Skills-Based Hiring: 6 Changes That Separate Doing From Saying

What to do: audit your question set and count how many answers could be satisfied by naming an employer or a qualification. Those are the ones to rewrite first.

3. Ask for Applied Skill Rather Than Claimed Skill

A list of skills on a CV is a claim, and a scored response to a task is evidence. The distinction matters most for candidates whose capability came from somewhere other than a formal program, which is the population skills-based hiring is supposed to reach.

A short scored video response works well here because it shows the skill in use rather than in summary. The candidate explains their reasoning, handles a constraint, and reaches a conclusion, and all three are visible to a reviewer in a way a bullet point never is.

What to do: pick the single capability that most predicts success in the role and build one task around it that every candidate completes under the same conditions.

4. Write Scoring Anchors Before Anyone Applies

This is where skills-based hiring most often collapses back into credentialism, and it happens quietly. Without written anchors describing what a weak, adequate, and strong answer looks like, reviewers fall back on their own sense of what good looks like, and that sense is shaped by the backgrounds they are used to seeing.

Anchors must be written in advance and describe evidence rather than impressions. A reviewer should be able to point at the part of the answer that produced the rating.

What to do: for each capability, write what a 1, a 3, and a 5 look like in concrete terms, and have two reviewers score five candidates independently before you use it on a live requisition.

5. Remove the Credential From the Reviewer's View at First Pass

If the degree still appears on screen during the first scored pass, it still affects the score, regardless of the policy. The cleanest test of whether your process is skills-based is whether the first assessment could be completed without knowing where somebody studied.

This is mechanically simple in most systems, and it is the change that keeps the others in place. It also makes the process defensible, because the rating is demonstrably based on the response rather than the background.

What to do: run first-pass scoring on the response alone, with education and employer fields hidden, and release them only after the score is recorded.

6. Check Whether Your Shortlists Changed at All

The HBS finding is that most organizations believed they had changed something, even as their hiring remained the same. The only way to know which group you are in is to measure the composition of your shortlists before and after.

Track the share of shortlisted candidates without the previously required credential, measured per role family rather than across the organization. An aggregate number hides precisely the pattern you are looking for, since one high-volume role can mask every other.

What to do: compare two quarters of shortlists per role family. If the composition is unchanged, the policy changed, and the screening did not.

Don'ts

• Do not remove the degree requirement and stop there. The research is clear that this is the most common version of skills-based hiring and the one that changes nothing, because the judgment the credential was carrying has to go somewhere.
• Do not replace a degree filter with a years-of-experience filter. Swapping one proxy for another reproduces the same exclusion with a different label, and tenure predicts less about capability than most teams assume.
• Do not let skills be self-reported. A skills checklist on an application form is a claim collected at scale, and it rewards candidates who are confident with the terminology rather than those who can do the work.
• Do not measure adoption by policy. Counting the job adverts you changed tells you about your wording. Counting the shortlists that changed tells you about your hiring.
• Do not assess a capability nobody has defined. If two reviewers cannot agree on what constitutes a strong answer, the score is an opinion with a number attached.

FAQs

1. Is skills-based hiring working anywhere?
   Yes, in the minority of organizations that changed their selection process rather than their job adverts. In that group, the share of workers hired without degrees rose by nearly a fifth, and those hires showed higher two-year retention than their graduate colleagues, suggesting the approach works when implemented rather than announced.
2. Does this mean degrees are worthless as a signal?
   No, and treating the question that way is why many programs stall. A degree is a weak, cheap proxy for a bundle of capabilities, and the argument for skills-based hiring is that assessing those capabilities directly is more accurate, not that the proxy was random.
3. How do we screen for skills at high volume without adding cost?
   Use one scored task per role family rather than a bespoke assessment per requisition, and run it as the first pass instead of as an extra stage. The cost problem usually comes from adding assessment on top of an existing CV screen rather than replacing it.
4. What is the first sign that our process reverted?
   Shortlist composition that looks identical to the year before. That is the measurement the HBS research effectively ran across the market, and it is the one most organizations never run on themselves.