Opens in a new tab

Error-budget calculator

Calculate the allowed and remaining failures for a request-based or time-based objective.

Free to use. Save a file draft or explicitly enable storage on this device.

Working draft · v2

Error-budget calculator

Save / export

Current result

Enter the objective and both event counts.

Incomplete or uncertain. Review the fields and warnings.

Review results

Start with your data or try an illustrative example.

File drafts are available without browser storage. Current draft matches the last save or initial state.

Inputs & calculation

Count eligible and bad events from the same period and SLI definition.

Enter the objective and both event counts.

Your policy checkpoint (optional)

Results

Review and export

This is incomplete or has review issues. You can save a draft; exports retain unknowns and warnings.

Error-budget calculator report

Error-budget calculator
User-entered draft; verify against source evidence.
Tool version 2 · https://aiopssre.com/error-budget-calculator/
Name: Untitled

Measurement: Request/event-based SLI
Scope: 
SLI: Not defined
Window: Not recorded
Exclusions: Not recorded
Enter the objective and both event counts.
Policy checkpoint: Not set% consumed
Policy action: Not defined
Owner: Not assigned
A calculation does not authorize a release. Request allowance is not downtime allowance.
MeasureValue
Target (%)99.9
Eligible events
Bad events
AllowedInvalid
RemainingInvalid

What the allowance counts

The selected budget definition determines the units. For requests, a 99.9% target allows 0.1% of eligible events to fail. Of 1,000,000 events, that is 1,000 allowed bad events; 250 observed failures leave 750 and consume 25% of the allowance. Time-based availability uses eligible and bad minutes instead. Its result describes that separate availability definition, rather than converting failed requests into downtime.

Connecting consumption to the service policy

A remaining allowance has meaning only for the population and window entered above. Record which requests or minutes count, what makes an outcome bad, and how missing telemetry is handled. The optional checkpoint then compares consumption with your chosen threshold and keeps the agreed action and decision owner beside the calculation. That makes the report useful in a release review without implying that the calculator can grant approval.

Read the accompanying guide · Source and further reading

Calculation and worksheet review: September 29, 2026.

Full Incident Response Toolkit · Get new guides by email