On Page Navigation

An Amazon SES Alternative For Teams Without An AWS Engineer

Amazon SES will send your mail for pennies. It will not publish your records, warm your subdomain, or tell you why a receipt bounced. We do all three. $60 a month, 25,000 delivered emails included.
Who this comparison fits
Is Amazon SES the right home for your application mail?
People usually land here having already tried it. SES is cheap enough that the decision looks obvious, and then the account sits in the sandbox for a week, or the bounce rate climbs and nothing is set up to notice. The sending itself is rarely the problem. The problem is that SES ships you an endpoint and considers its part done.
Icon representing deliverability that nobody on the team owns

Nobody on staff owns deliverability

The account exists and the API key works. Somebody published the records once and moved on. SES will keep accepting your mail while your reputation slides, because reporting that back to you was never part of the deal.
Icon representing support sold as a separate AWS plan rather than included

Support is a separate AWS purchase

SES pricing covers sending. Talking to a human about why sending is failing means an AWS Support plan bought alongside it. The month you need help is the month you find out which tier you are on.
Icon representing a search through delivery records for one message

There is no message log unless you built one

SES does not keep a searchable history of what it sent. Events go to SNS and land wherever you point them, which means somebody has to have designed that pipeline first. If nobody did, last month is simply gone.
Platform versus service
What you buy from Amazon SES, and what you buy from us
SES is infrastructure sold close to cost, and it is very good infrastructure. Amazon prices outbound sending from ten cents per thousand emails and runs it in every region they operate. It does exactly what you tell it to do, and nothing you did not tell it to do. What we sell is the telling, the watching, and the fixing, on a sending path of the same class.

Cascadia

We publish the records, warm the subdomain, route your application mail and read the delivery numbers every month. You get an invoice and an inbox that works.

Without Cascadia

You get an endpoint, an IAM policy and a documentation site. The DNS records, the sandbox exit, the warmup, the bounce pipeline and the suppression list are all yours to design, build and remember.

Comparison

Cascadia vs Amazon SES
Here is where a sending platform and a managed service part company. Every row below is about who does the work, not about whose servers are quicker. All of it is covered by our managed transactional email service.

Deliverability is our job, not yours

SPF, DKIM and DMARC published and kept aligned

Sending subdomain warmed before live traffic depends on it

Application mail routed and every template tested

Bounce and complaint rates watched on your behalf

A straight answer about one recipient, on request

Delivery history kept and searchable without building it

Existing reputation damage audited and repaired

Support included at the entry price

One monthly price covering the platform and the work

Generic logo representing a comparable provider.
Amazon SES

Where Amazon SES fits

Amazon SES is the cheapest credible way to send application mail, and it is run by a company that is exceptionally good at running infrastructure. If you have engineers who want direct control of sending, it is the obvious choice and this page will not talk you out of it.
Cascadia Web Services logo

Cascadia Web Services

Where Cascadia goes further

We are built for teams who need the mail to land and have nobody spare to make that happen.

Where Amazon SES is strong

Icon representing sending available from every AWS region
It lives where your application already lives
SES runs in nearly every AWS region. If your application already sits on AWS, sending through it is a short piece of work, and the mail leaves from the same account as everything else you operate.
Icon representing an API connecting an application to a sending platform
An API, plain SMTP, and IAM control
SDKs in every common language, an SMTP endpoint if you prefer it, event destinations through SNS, and permissions handled by IAM alongside the rest of your infrastructure. If your team wants to build directly against a sending platform, it is all there.
Icon representing a very low cost per message sent
Cheaper per message than anything else
A la carte outbound sending is $0.10 per thousand emails, and the Essentials plan is $0.16 per thousand with no monthly account fee. At 25,000 messages that is a few dollars. On raw sending cost there is no argument to have here.
Icon representing a dashboard of email delivery statistics
Reputation metrics in the console
Bounce and complaint rates, the account reputation dashboard and CloudWatch metrics all ship with the service. The numbers are there for anyone who wires up somewhere to send them and then goes and reads them.
None of that stops being true because you are reading a comparison page. If you already run on AWS and have an engineer who will own the bounce pipeline and the warmup, use SES. You will pay a fraction of what we charge and you will not be worse off for it. This page is written for the other situation.

Where our service is stronger

Icon representing ongoing monitoring of delivery numbers
Someone is watching the numbers
Delivery rate and bounce rate get read every month by a person, against thresholds that mean something. Above 99 percent delivered is the normal range for transactional mail. Under 98 gets looked into before you have noticed anything.
Icon representing a setup checklist worked through on your behalf
Setup is ours, not a link to the docs
We publish the records, warm the subdomain, route your application mail and test each template that goes out. You give us DNS access and roughly an hour.
Icon representing a single delivered message being traced
We answer for individual messages
Give us an address and a rough time, and we will tell you whether the message was accepted, bounced or rejected, and what the receiving server said about it. Without kept logs the honest answer to that question is a shrug.
Icon representing a damaged sending reputation being repaired
Broken reputation gets repaired
If the domain has been mailing bad addresses for years we audit what is published, correct it, request delisting where you have been flagged, and clear out the addresses causing bounces. Expect weeks rather than days.
If your mail matters more than your interest in email infrastructure, buying the outcome costs less than staffing it. The same reasoning runs through the rest of our managed email services.

Onboarding process

Moving your sending off Amazon SES in four steps

The work here is mostly ours and mostly DNS. The one part nobody can rush is warming, because mailbox providers form their opinion of a new sending domain slowly and on their own schedule.

1

We audit what is published today

We read your current SPF, DKIM and DMARC records, look at what the domain has actually been sending, and check whether anything has been flagged anywhere. Sometimes this turns up the real problem on day one.

2

We publish records and build the subdomain

A dedicated sending subdomain goes up with authentication aligned across all three records. Your primary domain keeps its own reputation, held separately from anything your application does.

3

We route your mail and test every template

Your site starts sending through the authenticated path instead of the web server. Each template gets sent and read by a person, WooCommerce order mail included, rather than assumed to render.

4

Volume rises and we watch it

Sending climbs over two to three weeks while providers learn the domain. Then it gets boring, which is the whole point, and you get a monthly delivery number instead of a support ticket.

Testimonials

Don't Take Our Word For It

What changes when you move your sending off Amazon SES
The stories are variations on one theme. The platform was never the problem. Nobody owned the strip of ground between the platform and the inbox, which is exactly what our transactional email service is for.
A laptop showing a dashboard that nobody has acted on

How this plays out

A password reset that quietly stopped arriving

A store sends a few thousand receipts a month without incident. Then resets start landing in spam, and support hears about it from a customer rather than from a dashboard. Nobody ever wired up event destinations, so there is no log to consult at all. The trouble started a fortnight ago and there is no record of the fortnight.
A man at a laptop working out how long something has been going wrong

The other version of this

A sending domain carrying years of bad addresses

An older site collected addresses with no validation and mailed all of them anyway. Delivery has been drifting down for a year without anyone plotting it. The fix is not a plan upgrade. It is an audit and a delisting request, then several weeks of rebuilding a reputation that took years to spend.
99%

Delivered rate we work to

For transactional mail, above 99 percent delivered is the normal range. Anything under 98 gets investigated rather than averaged into the quarter.
25

Thousand delivered emails included each month

Twenty five thousand delivered messages sit inside the $60 monthly price. Past that it is one dollar per additional thousand, on the same invoice, with no upgrade conversation attached.
2 to 3

Weeks of warmup, typically

Volume rises gradually over two to three weeks so providers learn the domain before live traffic depends on it. The records themselves go up within a day or two of getting DNS access.

Amazon SES alternative and managed transactional email FAQs

Frequently Asked Questions

What is an Amazon SES alternative?
Any other way of getting your application mail delivered. That covers rival platforms such as Postmark, Mailgun or SendGrid, and it also covers managed services like this one, where the sending is run for you instead of handed over as an endpoint and a documentation link.
How is Cascadia different from Amazon SES?
SES gives you a sending endpoint and expects you to build everything around it. New accounts start in a sandbox that will only send to addresses you have verified, until you request production access and are granted it. We handle that, along with the sending domain, the authentication records, the warmup schedule and the monthly delivery numbers, which are ours to get right and ours to explain when they slip.
How does the pricing compare?
It does not compare, and we are not going to pretend it does. SES charges $0.16 per thousand emails on the Essentials plan with no monthly account fee, or $0.10 per thousand on a la carte pricing. Twenty five thousand messages comes to about four dollars. Cascadia is $60 a month for the same volume. You are not paying us for the sending. You are paying for the setup, the authentication, the warmup, and the person who reads the bounce rate every month.
At what volume does Amazon SES become cheaper?
Immediately, and by a wide margin. SES is cheaper than us at every volume, including the first email you send. Sending runs $0.10 to $0.16 per thousand against our effective $1.00 per thousand once you are past the included allowance. What does not change at any volume is that somebody still has to run it, and an hour of an engineer's time each month costs more than the difference.
Does Amazon SES include support?
Not with the service. SES pricing covers sending, and reaching an AWS engineer about a delivery problem means an AWS Support plan bought separately from it. Here support is included at $60, because the support is the product.
How far back can I look at what was sent?
With SES that depends entirely on what somebody built, because the product has no searchable message log. Sending events publish to SNS and go wherever you route them, so if that was never set up, the history does not exist anywhere. We keep your delivery history and answer questions against it as part of the service, which matters more than it sounds until the first time somebody asks about last month.
Do I need a dedicated IP?
In our view most senders under a few hundred thousand messages a month do not, because a dedicated IP with too little volume behind it builds reputation slowly and can perform worse than a well-run shared pool. SES will sell you a standard one at $24.95 a month, or a managed one from $15 a month plus a per-message fee. We make the call on your actual volume rather than on what is easy to add to a bill.
Can you move me off Amazon SES without downtime?
Yes. The new sending path gets built and tested alongside the old one, and your application only switches over once messages are landing. Nothing is turned off before the replacement is demonstrably working.
Will my emails still go out during warmup?
Yes. Warming controls how fast volume climbs on the new domain, not whether your mail sends. Plan on two to three weeks before everything is running at full volume, and expect us to tell you where you are in that ramp rather than leaving you to guess.
What if my domain reputation is already damaged?
That is a repair job rather than a setup job, and it takes longer. We audit what is actually published, correct the records, request delisting anywhere the domain has been flagged, and clear out the addresses causing bounces. If the domain has been mailing bad addresses for years, expect a few weeks before the numbers settle.
Does this replace Amazon SES or sit on top of it?
It replaces the part you were operating. We own the sending path end to end, so there is no platform account left for anyone on your team to administer, and no second invoice to reconcile against the first.
Can I keep my existing templates?
Yes. We test each one during setup rather than assume it renders, and anything built inside a previous provider's editor gets rebuilt on the new path. The ones that break are almost always the oldest ones nobody has opened in two years.
What happens to my suppression list?
It comes with you. An accumulated list of addresses that bounced or complained is worth real money in reputation terms, and starting clean without it means mailing known bad addresses on your first day at full volume.
Is this only for WordPress and WooCommerce sites?
No, though those are the common case. Anything that can send through an authenticated path can be routed, and the setup work on our side is much the same whether the mail comes from WordPress, a custom application, or both at once.
How quickly does setup start?
Records go up within a day or two of getting DNS access, and templates are usually done the same week. The warmup sets the real timeline, so the honest answer to when everything is finished is two to three weeks.
What if I want to leave?
Thirty days notice ends it and everything is portable. The DNS records were always on your domain because they were always yours, and your templates, suppression list and sending history come back in a standard format. We will also tell you what to point your site at next so the mail does not simply stop.
​Contact

Ask Us Anything

We’d love to hear from you!