diff --git a/backlog.md b/backlog.md
new file mode 100644
index 0000000..fa2034a
--- /dev/null
+++ b/backlog.md
@@ -0,0 +1,57 @@
+# Feature Backlog — proxmox-monitor
+
+## Quick Wins (low effort, high value)
+
+### Console Link
+- Add a clickable name link per row linking to Proxmox web console: `/pve/console.html?node=X&vmid=Y`
+- **Status: done (pending push)**
+
+### Manual Refresh
+- Button in dashboard header that calls `mutate()` on all SWR caches
+- ~5 lines of code
+
+### Custom Poll Interval
+- Input field in header to adjust `refreshInterval` instead of env-only config
+
+### Group by Node
+- Collapse/expand VMs/LXCs by node with toggle
+
+### Uptime Duration
+- Format uptime seconds to "3d 4h 12m" via `formatUptime()` helper
+
+### Search / Filter
+- Client-side text input filtering table rows by name/node/status
+
+## Medium Effort
+
+### Storage Usage
+- Show disk usage per Proxmox storage backend (like node cards)
+
+### Status Change Alert
+- Toast/banner when a VM state changes between polls
+
+### Resource Allocation
+- Show vCPU count, total disk size alongside usage stats
+
+### Per-VM IP Addresses
+- Fetch guest network IP addresses (extra API call needed)
+
+### Per-VM CPU/Memory Trend
+- Periodic polling to build a small trend chart
+
+## Larger Features
+
+### Cluster Status
+- HA, quorum, and failover info per cluster
+
+### Network Health
+- Network interfaces and link status per node
+
+### Recent Proxmox Events
+- Audit log / recent events from the Proxmox cluster
+
+### Docker Deployment
+- Dockerfile + docker-compose for easy hosting
+
+### Health Check Endpoint
+- `/api/health` for reverse proxy / uptime monitoring