What is an infrastructure dashboard, and why do you need one?
An infrastructure dashboard is the single page that tells you — and everyone watching — whether your project is actually alive. Here's what goes on one, and why.
If you run a side project, an open source library, or a small SaaS, you probably have this problem: the status of your project is spread across fifteen tools, and none of them talk to each other.
An infrastructure dashboard is the page that brings those signals back into one view. Not a BI tool, not a monitoring stack — a shareable surface that answers, at a glance:
- Is the thing up?
- Is anyone using it?
- Is it making money?
- Is the team still working on it?
The four signals
Every useful infrastructure dashboard tracks some version of these four:
Availability. Uptime, latency, recent incidents. Your users already know when you're down — your dashboard should, too. Uptime Kuma, Beszel, and plain HTTP pings all work.
Usage. Pageviews, active users, conversion. Umami, Plausible, or Google Search Console give you the shape of traffic without dragging in the full Google Analytics surveillance kit.
Revenue. MRR, new subscribers, churn. For indie SaaS, Polar and Stripe numbers tell you whether the project is economically alive.
Development. Stars, commits, open PRs, Sentry errors. The pulse of the codebase — are people contributing, are things breaking, is the team shipping?
You don't need all four. You need the ones that answer the question you keep losing sleep over.
Who the dashboard is for
Dashboards fail when they try to be for everyone. Pick one audience and the widget choices fall out naturally:
- For yourself: error counts, deploy status, the metric you're currently moving. No fluff.
- For investors and prospects: MRR, traffic, uptime. The "yes this is a real company" page.
- For contributors: stars, open issues, recent commits, CI status. The "yes this project is alive" page.
- For customers: uptime, recent incidents, roadmap. The "yes we know what we're doing" page.
A single infrapage page can do any of these. Trying to do all four at once gets you the dashboard nobody opens twice.
What you don't need
An infrastructure dashboard is not a replacement for Grafana, Datadog, or Sentry. Those tools dig into individual problems. A dashboard tells you which problem to dig into.
If a widget needs a time range picker, alert thresholds, or a 30-day drill-down — it doesn't belong on a dashboard. It belongs in the underlying tool, and your dashboard should link to it.
Making one
infrapage is one way to build this kind of page. Pick widgets, drag them into a grid, publish the URL. There are others — a static site with embedded badges works fine, as does a Notion page with screenshots if you're willing to update it by hand.
The technology doesn't matter as much as the habit: pick the four signals you actually care about, put them on a page, and look at the page.