Protocol
Attach evidence to every skill you want others to believe
Self-rating is cheap; evidence is harder to fake and easier to discuss.
When it fits
- A profile says 'advanced', 'expert' or 'strong communicator' without inspectable proof.
When to avoid it
- Evidence must respect confidentiality, ownership and security; never export employer or client data to build a portfolio.
Why it matters
For each high-value skill, keep one or more evidence types: delivered artifact, before-and-after result, decision made, defect prevented, metric changed, stakeholder outcome or independent credential where relevant. Record enough context to explain your contribution without exposing confidential material.
Steps
- A reviewer can ask 'what proves this?' and receive a concrete answer.
An example
'Data migration' links to a sanitized methodology, scale handled, validation approach and measurable reconciliation result instead of a proficiency bar.
Check your result
A reviewer can ask 'what proves this?' and receive a concrete answer.
Keep this limit in mind
- Evidence must respect confidentiality, ownership and security; never export employer or client data to build a portfolio.
Evidence and sources
OECD 2025 and 2026 skills work supports transparent recognition of skills and more portable ways to signal learning and capability across jobs.
Skills-first hiring does not remove the importance of experience, regulated qualifications or occupation-specific requirements.
A Skills-First Labour Market · Skills-first labour market summary