Skip to content
steelabs

Template

Performance budget and accessibility remediation record

Record comparable performance observations and practical keyboard/accessibility checks without mistaking an automated score for user experience.

Download editable Markdown

How to use this

Useful for: Web teams investigating page speed or accessibility and documenting a targeted remediation.

  1. Keep baseline and follow-up conditions comparable. Document device, network, cache state, build and tool version, and separate laboratory observations from field measurements.

  2. Set a budget around the journeys that matter to users. A single homepage observation does not describe authenticated screens, checkout or another supported device.

  3. Use keyboard and assistive-technology review alongside automation. Record the actual interaction, expected behaviour and limitation rather than treating the absence of automated findings as conformance.

  4. Link each proposed fix to a retest and owner. Where the evidence is not available, leave a blank measurement field and a collection plan instead of inventing a before-and-after improvement.

Scope and limitations: The record contains measurement fields, not invented results. It does not establish WCAG conformance, legal applicability or a guaranteed Core Web Vitals outcome.

Document preview

# Performance and accessibility record

## Measurement scope
- Journey / route / candidate build:
- Device / browser / viewport:
- Network, cache and authentication state:
- Tool and version / observation date:
- Lab observation or field data source:
- Excluded journeys and collection limitations:

## Performance budget
| Signal | Agreed budget | Baseline / conditions | Follow-up / conditions | Evidence |
| --- | --- | --- | --- | --- |
| Largest contentful paint | Agree for the measurement method | Not measured | Not measured | |
| Interaction responsiveness | Agree for field or lab method | Not measured | Not measured | |
| Layout movement | Agree for the journey | Not measured | Not measured | |
| Images, fonts and scripts | Record scoped transfer budget | Not measured | Not measured | |

Keep lab and field results distinct. Record repeated observations and variability; a faster one-off result under a different network is not a supported improvement claim.

## Interaction checks
- Reach the primary action using a keyboard.
- Confirm visible focus and a meaningful control name.
- Trigger a form error and review announcement and recovery.
- Check zoom, content reflow and the supported narrow viewport.
- Review meaningful image alternatives and decorative images.
- Check the relevant journey with an agreed screen reader/browser combination.

## Remediation register
| Observed barrier | Expected behaviour | Fix | Owner | Retest setup / result | Limitation |
| --- | --- | --- | --- | --- | --- |
| Example: error message is not associated with its field | User can identify and recover from the error | Associate the message and review focus/announcement | | | |

## Decision
- Evidence supporting the scoped conclusion:
- Findings still open and accepted limitations:
- Further user or specialist review:
- Next measurement date:

The example row illustrates a remediation task, not a measured defect in a customer system. Use observed evidence before publishing a result.

Read alongside this worksheet

Get help applying itAll resources