AI is a gift – and a pest.
In 2025, anyone can spin up a role‑perfect résumé in minutes. Tools like ChatGPT make it trivial to tailor wording to any job, and “easy apply” means candidates can submit at scale. Surveys now show 45%+ of applicants use AI to complete job applications, and platforms like LinkedIn report double‑digit growth in application volume year‑on‑year. (eWeek)
The result: it’s common to see postings with thousands of applications. In some cases, literally bots are hammering “apply” on candidates’ behalf. (UW Homepage) Volume goes up; signal doesn’t. Meanwhile, recruiters say their biggest problem with online recruiting is too many unqualified applicants – 62.6% report this as a top challenge. (iHire)
If the playbook is “post a JD and sift,” AI just hands you a faster queue – more noise for your teams, more misses for great talent.
Candidates still pay the friction tax
On the other side of the table, the experience is just as broken.
Every application is “day one” all over again: new portal, new forms, no history carried forward. One survey found 49% of job seekers think applications are too long and complicated, and a third will outright abandon a process that feels clumsy or repetitive. Another report suggests 57% abandon applications due to complexity and lack of transparency. (JobScore)
Now layer staffing on top: land a contract through an agency or MSP and you get new onboarding, new paperwork, new ways of logging time, even if you just did a nearly identical assignment last month. Same person, same skills, same client type – completely new maze.
We’ve optimized employer funnels and left candidates to repeat their lives across systems. That’s not a process; it’s a penalty- and it shows up in your funnel as abandonment, slower starts, and avoidable backfills.
Metadata over marketing: why proof beats promises
In a world of AI‑generated résumés, what actually matters is evidence.
Most AI hiring systems still lean heavily on self‑reported information – titles, bullets, and keywords. (World Economic Forum) When everyone can produce “perfect” wording, there’s no real edge. We don’t have a résumé problem anymore; we have a proof problem.
The real unlock is metadata and transaction history:
- How many hours has this person actually worked in similar roles?
- What outcomes were delivered—features shipped, tickets closed, assets accepted, milestones hit?
- Were timesheets on time? Were SOWs delivered without escalation?
- What did previous clients actually say when the project ended?
For candidates, this becomes portable reputation. Not “I claim I did X,” but “the system can show that I delivered X, for Y client, across Z weeks, with these approvals and this feedback.” That should travel with them.
For clients, it’s risk removed and time saved. You’re not just looking at a nicely worded PDF; you’re seeing a ledger of real work: “This engineer merged 40 PRs tied to production incidents; this designer delivered 12 accepted assets across three sprints; this PM hit four SOW milestones on time.” In a noisy AI world, that kind of evidence is gold.
This is where I think recruiting has to go: one verified profile, backed by transaction history that compounds trust every time someone works.
What “good” should feel like (and why it matters to you)
In that world, the basics are simple:
One verified profile, with skills validated by actual work.
A portable history – hours, outcomes, references – that composes into a track record across assignments.
Transparent matching that explains why a candidate is a fit and how to improve the fit.
Human at the moments that matter.
It should feel as simple as ordering on Amazon: your identity is set, quality is clear, and the next step takes a click. When candidates move through a lane like that, conversion jumps, signal cleans up, and time‑to‑start shrinks.
AI has a clear role here – but it has to clear the road, not crowd it. Used well, AI sources, screens, schedules, and explains matches. Used poorly, it amplifies spam, hides bias in black boxes, and erodes trust. (For context: nearly 88% of companies now use some form of AI in screening, yet over two‑thirds of adults say they’re wary of applying to jobs that rely on AI for decisions. (World Economic Forum))
The difference isn’t the model—it’s the core: one source of truth, verification built into the flow, and a design where agents handle volume while humans keep judgment. That blend earns trust from candidates and hiring managers—and it shows up as throughput.
What we shipped – and the impact for your teams
Against that backdrop, we just launched the new WorkGenius candidate application. It’s purpose‑built to remove friction on the candidate side so you see speed and quality on yours.
The experience is new end‑to‑end:
- Fewer clicks, fewer drop‑offs. A cleaner path from interest to shortlist increases completed applications from qualified talent and reduces ghosting.
- Richer relevance. Discovery now surfaces more precise roles based on a candidate’s skills, history, and preferences, so your shortlists start closer to “right.”
- Simpler timesheets & assignment UX. Admin stops being a blocker; projects keep momentum post‑offer instead of stalling on forms.
- Explainable matching. Candidates can see their match score and what would improve it; your teams see why someone is a fit, not just a number.
- AI that candidates control. They can chat with an AI recruiter to sharpen experience bullets and tailor a résumé to your role with explicit review before sending. This raises quality without inflating noise.
- Human where it counts. Our recruiters stay in the loop for context, calibration, and the final “yes”, because trust still closes great talent.
Under the hood, more of that metadata and transaction history is being captured by default – hours, milestones, feedback – so over time, your shortlists come with receipts, not just marketing copy.
Why this matters to you: higher conversion, cleaner signal, faster starts, and fewer backfills. When candidates aren’t punished by process, they move faster and more honestly; when matching is transparent and evidence‑based, hiring managers decide with confidence. Built‑in contracts, compliance, and time capture keep the runway clear so accepted offers turn into day‑one delivery.
Where we’re going next
Launch day isn’t a finish line. We’re pushing toward:
- Deeper verification (skills tied to real work artifacts and on‑platform outcomes).
- Richer transparency around why a match scores the way it does.
- Greater portability of candidate history across clients and roles.
- More human touchpoints exactly where they make the difference.
Recruiting won’t be fixed by posting more jobs or reviewing more résumés. It changes when we treat signal, speed, and trust as one design problem—starting with the candidate experience, anchored in hard proof of work, and flowing all the way through to your team’s throughput.
Onward,
Daniel
