Docs Metrics Report Template (Privacy-First Docs Health)

Docs Metrics Report: [Documentation set] [Period]

Monthly or quarterly health report for a documentation set: an executive summary, the measurement approach and privacy rules, and six measure groups (coverage, freshness, quality, findability, usefulness and support impact), followed by a scorecard with targets, actions and the metric definitions. No formal standard prescribes docs metrics; the structure is aligned with the measurement and quality-evaluation activities in ISO/IEC/IEEE 26511:2018 (requirements for managers of information for users), and accessibility measures refer to WCAG 2.2 (W3C Recommendation, 12 December 2024). Measurement is privacy-first: aggregate, cookieless and consent-respecting, honouring Global Privacy Control, with no personal data required. Replace every [bracketed] value and every row marked Example.

Document control

Approvals

Revision history

Contents

1. Executive summary
2. Measurement approach and privacy
3. Coverage
4. Freshness
5. Quality
6. Findability
7. Usefulness
8. Support impact
9. Scorecard and targets
10. Actions
11. Metric definitions
12. Glossary and references

1. Executive summary

Example (Corvid Commerce developer docs, Q3 2026, 2026-07-01 to 2026-09-30): Freshness improved after the content audit: stale pages fell from 24.3% to 16.8%. Searches with no results fell from 7.8% to 6.0%, mainly because the secret rotation how-to was published. Page feedback is up to 74.0% positive. Coverage is flat at 87.5% with six released features still undocumented, including partial captures. The next quarter focuses on those six features and on the 37 stale pages.

2. Measurement approach and privacy

2.1 Data sources

2.2 Privacy rules

1. Report aggregates only. No measure in this report is attributed to an identifiable person, account or device.
2. Use cookieless, first-party measurement. Do not load third-party analytics, advertising or session-replay scripts on documentation pages.
3. Respect consent and honour the Global Privacy Control signal: when a browser sends GPC, record nothing beyond what is strictly needed to serve the page.
4. Do not store IP addresses, user agents or account identifiers with search queries or feedback. Strip them at collection, not later.
5. Scrub free-text search terms and feedback comments for personal data (email addresses, names, order numbers) before they reach the report.
6. Suppress any table cell below a minimum count of [10] so small groups cannot be singled out.
7. Any change to collection methods needs privacy lead approval and a note in the revision history. Use the Analytics Tracking Plan (data/analytics-tracking-plan) to document event definitions.

2.3 Data quality notes for this period

3. Coverage

Undocumented features:

4. Freshness

Stale pages by owner:

5. Quality

Accessibility issues by WCAG 2.2 success criterion:

Automated checks find only part of the barriers. Pair these counts with periodic manual evaluation in the Accessibility Audit (quality/accessibility-audit).

6. Findability

Top zero-result terms (after scrubbing personal data; terms below the minimum count are suppressed):

7. Usefulness

Lowest-rated pages (minimum 20 responses):

8. Support impact

Treat ticket deflection as an estimate: a reader who solved a problem with the docs rarely tells you. Report the self-service count as a lower bound, never as savings presented as fact.

Ticket themes that need docs:

9. Scorecard and targets

Status rule: Green when the measure meets the target, Amber when it is within [5] percentage points of the target, Red otherwise.

Status check for the example rows: coverage misses by 7.5 points (Red); stale pages exceed the target by 6.8 points (Red); pages with accessibility issues exceed it by 5.7 points (Red); internal broken links stand at 3 against a target of 0 (Red); zero-result searches by 1.0 point (Amber); positive feedback misses by 6.0 points (Red); task success by 5.0 points (Amber); docs-tagged tickets by 2.0 points (Amber). Recheck every status the same way before publishing.

10. Actions

10.1 Actions from the previous period

10.2 New actions

11. Metric definitions

Worked calculation (Example, DM-FND-01): 1,104 zero-result searches divided by 18,400 searches equals 0.060, reported as 6.0%.

12. Glossary and references

12.1 Glossary

12.2 References

Standard and licence note

Structure aligned with the measurement and quality-evaluation activities named in ISO/IEC/IEEE 26511:2018; no ISO text, table or definition is reproduced, and no formal standard defines the measures in this report. Accessibility measures cite WCAG 2.2 success criteria by number and short name only. The privacy approach references the Global Privacy Control specification, which is a W3C Working Draft and may change. Diátaxis type names (diataxis.fr, CC BY-SA 4.0) are used for classification only. All example data refers to a fictional company. This template is original text written for this collection. Check the current official versions of every cited standard and specification before use. Template licence: MIT.