What this comparison means in 2026
Cloudflare Email Routing forwards or processes incoming email. Cloudflare Email Sending, now documented in beta, sends transactional email through Workers, REST or SMTP. Resend also handles outbound and inbound email. Neither the claim that Cloudflare cannot send nor the claim that Resend cannot receive is current.
If you only need an alias forwarded to your inbox, start with Cloudflare Routing. If you need an application sender, decide who will own templates, delivery diagnostics and event processing.
Cost for a small app and a growing app
| Outbound emails/month | Cloudflare email usage | Resend transactional |
|---|---|---|
| 1,000 | Included with Workers Paid | Free, subject to 100/day |
| 10,000 | $2.45 plus Workers Paid | $20 Pro |
| 100,000 | $33.95 plus Workers Paid | $35 Pro |
Cloudflare's usage calculation includes 3,000 sends and then $0.35 per 1,000. Its required Workers Paid subscription is separate. Resend's $90 Scale plan also covers 100,000, but buys a different feature tier; Pro at $35 is the lower priced volume match. Resend inbound usage shares its transactional allowance, so include received mail in your estimate.
At 10,000 marketing contacts, Resend lists an $80/month marketing plan, separate from transactional sending. Cloudflare's transport cost does not include an equivalent finished contact-management and campaign product. For four newsletters, budget 40,000 recipient deliveries and the software you will use to manage consent and unsubscribes.
A worker and a dashboard solve different jobs
Cloudflare's attraction is deployment locality: sending and inbound processing can sit beside your existing Workers code. Resend's attraction is a dedicated email workflow, including message logs, webhooks, React Email and Broadcasts. Its current Automations documentation also includes custom events, delays and branching.
A team with a messaging service already built may prefer Cloudflare's primitives. A team debugging individual customer receipts may value Resend's email-specific dashboard more than a smaller usage bill. Before choosing the beta, check its current limits, availability and support commitments rather than assuming it matches another provider's contract.
A migration test that catches the expensive mistakes
Use a fresh subdomain and a representative receipt template. Send to accounts you control at several mailbox providers. Check authentication, links, attachments, reply behavior and the events produced by an invalid recipient. This is a functional test; a handful of test inboxes does not establish comparative inbox placement.
Next, simulate a timeout after submission. Verify that your application does not send twice just because it lacks a response. Make your event handler tolerate duplicates and preserve complaint and hard-bounce suppressions when changing providers.
For inbound mail, map every old address and route before moving MX records. Keep your team's mailbox domain separate from a test receiving subdomain. A successful outbound send tells you nothing about whether replies arrive at the right place.
Compare the whole developer workflow
A Cloudflare Worker can be an attractive place to send a message if the business event already happens there. Resend is attractive when the team wants a focused email API and dashboard alongside application-owned templates. The useful question is where developers and operators spend time after the first request succeeds. Trace domain verification, template changes, failed-message investigation and credential rotation in both candidates.
For a Worker-first application, prototype the actual deployment path instead of a standalone script. Confirm that rendering works in the chosen runtime and that secrets are available only on the server. For a Resend integration, evaluate the SDK or HTTP request with the same renderer and business notification ID. Keep the content equivalent so the comparison measures workflow differences rather than a simplified example on one side.
Decide whether you want a sender or an email application
Cloudflare's outbound sending capability should be evaluated as its own service, distinct from Email Routing. Resend also provides a broader email workspace with inbound, Broadcasts and Automations. If marketing uses those features, compare how they would operate after a move to Cloudflare. A low transport bill does not include an automatic replacement for contact management, campaign authoring or journey editing.
Conversely, a product that only sends a handful of code-generated notifications may not need the broader workspace. List the features actually used today and those required for the next planned release. Avoid charging Cloudflare with the cost of replacing tools you do not use, but do not give it credit for features your team would still have to build. That inventory makes the architecture decision more concrete than a binary marketing-versus-transactional label.
Plan for account limits and a concentrated send
A monthly usage allowance does not tell you what happens during a short burst. Suppose a service incident creates notifications to all 10,000 customers. Define how the queue releases that work, how provider errors are retried and how support can see the backlog. Confirm the current limits of the exact Cloudflare sending service or Resend tier you plan to buy.
Cloudflare's outbound service is documented as beta, so verify current eligibility and constraints before moving a critical message category. With Resend, inspect the selected account and domain limits rather than assuming every feature tier differs only in send volume. A successful prototype proves the request shape, not production capacity, support coverage or the behavior of your own retry code under a peak workload.
Make the contact and send calculations comparable
At 10,000 subscribers, a weekly campaign is approximately 40,000 recipient sends across four editions. Transactional activity adds another workload. On Resend, distinguish the marketing contact charge from transactional sends. On Cloudflare, add the required Workers plan and the software or engineering work that will select recipients and manage campaign preferences. Price the same business outcome on both sides.
Also calculate a quiet month and a busy month. An event-driven product may send few messages in a normal week and many during onboarding, billing or an incident. Estimate actual deliveries rather than counting template types. A product with five templates can send millions of messages, while a large template library can generate very little volume. The usage model should come from recent application data or a clearly labeled forecast.
Keep domain and recipient safety portable
Start with a dedicated sending subdomain if that matches your existing domain policy, and preserve the mailbox records used for ordinary staff email. Configure the chosen service's authentication exactly as documented. Domain verification proves control of DNS; it does not establish consent to email every address in the database. Bring hard-bounce and complaint exclusions into the new routing decision before switching traffic.
If the new platform does not store the same audience state, the application must do it. Keep suppression categories explicit: a marketing opt-out, an invalid mailbox and a complaint can require different handling. Do not erase those differences during an export or treat the absence of a record at the new provider as permission to send. Verify that an opted-out test contact remains excluded after a subsequent profile update.
Test observability with a real support question
Ask a teammate to answer: why did this customer not receive an invitation sent yesterday? They should be able to locate the business notification, see whether it reached the provider and interpret the latest delivery status. If answering requires opening several unrelated logs and guessing which request belongs to the customer, include that friction in the comparison.
Retain provider identifiers and a small internal status history. Define how long you need those records and which personal data they contain. Resend dashboard convenience may be valuable to a small support team; a Cloudflare-oriented team may prefer its own operational tools. Either approach needs enough information to distinguish an application bug, a queue delay, a provider rejection and a receiving-server failure without making unsupported claims about inbox placement.
Choose a reversible first deployment
Move a single message type with limited business risk and route every notification through exactly one sender. Keep the previous integration available for new work if the pilot fails, while continuing to handle events for messages already sent. Do not replay every attempted notification when changing the route back. Rollback changes the future sending path; it should not manufacture duplicate receipts or invitations.
Choose Cloudflare when its current service fits a Worker-centered architecture and you accept the operational responsibilities around it. Choose Resend when its email workspace and developer workflow save enough time to justify the total bill. If the existing service works and the price difference is small, improving your own rendering, retries or event reconciliation may deliver more value than switching providers.
What would justify replacing a working integration?
A switch is easier to justify when it removes a measurable operational problem. A Worker-centered team might want to simplify its deployment boundary. A team using Resend might value faster investigation through its email workspace. Name that benefit and test it with the people who will operate the result. Avoid measuring success only as a reduced provider invoice if the new architecture adds recurring development or support work.
Write down the features you would lose as well as those you would gain. For a Resend user, include any contact, inbound and automation workflows currently active. For a Cloudflare user, include the way the sending code fits existing Workers infrastructure. Mark each lost capability as replaced, unnecessary or a reason to stop the migration. This turns a broad provider comparison into a reviewable decision about your actual application and keeps an attractive transport price from concealing unfinished work.
Recommendation
Choose Cloudflare for a Workers-centered application when you can own the surrounding email behavior and accept the beta terms. Choose Resend for its dedicated developer email tooling. Consider Sequenzy when subscriber journeys and billing-driven campaigns are the deciding workflow, while remembering Resend now offers automation too.
Sources and scope
Updated November 3, 2026. We publish Sequenzy and have a commercial interest in this comparison. This is a documentation-based evaluation, not a hands-on benchmark or independent ranking. Prices are USD list prices or explicitly labeled calculations; taxes, promotions and optional services can change the total. Confirm the selected plan before purchase.

