Courier.

San Francisco · Filed from 140 2nd Street

Somewhere right now, a password-reset email lands in an inbox. A delivery alert buzzes a phone. A "you were mentioned" badge lights up inside an app. None of those moments feel like infrastructure. All of them are. And a growing share of them quietly pass through Courier first.

Courier sells the least glamorous product in software: the notification. The thing users barely notice when it works and curse loudly when it doesn't. The company's pitch is unfashionably simple - notifications are universal, they are a pain to build well, and almost no engineering team should be building them from scratch. So Courier built the layer once, behind a single API, and rents it out to everyone else.

The Problem They Saw

A solved problem nobody actually solved

Here is the uncomfortable truth Courier noticed: every app needs to tell its users things. New comment. Failed payment. Shipment out for delivery. And every engineering team, independently, rebuilds the same machinery to do it. They wire up SendGrid for email. Twilio for SMS. A push service for mobile. Slack for internal alerts. Then they glue it all together with brittle code, a tangle of templates, and a prayer.

It works, more or less, until it doesn't. A template breaks. A channel goes down. Marketing wants to edit copy but has to file a ticket and wait for a deploy. Multiply that across email, SMS, push, in-app and chat, and "just send a notification" becomes a quarter of someone's roadmap. The problem looks solved from a distance. Up close, it is a recurring tax on every product team in the world.

The Founder's Bet

Build it once, so they don't have to

Troy Goode founded Courier in 2019 and took it through Y Combinator's Summer 2019 batch. The bet was that notifications deserved to be infrastructure - a category, not a side feature. If Stripe could turn payments from a months-long integration nightmare into a few lines of code, why couldn't someone do the same for messaging?

It is the kind of idea that sounds obvious in hindsight and slightly mad at the time. Plenty of companies were happy enough duct-taping their own notification stacks together. Convincing them to hand that over required proving two things at once: that Courier could be more reliable than their in-house version, and that it could give non-engineers control without handing them the keys to production. Investors bought the thesis. The product had to earn the rest.

Four numbers that explain why a "boring" notification company keeps getting funded.

The Product

An API for engineers, a canvas for everyone else

Courier ships two things that rarely live in the same product. For developers, there is the API: send a notification once, and Courier decides how it gets there - routing across email, SMS, push, in-app, chat and Slack, while plugging into the providers teams already use, like SendGrid, Twilio and Postmark. For the people who actually write the copy, there is a drag-and-drop designer that turns a notification into something you can assemble in minutes, no deploy required.

The unglamorous toolkit that lets a team delete a few thousand lines of glue code and never miss them.

The clever part is the routing. Courier can weigh signals - is the user online, which channel has earned the most engagement before - and pick the path most likely to actually reach a human. Alongside the 2022 Series B, the company added a mobile notifications SDK, pushing further into the in-app experience where so many notifications now live.

The Proof

When the customers are the ones who could build it themselves

The most telling thing about Courier's customer list is who is on it. New Relic. LaunchDarkly. Lattice. Vanta. Contentful. ApartmentList. These are companies with serious engineering benches - the kind that absolutely could build their own notification stack. They chose not to. That is the whole argument for infrastructure: even teams who can build it would rather not maintain it.

Each bar is an investor deciding that notifications are, in fact, a real business.

There is a wink hiding in the cap table, too. Both Slack Fund and Twilio Ventures backed the Series B - two companies whose own channels Courier helps orchestrate. When the channel owners invest in the layer that sits above them, the category has arrived.

The Mission

Make notifications delightful, of all things

Courier titled its Series B announcement "to make notifications delightful." It is an almost absurd ambition - delight is not a word usually attached to transactional email. But it captures the point. A notification that arrives on the right channel, at the right moment, in the right voice is invisible in the best way. The mission is not to send more messages. It is to make the ones that matter land, and to spare engineering teams the tedium of getting there.

Why It Matters Tomorrow

More channels, more noise, same need

The number of ways to reach a person keeps multiplying - new channels, new devices, new expectations about timing and consent. That is bad news for any team trying to keep up on its own, and good news for a company whose entire job is to absorb that complexity behind one API. As messaging gets smarter about when and how to reach someone, the orchestration layer only gets more valuable.

So back to that password-reset email landing in an inbox, the alert buzzing a phone, the badge lighting up an app. The user still does not think about any of it. That is exactly the win. Courier's ambition was never to be noticed - it was to make sure the message is, and to take the rest off everyone else's plate. A solved problem, finally solved by someone whose only job is to solve it.