competitors don't match — and we'll be honest about what we don't replace as well as what we do replace better.
Mailgun versus Authorize Hosting on the dimensions that actually shape the bill and the deliverability
| Dimension | Mailgun (Sinch) | Authorize Hosting |
|---|---|---|
| Entry plan (paid) | Foundation $35/mo (50K emails) or Flex $2/1K (doubled Dec 2025) | SMTP Relay Starter €399/mo (Email API €469/mo) |
| Pay-as-you-go path | Flex doubled to $2/1K in December 2025; impractical above 17K/mo | No pay-as-you-go; tier-based monthly pricing |
| Dedicated IPs included | Scale tier only ($90/mo+); $59/mo per dedicated IP added | 5-30 dedicated IPs included by plan tier |
| Deliverability tools | Optimize suite separate at $49/mo+ on top of base plan | Bundled in plan; not a separate add-on |
| Subaccount management | Requires Scale plan minimum ($90/mo) | Subaccount support from entry plan; per-subaccount IP allocation available |
| EU data residency | Must be set at account creation; migration between regions causes service disruption | EU-only routing by default; no region flag, no US-region trap |
| Jurisdiction | US (Sinch parent in Sweden, but Mailgun infra+entity US-based); CLOUD Act exposure | EU (Sweden HQ, Stockholm + Frankfurt routing); not subject to CLOUD Act |
| Inbound routing | Strength of the platform; sophisticated rules engine | More limited; we pair with dedicated inbound services for complex workflows |
| Pricing stability | Stable 2018-2024, then Flex doubled Dec 2025 — recent volatility | Tier-based EUR pricing with no recent changes; multi-year contracts available |
| Account-level support | Email support on Foundation; phone on Scale; dedicated on Growth | Named operator from first plan; EU business hours phone + Slack channel |
| Shared IP placement | Independent tests show ~71% inbox placement on shared IPs | Dedicated IPs from plan entry; no shared-pool contamination risk |
Pricing data current as of Q1 2026 from public Mailgun pricing pages and independent reviews (Mailflow Authority, SaaSPricePulse, Sequenzy). The Flex rate increase is documented across multiple independent sources as of December 2025. We'd recommend re-verifying current pricing on Mailgun's official pricing page before a final decision.
How a typical Mailgun migration runs — week by week, with no deliverability drop
Mailgun migrations run on the same 30-day cadence we use for SendGrid migrations, with one specific Mailgun consideration: if you've been using the inbound routing feature, the inbound side needs its own migration sub-plan (typically running 1-2 weeks after the outbound migration completes). For pure transactional outbound, the standard playbook applies.
DNS preparation and dual-provider authentication setup
We provision your dedicated IPs and generate fresh DKIM keys under your sending domains. You add the new DKIM records alongside your existing Mailgun records — multiple DKIM selectors can coexist. SPF gets updated with our include: directive added to your existing record (Mailgun's mailgun.org include can stay during the migration period). DMARC stays at whatever policy you currently have. By end of week, every sending domain has both providers' authentication records published.
Gradual ramp on the new dedicated IPs while Mailgun stays primary
Controlled warmup on the new IPs: day 1 sends 50 messages, doubling every 2-3 days through a 14-day curve to reach the steady-state daily volume. The warmup traffic is your highest-engagement segment — typically transactional emails to active users (notifications, receipts, password resets) rather than digest or marketing content. This is critical for teams previously on Mailgun shared IPs: dedicated IPs need real engagement to establish their own reputation, and using your best-engagement segment during warmup is what makes the curve work.
10-30% of traffic shifted, deliverability compared in real time
We start routing a measurable percentage of your transactional traffic through us — typically 10-20% via a feature flag in your application code or an environment variable that selects the provider. The dashboards on both providers show comparable metrics; we line them up side-by-side. Mailgun's analytics granularity is genuinely good, so this is a clean comparison. By end of week 3, most teams reach 30-50% on our infrastructure.
100% traffic on Authorize Hosting, Mailgun as standby
By end of week 4, new IPs are at steady-state volume, deliverability matches or exceeds Mailgun baseline, and 100% of outbound traffic shifts to us. Mailgun stays as hot standby for 2-4 weeks — keep the contract active so the failover-back option exists. Most teams cancel at end of week 8 once they've seen a full billing cycle of steady-state performance. For Mailgun customers on Flex specifically, the cancellation produces immediate cost savings since there's no contract term to wait out.
For teams using Mailgun inbound: dedicated migration sub-plan
Mailgun's inbound routing — the sophisticated rules engine that parses incoming mail and routes to webhooks or storage — is a genuine strength of the platform. Our inbound is more limited. For teams that rely on Mailgun inbound for parsing replies to transactional mail, processing inbound webhooks, or routing user-replies into the application, the migration path is one of: (1) keep Mailgun for inbound only at a minimal plan ($35/month Foundation) while we handle outbound, (2) move inbound to a dedicated inbound service like CloudMailin or Postmark Inbound, or (3) build inbound on Postfix directly. We'll discuss the right answer for your specific inbound workflow during the migration scoping.
The Mailgun feature set, mapped to what we ship — and what we don't
- REST API + SMTP relayBoth endpoints; Mailgun-style API request structure with per-domain credentials; SMTP relay with STARTTLS, TLS 1.3, OAuth options
- Webhook eventsDelivered, opened, clicked, unsubscribed, complained, bounced, dropped, failed — same Mailgun event taxonomy for drop-in replacement of event handlers
- Multi-domain supportSending domains added through API or dashboard; per-domain DKIM keys; per-domain SMTP credentials matching Mailgun's per-domain model
- Email validationAddress validation with syntax, MX, role-based detection, disposable-domain check; included rather than per-address billed as separate add-on
- Suppression listsBounces, complaints, and unsubscribes maintained automatically; manual additions via API; per-domain or account-wide scoping
- Template engineVariables (
%recipient.name%-style) and conditional logic supported; Mailgun template syntax converts straightforwardly - Tagging and analyticsPer-message tags, custom variables, log retention 30+ days, analytics dashboards comparable to Mailgun's granularity
- Inbound routing rules engineMailgun's sophisticated routing (regex-matched, action-chained) is more powerful than ours. For complex inbound workflows we recommend keeping Mailgun inbound-only at Foundation tier, or pairing with CloudMailin / Postmark Inbound / Postfix-direct
- Mailgun Optimize toolsOptimize suite features (inbox placement testing, complaint feedback acceleration, suppression management upgrades) are bundled in our base plans rather than sold as separate add-on
- EU data residencyMailgun has US and EU regions selectable at account creation. We're EU-only by default — no region toggle, no migration between regions. Simpler for EU customers; if you specifically need US-based routing for latency reasons, we're not the fit
- Pricing modelTier-based monthly pricing in EUR; no pay-as-you-go option like Mailgun Flex. Predictable but doesn't support the very-low-volume use case Flex used to fit
- Account onboardingNamed operator engagement rather than self-service signup; first 30-60 minute conversation covers requirements, IP allocation, integration architecture
How Mailgun migrations typically size on the new infrastructure
SMTP Relay Starter · €399/mo
5 dedicated IPs (Foundation has 0), up to 200K messages/month, full authentication suite, deliverability tools bundled (no separate Optimize add-on). Sized for teams currently on Mailgun Foundation at $35/mo + Optimize $49/mo + occasional Flex overages.
See SMTP RelayEmail API Pro · €859/mo
20 dedicated IPs (Scale has 1, then $59/mo each additional), up to 2M messages/month, full Postmaster Tools integration, subaccount management included, dedicated deliverability engineer. Landing spot for teams on Mailgun Scale at $90/mo + 3-5 dedicated IPs at $59 each + Optimize $49 + subaccount tier = $400-500/mo.
See Email APIPowerMTA Enterprise · from €2,799/mo
PowerMTA-based architecture for teams above 2M messages/month, 30+ dedicated IPs, custom IP pool segmentation, named deliverability engineer, EU-only routing across Stockholm + Frankfurt. For teams currently on Mailgun Growth plan or negotiating Enterprise custom pricing.
Open the conversationWhat teams ask before kicking off the Mailgun migration
We rely on Mailgun inbound routing — what's the migration path?
+
Honest answer: Mailgun's inbound routing is a genuine strength of the platform and our inbound is more limited. If you depend heavily on the routing rules engine, three options:
(1) Keep Mailgun for inbound only. Foundation plan at $35/month gives you the inbound routing while we handle outbound. The split-provider pattern is clean architecturally — outbound goes through us with our dedicated IPs and EU routing, inbound stays on Mailgun where the routing rules already work. (2) Migrate to a dedicated inbound service. CloudMailin and Postmark Inbound both offer rule-based inbound parsing comparable to Mailgun's. The migration involves rewriting your rules in the new provider's syntax. (3) Build on Postfix directly. If your inbound workflow is well-defined and you have engineering resources, running Postfix with custom processors is more flexible than any hosted inbound. Higher operational burden, but no per-message inbound costs.
We'll work through the right answer for your specific inbound workflow during the migration scoping. For most teams, option 1 (keep Mailgun for inbound only) is the lowest-friction path.
We're on the Mailgun EU region — does that matter for migration?
+
It actually makes migration easier rather than harder. Teams that landed on Mailgun's EU region already made the architectural decision that EU routing matters — they're past the residency question and just need an EU-native provider that doesn't gatekeep that decision at account creation. The migration runs identically; the dedicated IPs we provision are EU-native from day one.
The bigger consideration is for teams on the Mailgun US region who realized later they needed EU. Moving from Mailgun US to Mailgun EU causes the service disruption we mentioned (re-provision account, re-add domains, re-warm IPs). The same effort moving from Mailgun US to us gets you to the EU-jurisdictional infrastructure without the within-vendor migration overhead.
Will we lose the sender reputation we've built on Mailgun?
+
Domain reputation transfers; IP reputation doesn't. Sender reputation at receiving mailbox providers (Gmail, Yahoo, Microsoft) is largely tied to the combination of sending IP and From: domain. Your domain reputation stays attached to the domain — when emails start flowing through our new IPs but the same From: domain, receivers see "this domain has historical engagement, but the IP is new." That's the warmup signal they expect, and the 14-day ramp on new dedicated IPs handles it.
What we don't transfer: any historical IP-warmup data Mailgun built on the shared pools they used for your account. Independent tests showed Mailgun shared IP inbox placement around 71%, which is decent but a ceiling rather than a foundation — moving to dedicated IPs is what unlocks the next placement tier.
How does this compare against migrating to Amazon SES?
+
Amazon SES is the lowest-cost option at high volume, especially after the December 2025 Flex doubling made Mailgun's previously-cheap entry point less compelling. SES costs $0.10 per 1,000 emails vs Mailgun's $2/1K (Flex) or $0.70-$1.80 per 1K depending on plan tier. For teams sending 1M+ emails/month with AWS engineering already in-house, SES is genuinely competitive and Tenants (launched August 2025) addresses some historical scaling concerns.
Where we differ from SES: managed deliverability practice versus metered API. SES expects you to own warming, monitoring, complaint-rate management, list hygiene — all the operational work that Mailgun's Optimize suite and our platform handle. For teams without dedicated deliverability staff, our pricing typically beats SES + the cost of the internal deliverability lead. For teams that have that capability in-house, SES is a real alternative and we'd be honest about that.
We're a multi-tenant SaaS on Mailgun subaccounts — can you handle that?
+
Yes, and the per-tenant isolation model on our infrastructure is more flexible than Mailgun's. The Mailgun subaccount feature requires the Scale plan minimum ($90/month) and provides logical separation but tenants share underlying IP pools by default. Our model lets you allocate dedicated IPs per tenant on the higher plan tiers, which means a tenant's complaint rate doesn't contaminate other tenants' deliverability.
The migration mapping: Mailgun subaccount → AH subaccount with optional dedicated IP allocation per high-volume tenant. For agencies, white-label dashboards are available; the per-tenant API tokens, suppression scoping, and webhook isolation work the same way Mailgun's do.
What about Mailgun analytics and historical data?
+
Export before cancellation. Mailgun retains log data for 30 days on Scale plan and longer on higher tiers. Before cancelling the Mailgun account, we recommend exporting your last 30-90 days of event data via the Mailgun API (the /events endpoint) or the dashboard CSV export. Our dashboards cover the same event types Mailgun reports (delivered, opened, clicked, bounced, complained, unsubscribed, dropped, failed) plus per-IP reputation data and Postmaster Tools integration that requires dedicated IPs to be meaningful.
For teams that depend on Mailgun's analytics for product analytics workflows (passing event webhooks into Mixpanel, Segment, or a data warehouse), the webhook format compatibility means most of these integrations work unchanged after migration — the event taxonomy aligns and the payload structure is similar enough that schema mapping is a small effort rather than a rewrite.
Is there a contractual minimum or long-term commitment?
+
Month-to-month on all standard plans with 30-day notice for cancellation. Annual contracts available with 10-15% discount if pricing certainty matters. No trial-then-paid-then-overage waterfall like SendGrid; no Flex-style pay-as-you-go that could double overnight. The plan you choose is the price you pay for the volume that plan covers, and overage discussions happen before they appear on a bill.
For migration projects specifically, the first 60 days are structured so that you can route back to Mailgun at any time if the new infrastructure doesn't meet your standards. We don't lock you in during the period when you're still validating.
Open the migration conversation
Tell us about your current Mailgun setup: which plan tier (Flex, Foundation, Scale, Growth), monthly volume, dedicated IPs in use, region (US or EU), whether you use inbound routing, and what specifically prompted you to look at alternatives. We'll come back with a sized proposal, a migration timeline, and the inbound-routing-specific guidance for your workflow.