Dashboards
Operational dashboards built from any metric AccelerUp holds.
Analytics → Dashboards builds views across every source AccelerUp holds — services, servers, Cloudflare, logs, errors, databases, deploys.
Tiles
| Tile | For |
|---|---|
| Timeseries | A metric over time |
| Single value | A number with its period-over-period change |
| Status grid | Many things, current state |
| Table | Top-N |
| Text | Context, links, what to do |
That last one is underused. A dashboard with a line of text saying "if the red number is above 5, page the payments on-call" is worth more than another chart.
Domain health
Dashboards → Domain health is a prebuilt view per domain: DNS, certificate, Cloudflare status, origin health and recent errors in one place.
Certificate expiry is on it deliberately. It is the failure that is completely predictable, entirely preventable, and still takes sites down every year.
Building a good one
- One question per dashboard. "Is checkout healthy" is a dashboard. "Everything" is a wall.
- Most important at the top left. People read one screen and leave.
- Absolute time ranges during an incident. A relative range reinterprets itself while you are discussing it with somebody.
- Say what "bad" looks like. A chart without a threshold line requires the reader to already know the answer.
Dashboards are not alerts
A dashboard tells you when somebody is looking. If a number on it matters enough to check every morning, make it an alert as well.
Sharing
Dashboards are shared within the organisation. Tag them — a flat list of forty dashboards is a list nobody uses.
Where this behaviour lives: frontend/src/app/dashboards/. If the code and this page disagree, the code is right — please fix the page.
Part of Analytics and cost — Dashboards, SLOs, reports and what the infrastructure costs.