In short
- Definition: The ease with which a system can be operated and kept running in production, including monitoring, alerting and routine tasks.
- Goal: Make day-to-day operations simple, observable and low-risk.
- Example S.M.A.R.T. requirement: 100% of services with health checks, dashboards and runbooks, and under 1 manual intervention per week.
1. Definition
As a non-functional requirement, operability describes how easily a system can be run and kept healthy in production: monitored, diagnosed, scaled, updated and recovered [1].
Build it so the person on call at 3 a.m. can understand it in five minutes.Key takeaway
It relies on observability (metrics, logs and traces that show what the system is doing), health checks, dashboards, alerts, runbooks and automation of routine tasks. Google’s SRE book recommends four golden signals for monitoring user-facing systems: latency, traffic, errors and saturation [2], [3].
Operability is often forgotten during development and discovered by the operations team after go-live. Writing it as a requirement brings operators into the design.
2. Goal
The goal is simple, observable and low-risk day-to-day operations.
Poorly operable systems need heroics: long diagnoses, manual restarts, fragile procedures and dependence on the original developers.
3. How to Measure It (S.M.A.R.T.)
| Metric | Unit | Typical target | How to measure |
|---|---|---|---|
| Services with health checks, dashboards and runbooks | % | 100% | Operational readiness review |
| Manual interventions | Per week | < 1 | Operations log |
| Time to diagnose | Minutes to find the cause of a typical incident | < 15 | Incident records |
| Golden signals monitored | Latency, traffic, errors, saturation | All four for every user-facing service | Monitoring configuration |
4. Example Requirements
Each requirement below is ready to copy into a specification: specific, measurable, achievable, relevant and time-bound.
- Readiness: “Before go-live, every service shall expose health checks and the four golden signals, have a dashboard and a runbook for each alert, and pass an operational readiness review.”
- Toil: “Routine operations shall require less than 1 manual intervention per week, measured over each month.”
- Tracing: “Every user request shall carry a correlation ID visible in logs and traces across all services.”
5. Questions to Ask Stakeholders
- Who will operate the system, and what tools do they use?
- What must operators see to know the system is healthy?
- Which routine tasks exist, and which can be automated?
- How will operators be alerted, and with what information?
- Who writes and maintains runbooks?
- How will operators be trained before go-live?
6. Common Pitfalls
- Logs without structure. Free-text logs are hard to search. Use structured logs with correlation IDs.
- Alerting on causes, not symptoms. Alert on what users feel; use other signals for diagnosis.
- No operational readiness review. Go-live without runbooks and dashboards shifts the work to the first incident.
7. Related NFRs
- Management — administering and configuring the system.
- Failure Management — detecting and recovering from failures.
- Supportability — diagnosing user problems.
- Deployment — operators need safe, repeatable releases.
- Documentation — runbooks are operational documentation.
8. FAQ
The ability to understand what a system is doing from its outputs, such as metrics, logs and traces.
Latency, traffic, errors and saturation, recommended by Google’s SRE book for monitoring user-facing systems.
Manual, repetitive operational work that could be automated and grows with the system.
9. Acronyms, Initialisms & Terms
Acronyms are abbreviations pronounced as words (like NASA); initialisms are spelled out letter by letter (like FBI). The classification below follows common pronunciation.
Acronyms
| Abbreviation | Stands for | Definition | Ref. |
|---|---|---|---|
| SMART | Specific, Measurable, Achievable, Relevant, Time-bound | Criteria for writing a goal or requirement so that anyone can check whether it has been met. | [4] |
Initialisms
| Abbreviation | Stands for | Definition | Ref. |
|---|---|---|---|
| FAQ | Frequently asked questions | List of common questions with their answers. | [5] |
| ID | Identifier | Unique value used to recognize an item, such as a request or a user. | [6] |
| NFR | Non-functional requirement | Requirement that describes how well a system must work (its qualities, such as speed or security) rather than what it does. | [1] |
| SRE | Site reliability engineering | Approach, started at Google, that applies software engineering to running reliable production systems. | [7] |
Terms
Plain-language definitions of technical and project-management terms used in this article.
| Term | Plain-language definition | Ref. |
|---|---|---|
| Dashboard | Screen that summarizes key measures at a glance. | [8] |
| Deployment | Installing a new version of software in the environment where people use it. | [9] |
| Incident | Unplanned interruption or degradation of a service. | [10] |
| Latency | Delay between a request and the start of the response. | [11] |
| Log | Time-stamped record of events written by software, used for troubleshooting and audit. | [12] |
| Observability | Ability to understand a system’s internal state from its logs, metrics and traces. | [3] |
| Production environment | The live system that real users use, as opposed to test environments. | [13] |
| Runbook | Step-by-step procedure for routine operations or for handling a known problem. | [14] |
| Stakeholder | Person or group with an interest in the system: users, sponsors, operators, regulators. | [15] |
| Trace | Record of one request’s path through all the services that handled it. | [16] |
10. Sources and Further Reading
- [1] “Non-functional requirement.” Wikipedia. https://en.wikipedia.org/wiki/Non-functional_requirement (accessed Nov. 18, 2026).
- [2] Google. “Site Reliability Engineering, chapter 6 ‘Monitoring Distributed Systems’.” Google SRE. https://sre.google/sre-book/monitoring-distributed-systems/ (accessed Nov. 18, 2026).
- [3] “Observability (software).” Wikipedia. https://en.wikipedia.org/wiki/Observability_(software) (accessed Nov. 18, 2026).
- [4] “SMART criteria.” Wikipedia. https://en.wikipedia.org/wiki/SMART_criteria (accessed Nov. 19, 2026).
- [5] “FAQ.” Wikipedia. https://en.wikipedia.org/wiki/FAQ (accessed Nov. 19, 2026).
- [6] “Identifier.” Wikipedia. https://en.wikipedia.org/wiki/Identifier (accessed Nov. 19, 2026).
- [7] “Site reliability engineering.” Wikipedia. https://en.wikipedia.org/wiki/Site_reliability_engineering (accessed Nov. 19, 2026).
- [8] “Dashboard (computing).” Wikipedia. https://en.wikipedia.org/wiki/Dashboard_(computing) (accessed Nov. 19, 2026).
- [9] “Software deployment.” Wikipedia. https://en.wikipedia.org/wiki/Software_deployment (accessed Nov. 19, 2026).
- [10] “Incident management.” Wikipedia. https://en.wikipedia.org/wiki/Incident_management (accessed Nov. 19, 2026).
- [11] “Latency (engineering).” Wikipedia. https://en.wikipedia.org/wiki/Latency_(engineering) (accessed Nov. 19, 2026).
- [12] “Logging (computing).” Wikipedia. https://en.wikipedia.org/wiki/Logging_(computing) (accessed Nov. 19, 2026).
- [13] “Deployment environment.” Wikipedia. https://en.wikipedia.org/wiki/Deployment_environment (accessed Nov. 19, 2026).
- [14] “Runbook.” Wikipedia. https://en.wikipedia.org/wiki/Runbook (accessed Nov. 19, 2026).
- [15] “Project stakeholder.” Wikipedia. https://en.wikipedia.org/wiki/Project_stakeholder (accessed Nov. 19, 2026).
- [16] “Tracing (software).” Wikipedia. https://en.wikipedia.org/wiki/Tracing_(software) (accessed Nov. 19, 2026).
Written by Philippe Huard · LinkedIn · Last updated: November 18, 2026 · Header photo by @ttnspn; inline photo by @tashakostyuk, both on Unsplash.




