01 Play 01
Cut the Wish List
Every unnecessary requirement is a candidate who never clicks apply.
- Keep only true requirements. Move everything else to "preferred," or cut it. Research from Harvard Business Review found women, in particular, tend to apply only when they meet every listed qualification. A long wish list filters out strong candidates before a recruiter ever sees them.
- Translate the jargon. If a requirement only makes sense inside your own walls, a candidate outside those walls will skip right past it.
- Trade years for capability. "5+ years" is a proxy, not a requirement. Ask for the actual skill you need instead.
- Scrub the loaded language. Words like "aggressive" or "ninja" skew male. Phrases like "digital native" skew young. Cut both.
Bottom line An honest, specific job description attracts more of the right candidates than an exhaustive one ever will.
02 Play 02
Choose Your Filter
Skills and competencies are not the same thing, and mixing them up is a quiet source of bad fit.
Lead with skills when
- The role is highly technical or tool-specific (engineering, design, finance ops)
- You need to widen the funnel fast, one widely cited study found shifting from qualifications to skills more than doubled an organization's talent pool
- The tools and tasks aren't likely to change soon
Lead with competencies when
- The role is leadership, client-facing, or cross-functional
- The work itself will keep evolving
- Judgment, adaptability, and communication matter as much as any single tool
Bottom line Match the model to the role. Don't force every job through the same lens.
03 Play 03
Let AI Draft, Never Decide
AI is a legitimate shortcut for the time problem when used the right way.
Where it earns its keep
- Scanning hundreds of job titles to flag duplicates and overlap
- Drafting a first pass at a skills section
- Benchmarking your roles against what competitors are posting
Where it quietly backfires
- Reproducing the same biased language it was trained on
- Hallucinating stats, quotes, or requirements with total confidence, a widely cited industry study found a meaningful share of AI chat outputs contain fabricated information
Bottom line AI buys back time. A human protects the brand and the legal exposure. Skip the second part, and the first one doesn't matter.
04 Play 04
Kill the Manual Bottlenecks
If a new job description still starts with an old Word doc, the slow part of hiring isn't the interview stage.
- Start from a template, not a blank page. A library of approved boilerplate turns JD creation into fill-in-the-blank, cutting time from request to posting from days to hours.
- Collect feedback in one place. A login-free feedback link beats chasing hiring managers and legal through email threads that stall out.
- Make org-wide changes in minutes, not weeks. When a policy update needs to touch hundreds of roles, doing it by hand can cost a team a week or more. The fix isn't writing faster, it's removing the busywork between knowing what needs to change and getting it live.
Bottom line The fastest hiring teams aren't writing better job descriptions one at a time. They've removed the manual work standing between them and posting.