---
title: Compare
description: How sently fits next to vendor SDKs, Nodemailer, and orchestration platforms.
icon: Scale
source: "README.md"
---

Teams often start with one channel — usually email via SendGrid or Resend — then add SMS and push.
Each new vendor SDK brings its own auth, retries, and error shapes into app code.

<Callout title="The one rule">
  Treat sently as the channel-delivery layer. Put preference centers, digests, workflow builders, and in-app inboxes on top of it — custom code or a tool like Novu, Knock, or Courier — not instead of it.
</Callout>

## What sently replaces

| Instead of… | Use sently for… |
| --- | --- |
| A growing pile of vendor SDKs in call sites | Channel senders + pluggable transports |
| Per-channel retry and error glue | Shared `SentlyError` codes and retry/fallback decorators |
| Nodemailer when you need multi-channel or non-Node runtimes | `createMailer` / `createSMTPMailer` plus SMS, WhatsApp, and push |

## What sently does not replace

| Need | Where it lives |
| --- | --- |
| Preference centers and unsubscribe UX | Your product, or Novu / Knock / Courier |
| Digest / batching product logic | Your product, or an orchestration platform |
| Workflow builders and multi-step journeys | Your product, or Novu / Knock / Courier |
| In-app notification inboxes | Your product, or an orchestration platform |

Those tools (or your own routing) call sently — or any transport — to deliver. sently does not ship dashboards or per-send SaaS.

## Versus a vendor SDK pile

| Concern | Typical stack | sently |
| --- | --- | --- |
| As you add channels | New SDK, new auth, new retries | Same sender factories |
| Failures | Per-vendor exceptions | Stable `SentlyError` codes |
| Reliability | Ad-hoc loops per client | `RetryTransport` + `FallbackTransport` |
| Provider swap | Rewrite call sites | Swap the transport |

## Versus Nodemailer

| Concern | Nodemailer | sently |
| --- | --- | --- |
| Channels | Email | Email · SMS · WhatsApp · Push |
| Runtimes | Node.js | Node · Bun · Deno · CF Workers |
| Module format | CommonJS | ESM only |
| Edge / Workers weight | Full mail stack on import | Tree-shakeable subpaths (~6.3 KB HTTP) |

Bundle size matters most on Workers and cold starts. It is supporting evidence, not the primary reason to adopt.

## Versus Novu, Knock, and Courier

| Concern | Orchestration platforms | sently |
| --- | --- | --- |
| Role | Workflows, preferences, digests, in-app | Channel delivery / transports |
| Hosting | Usually hosted SaaS | Library in your process |
| Pricing model | Often per-send or seat | MIT package; you pay providers |

Use both when you need product orchestration **and** a typed, multi-runtime delivery layer underneath.

## Troubleshooting

<Accordions>
  <Accordion title="Do I need sently if I already use Novu or Knock?">
    Only if you want local, typed transports under your own routing — or to send outside that platform. Many teams use an orchestration tool for preferences and journeys, and a library like sently for direct transactional sends.
  </Accordion>
  <Accordion title="Is sently a Nodemailer drop-in?">
    No. Migrate email with [Migrate from Nodemailer](/docs/get-started/migrate-nodemailer), then add other channels with the same sender pattern.
  </Accordion>
</Accordions>

## Learn more

- [Introduction](/docs/get-started/introduction)
- [Stability policy](/docs/get-started/stability)
- [Support matrix](/docs/get-started/support-matrix)
- [Migrate from Nodemailer](/docs/get-started/migrate-nodemailer)

## Next

<Cards>
  <Card title="Introduction" href="/docs/get-started/introduction" />
  <Card title="Support matrix" href="/docs/get-started/support-matrix" />
</Cards>