{
  "id": 20935789,
  "title": "We Designed a Monitoring Stack, Then Decided Not to Build It",
  "url": "https://urgent.news/2026/11/20/we-designed-a-monitoring-stack-then-decided-not-to-build-it",
  "topic": "tech",
  "section": "Tech",
  "published": "2026-11-20T19:06:50.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/fettekette/we-designed-a-monitoring-stack-then-decided-not-to-build-it-216h"
  },
  "original_language": "en",
  "account": "The plan for continuous frontend performance monitoring for 30 to 40 WordPress sites involved using sitespeed.io runners on Hetzner Cloud, a time-series store, artifacts in object storage, Grafana on top, all set up through Terraform. After working through the architecture decision records (ADRs) on runner topology, sizing, and a two-tier regression detector with budgets and baseline deviation, the fourth ADR questioned whether building this stack was worth it, ultimately retiring the entire build. Compared to buying Grafana Cloud Synthetic Monitoring, the self-hosted stack would run around €4,900 in year two, with only €1,300 being infrastructure costs. The break-even point for building the stack would be around 600 URL profiles, roughly 100 sites, making it about two and a half times the current estate. The mistake in earlier calculations was not considering the cost of own time, which was implicitly priced at zero but should have been priced at €800 per day. The self-hosted option's break-even point depends on scale, and only makes sense at a scale of around 600 URL profiles, or about 100 sites. The decision record for the build should include this condition, along with the fact that it is not a decision to be made, but rather a build instruction for when the condition is met. Arguments for building the stack included data sovereignty and the learning effect of building a Terraform/Hetzner stack, but these arguments did not survive as they were not relevant to the current scale of operations.",
  "summary": "The plan was a good plan. Continuous frontend performance monitoring for 30 to 40 WordPress sites: sitespeed.io runners on Hetzner Cloud, a time-series store, artifacts in object storage, Grafana on top, all of it in Terraform. Three architecture decision records worked through runner topology and sizing, the split between time series and artifacts, and a two-tier regression detector with budgets…",
  "key_points": [
    "Planned monitoring stack for 30-40 WordPress sites",
    "Architecture decision records questioned stack viability",
    "Self-hosted stack cost €4,900/year two, break-even at 600 URL profiles"
  ],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}