Alexandre Bouchard
Introducing Hookdeck Radar: Real-Time Webhook Performance Monitoring
Introducing Hookdeck Radar: Real-Time Webhook Performance Monitoring
We've processed over 100 billion webhooks at Hookdeck, and we've always been curious: what does webhook latency actually look like at the producer level? How fast are Shopify webhooks, really? What's normal versus what's degraded?
Individual developers only see their own webhooks. When Shopify webhooks take 2 seconds to arrive, is that normal? Is that slow? You have no frame of reference. And on those rare occasions when webhooks do slow down unexpectedly—your database looks fine, CPU usage is normal, logs show nothing unusual, but Shopify webhooks are taking 10 seconds instead of 2—you're debugging in the dark. Is it you? Is it Shopify? There's no way to know without comparing to what others are experiencing.
Today, we're launching Hookdeck Radar to solve both. It's a free public dashboard that tracks real-time latency and performance for major webhook producers. We're starting with Shopify webhooks, and you can see what thousands of developers are experiencing with the same producer—no signup required.
Why This Exists
Honestly, this started from curiosity. We process millions of webhooks daily and have always wanted to know what producer latency looks like across the ecosystem. Are Shopify webhooks typically fast? How do they behave during peak events? What's the baseline?
But we've also had moments where this would have been useful. A customer opens a support ticket because their Stripe webhooks are slow. We dig into their infrastructure, check their destination endpoints, review their logs. Everything looks healthy. Then we check our own metrics across all Stripe webhooks and realize: Stripe's delivery is degraded right now. Everyone is experiencing the same slowdown.
When that happens, individual developers can't see it. You only see your own webhooks. When Shopify slows down, you see your Shopify webhooks slow down—but you don't know if it's Shopify or you. By aggregating data across millions of webhooks flowing through Hookdeck, we can show you what's actually happening at the producer level.
Radar solves both: it satisfies the curiosity of understanding producer performance, and it gives you a tool when those rare debugging scenarios come up.
How It Works
Hookdeck processes hundreds of millions of webhooks daily. For Radar, we aggregate anonymized latency data across all webhooks for a given producer—measuring from receipt to successful delivery—and calculate percentile metrics (p50, p90, p99) that update in real-time.
If Shopify webhooks suddenly spike from 2 seconds to 30 seconds across our entire network, Radar shows that immediately. You'll know it's not your infrastructure—it's Shopify experiencing a wide-spread issue. Conversely, if Radar shows normal performance but your webhooks are slow, that's when you investigate your own systems.
All data is aggregated and anonymized. We don't expose individual webhook details, destinations, or any identifying information—just anonymized performance metrics.
Case study: Shopify Webhook Performance
Looking at Shopify's webhook performance over the past week reveals some interesting patterns. Average latency (p50) holds steady around 2.5 seconds—that's Shopify's typical delivery speed. But the p90 and p99 percentiles tell a more nuanced story.
The p90 latency varies between 4-8 seconds, while p99 can spike to 15+ seconds during peak periods. This variance is normal for high-volume webhook systems, but it's helpful to understand baseline behavior.
When your monitoring shows latency matching Radar's p99, the issue is upstream with the producer. When it exceeds Radar's p99, investigate your own systems. During peak events like Black Friday, if Radar shows elevated latency across the board, you know it's the producer scaling under load—not your infrastructure failing.
Getting Started
Visit Hookdeck Radar to view current Shopify webhook performance—no signup required. You can subscribe to alerts via email or webhook to get notified when producers experience performance degradation.
We are planning to add more producers based on demand and traffic volume such as Stripe, GitHub, Twilio and Facebook (Whatsapp). If you want a specific producer tracked, let us know.