Cross-cutting

Interactive tools

Reading mode

Some mechanisms are easier to grasp by playing with them than by reading about them. The tools run entirely in your browser, without sending any data: those that remember your answers (RACI, NIS2, solution scorecard) keep them in this browser only. Every assumption is shown and editable, and I encourage you to replace them with your own figures.

Reliability and SLOs

Error budget and burn rateSet an SLO, trigger an incident, watch the budget melt and multi-window alerts fire (or not).

Quality SLO and the Wilson boundA quality score measured on a sample: is the SLO proven, violated or merely hoped for?

Telemetry cost

Cardinality explosionAdd a label to a metric and watch series count, memory and the bill jump orders of magnitude.

Tail samplingAt equal volume, tail sampling keeps every error; head sampling loses most of them.

Cost is decided at the CollectorThe bill before and after each lever: labels, aggregation, sampling, filtering, cold routing.

For leadership, finance and management

MTTR business caseWhat faster detection and recovery are worth: return on investment, payback period, sensitivity.

GenAI economic exposureQuantify in tokens, then in euros with your prices, the cost of not monitoring an AI service.

Alert fatigueAlerts, noise, night wake-ups, rotation size: is on-call sustainable?

Governance and compliance

RACI matrixTwelve activities, seven roles: one accountable owner per line, load per role, CSV export.

NIS2 Article 21 self-assessmentThe ten measures, the evidence telemetry can provide, your status and a prioritized action plan.

Solution scorecardThree options, six weighted criteria, the same requirements for SaaS, open source and hybrid.

High performance computing

Straggler effectA training step lasts as long as its slowest GPU: lost efficiency, wasted GPU-hours, cost.

Going further

Interactive diagramsTrace explorer, clickable architecture, animated Collector pipeline and training step.

Maturity self-assessmentPlace your organization on the four dimensions and leave with an action plan.

BridgesWhat this page changes beyond the technical team.

BusinessValue, cost, risk: Each simulator turns a technical decision into figures, then into euros with your prices: a user_id label into billed series, a sampling policy into stored gigabytes, a noisy on-call rotation into paid hours. These are figures to put in front of the finance department, assumptions included.
OrganizationWho decides, who owns: The RACI matrix and the solution scorecard are filled in during a workshop with the teams involved: they settle who decides (cardinality budget, alert thresholds, choice of a solution) before the first incident, not during it.
PeopleTeams, on-call, skills: The alert fatigue and error budget simulators provide figures to argue for fewer night-time pages; playing with a mechanism also helps a newcomer understand it before their first on-call shift.
Business linesProduct, finance, risk, customer: The MTTR business case speaks to leadership, the NIS2 self-assessment to the CISO and compliance, the scorecard to procurement: each tool displays a result that reads outside the technical team.

Why bridges matter →