Hookdeck
Managed event gateway that sits in front of your webhook endpoints — queuing, retrying, filtering, and surfacing every failure so your integrations don't silently drop data.
Operator's take
Hookdeck has two products: Event Gateway (receive webhooks reliably) and Outpost (add outbound webhook delivery to your own API or SaaS, managed or self-hosted). Most teams reach for it on the inbound side first — when they've been burned by payment webhooks missed during a traffic spike, a third-party integration that sends no replay, or a failed delivery with no trace in any log.
It's a developer-shaped tool — you configure sources and destinations via dashboard or API, point your vendors at Hookdeck's inbound URL, and Hookdeck fans traffic into your actual endpoint with configurable retry schedules, dead-letter queuing, and a full event log. The platform explicitly positions itself as "built for code and agents," which matters if you're running AI workflows that consume webhooks and need guaranteed delivery without babysitting the queue. Not a no-code workflow builder; pairs with one if you need business logic on top.
What it's good at
- Guaranteed delivery with automatic retries — configurable retry policies and queuing absorb traffic spikes without dropping events; every undelivered payload is tracked, not silently lost.
- Event replay — replay any past webhook from the dashboard without asking the sender to retrigger; speeds up debugging and recovery after a bad deploy.
- Centralized visibility — one dashboard shows delivery status, latency, and failure history across all webhook sources and destinations.
- Payload-based routing and filtering — route events to different destinations by header or body content; useful when a single inbound source fans out to multiple services.
- Drop-in infrastructure — no code changes required on the sending side; just change the destination URL to Hookdeck's inbound endpoint.
What it's not
- Not a workflow automation platform — handles the delivery and reliability layer only; anything resembling business logic (conditionals, transforms, multi-step flows) still needs n8n, Make, or a custom handler.
- Not free at meaningful scale — the Developer tier covers 10,000 events/month (dashboard locks on overage until upgrade); production workloads typically land on Team ($39/month) or Growth ($499/month) for longer retention and SLAs.
- Not the right tool if EventBridge already fits your stack — AWS-native organizations with existing EventBridge investment get broader orchestration and tighter IAM there; Hookdeck wins on out-of-the-box observability and no-AWS-dependency.
- Not a drop-in for general-purpose message brokers — optimized for webhook reliability and observability; not a Kafka replacement for high-throughput event streaming architectures, even though it processes billions of events per week at scale.