The Talent Isn't Missing. Your Role Definition Is.

Slow hiring is usually blamed on a talent shortage. The data points somewhere else. Cybersecurity roles already take 21% longer to fill than the baseline (CyberSeek, 2025), but the same year’s research says the bottleneck isn’t a missing candidate, it’s a missing definition: 60% of organizations call skills gaps a worse problem than headcount, and only 17% blame a lack of qualified candidates (SANS, 2026). The req that’s been open for 90 days is rarely empty because the talent left. It’s open because nobody wrote down what the job actually is.

That reframe is uncomfortable because it moves the problem from the market, which you can’t control, to the requisition, which you can fix this week. This article gives you the honest evidence behind the reframe, shows exactly how a vague req slows you down in ways that never show up as a line item, and hands you a concrete role-scoping checklist you can run before you post the next role.

Why do roles take so long to fill? It’s the definition, not the market

The most-cited workforce research of 2026 argues the opposite of the shortage narrative. When the SANS 2026 Cybersecurity Workforce Research Report (947 global respondents) forced organizations to choose, 60% said skills gaps were the bigger problem versus 40% citing staffing shortages, a 20-point gap that had widened from just 4 points a year earlier. Only 17% named a lack of qualified candidates as their main hiring challenge, against 36% pointing to budget limits.

The 21%-longer figure everyone reads as proof of scarcity comes from CyberSeek (June 2025), and CyberSeek is careful: it lists slower recruiting alongside several market factors and never claims vague reqs cause it. So hold the two facts side by side honestly. The number says roles take longer. The independent SANS data says the dominant problem is alignment and capability, not a depleted pipeline. As SANS Chief AI Officer Rob T. Lee put it, “Organizations have people. But those people are overwhelmed, under-resourced, and unable to develop the capabilities they need.”

Cybersecurity is the sharpest example because its roles are the most prone to scope creep and it has the cleanest published data. But Frankenstein reqs and kitchen-sink must-have lists show up in engineering, design, and ops hiring at every startup. The playbook generalizes.

The “Frankenstein role” is the tell

The archetype SANS and the broader 2026 commentary keep describing is one title fused from several jobs. A single “Security Engineer” req owns cloud, GRC, incident response, DevSecOps, and tooling, with AI, zero trust, red teaming, and compliance all listed and none ranked. No single human matches that, so the pipeline looks empty. It isn’t empty. It’s mis-specified. You are sourcing for a person who doesn’t exist and then concluding the talent isn’t out there.

When every skill is “required” and nothing is prioritized, the req stops being a filter and becomes noise. Per the analysis that reframed CyberSeek’s number around clarity, poorly defined roles slow hiring, increase mis-hires, and leave critical functions exposed even when qualified professionals are on the market.

How a vague requisition actually slows you down

A vague req doesn’t fail loudly. It fails as silent rework, a drifting bar, and good candidates rejected on gut calls, none of which appear on a dashboard until the clock has already reset.

The silent-rework loop

Misalignment is invisible until candidates are already in the pipeline. The recruiter sources to one mental model. The hiring manager interviews against a different one. The candidate who clears the recruiter’s bar fails the manager’s unstated bar, and the fix is a mid-search rewrite of the req. That rewrite re-screens the funnel, resets sourcing, and quietly adds weeks. The expectations were real the whole time. They were just never surfaced before posting, so they surfaced as rejections instead.

Picture a team with a 60-day time-to-fill target. Misalignment that nobody named pushes the real search to 90 days, not because of any one bad decision, but because the definition kept changing underneath everyone. That gap is the cost of skipping the conversation that should have happened before the role went live.

The bar drifts between interviewers

With no shared rubric, “we’ll know it when we see it” becomes the standard, and that standard moves. One interviewer weights system design, the next weights culture fit, the loudest voice in the debrief weights whatever they remember most vividly. The same candidate would pass on Tuesday and fail on Thursday. You are not measuring the candidate. You are measuring which interviewer happened to anchor the room.

Good candidates self-select out before they apply

A long, undifferentiated must-have list deters qualified people at the top of the funnel. When candidates can’t tell which requirements are real, many assume they need to match nearly every line, and the ones who could do the job never apply. There’s an often-cited, if anecdotal, observation, originally an internal Hewlett-Packard finding popularized by HBR, that men tend to apply when they meet about 60% of listed qualifications while women wait until they hit closer to 100%. Treat it as illustrative, not gospel. The directional point survives: every “required” line you can’t defend is a filter that removes people you’d have wanted to talk to.

The role-scoping checklist: lock the role before you post

The proven antidote is to surface expectations before you post, not after candidates start failing. Run these four questions with the recruiter and hiring manager in the room, and write the answers down. This is the intake meeting, and it is the cheapest weeks-saver in hiring.

1. What problem does this hire solve? Not the title, the problem. “Our deploys break on Fridays and nobody owns the pipeline” is a role. “Senior Platform Engineer” is a label. If you can’t state the problem in one sentence, you are not ready to post.
2. What are the first-90-day outcomes? Name two or three concrete things this person will have shipped or owned by day 90. Outcomes force specificity that adjectives (“strong,” “seasoned,” “rockstar”) never do, and they double as the success criteria you’ll evaluate against.
3. What are the real must-haves versus nice-to-haves? Rank everything. Three to five true must-haves, and an explicit nice-to-have bucket for the rest. If a skill can be learned in the first quarter, it is not a must-have.
4. Who decides at each stage? Name the owner for every step and what each step is allowed to reject on. “We’ll all weigh in” is how the bar drifts. One decider per stage, with a clear criterion, is how it holds.

Answer those four, and the Frankenstein role usually splits itself: you discover you’ve described two jobs, or that nine of your twelve “requirements” are nice-to-haves, or that nobody can name a single 90-day outcome, which means the role isn’t ready to exist yet.

Must-have vs nice-to-have: stop writing kitchen-sink reqs

Rank requirements, don’t pile them. The fastest fix for a slow req is cutting the must-have list to the three to five things a person genuinely cannot succeed without, and moving everything else to nice-to-have.

A ranked list does two jobs at once. It widens the top of the funnel, because qualified people can now see they’re a realistic fit instead of self-selecting out against a wall of “required” lines. And it sharpens the bottom, because interviewers know which gaps are disqualifying and which are coachable. The kitchen-sink req does the reverse on both ends: it scares off the people you want and gives the people who do apply no way to be evaluated consistently.

A simple test for every line on your must-have list: can you name the stage in your process that will verify it? If you can’t, it isn’t a must-have. It’s a wish.

Map every stage to an explicit must-have

A scoped role isn’t finished until each must-have is tied to a stage that verifies it. This is what turns “we’ll know it when we see it” into a defensible, repeatable filter.

Walk your ranked must-haves and assign each one a verification method. Can they actually do the work? Use a paid work sample or code assignment. Can they communicate a decision under questioning? Use a structured interview with fixed questions. Can the team agree they meet the bar? Use independent scoring before the debrief, not a freeform conversation after it. Every stage should answer exactly one question: which must-have does this verify? A stage that verifies nothing is a stage that only adds days, and long loops are their own way to lose the best candidates, who are off the market fastest. (We’ve written about how many interview rounds is actually normal if your loop has crept past five.)

The scorecard is what holds the bar still across that whole sequence. Reviewers score independently against criteria fixed at intake, so the recruiter and hiring manager stop talking past each other and the standard stops drifting between interviews. If you want the evidence on why this works, structured scorecards roughly double predictive validity over gut-feel debriefs.

How to get recruiters and hiring managers to stop talking past each other

The intake meeting is the single mechanism that prevents misalignment, because it forces the hiring manager’s unstated bar into the open before anyone sources a single candidate. The recruiter leaves with the problem, the 90-day outcomes, the ranked must-haves, and the decision rights. The hiring manager leaves having committed to them in writing.

The trap is treating intake as a one-time conversation. The alignment decays the moment the doc gets filed and the search drifts back to instinct. What keeps it durable is making the intake’s output the actual process: the same ranked must-haves become the stages, the same success criteria become the scorecard, and the same deciders own the steps. When the artifact is the workflow, there’s nothing to forget and nothing to quietly rewrite.

Make the role clarity durable with a process template

This is exactly the gap Kit is built to close. The role-scoping checklist is only as good as how long it survives contact with a real search, and a doc doesn’t survive. A template does.

Kit’s process templates turn the checklist into permanent, ordered stages. The problem the hire solves, the first-90-day outcomes, the real must-haves, and who decides at each step stop being a one-off intake note and become the process itself. Each stage maps to one thing being verified: an application_form, a paid code_assignment with a GitHub repo and a real deadline, a portfolio_upload, a structured live_interview, a team_review, a reference_check, an offer. Every stage forces the question the kitchen-sink req never asks, which must-have does this verify.

The team_review stage is the scorecard built in: async, independent scoring against criteria fixed up front, so the bar can’t drift and the loudest voice can’t win by default. A paid work sample can be a default gate rather than an afterthought, putting the highest-validity signal, real work, early in the process. And the job posting inherits the template’s stage logic, so the public-facing role is tied to an explicit, ordered set of must-haves instead of a buzzword pile. Greenhouse sells structured hiring like this at enterprise prices. Kit ships it at $6 per seat, with the role clarity built into the templates.

The talent is there. Go define the job.

Stop blaming the market and read your own requisition. A role that’s been open for months is usually a role nobody fully defined: a Frankenstein title, an unranked must-have list, and a hiring manager’s bar that lives only in their head. The evidence is consistent. The dominant hiring bottleneck in 2026 is clarity and alignment, not a depleted talent pool, and clarity is the one variable entirely under your control.

Run the four-question intake before your next post. Cut the must-haves to the ones that matter. Tie each one to a stage that verifies it. Then make it durable by encoding it as a process template, so the recruiter and the hiring manager are held to the same definition for the entire search. When you’re ready to turn the checklist into a pipeline that doesn’t drift, start a free trial and build the template before you open the role.