People & Exec SupportFree
Resume Screening and Shortlist Report
One row per candidate showing which requirements are met and missed, with the line from their resume that proves each call.
River's resume screening and shortlist reads a pile of applications against the requirements you actually set and returns one row per candidate. Each requirement is marked met or missed, and next to each call sits the line from that person's resume that produced it. What comes back is a sheet you can filter, a shortlist with a written reason for every name on it, and a separate borderline set with the specific question that would settle each one.
Unlike the screening products that rank for this search, it does not hand you a number. Those return a match percentage or a score out of a hundred assembled from weighted pillars, and a hiring manager cannot check a weighted score against a resume. Evidence can be checked in seconds. A score can only be believed or ignored. The requirements you screen against come from the role itself, and the job description template is where to fix them if they were never written down properly.
Written for founders doing their own hiring for the first time, recruiters covering more open roles than they have hours, and hiring managers who inherited a pile after a posting went further than expected. Reach for it the day applications close, or partway through when the volume stops being readable. If the role has been open for months and screening may not be the real constraint, recruiting funnel analysis settles that question first. Once the shortlist converts, the offer letter is the next document you need writing.
Screening is a selection procedure, whoever does the screening
The tools competing for this search describe themselves as time savers, and one advertises that its scoring is bias-free. That is not a claim any scoring system can make about itself, because bias in selection is measured in outcomes rather than intentions. Under the Uniform Guidelines on Employee Selection Procedures, the four-fifths rule treats a selection rate below eighty percent of the highest group's rate as evidence of adverse impact. A resume screen is a selection procedure. Consistency does not exempt it, and a consistent rule can produce a disparate rate perfectly reliably.
The arithmetic is quick enough to run yourself. Suppose 200 applications arrive and you shortlist 30. Of the applicants, 120 are in one group and 80 in another. If 24 of the first group advance and 6 of the second, the selection rates are 20 percent and 7.5 percent. Divide the lower by the higher and you get 37.5 percent, well under the threshold, on a screen that felt entirely consistent while you ran it. Nobody spots that from a ranked list. You spot it by keeping the rows and doing the division.
This is why the output format is not a cosmetic preference. The same guidelines expect employers to keep records showing the impact of their selection procedures. Where a user has not kept that data, they note, enforcement agencies may infer adverse impact from the absence of the records. A screen that outputs a score and nothing else leaves no record of why anyone was cut. One that outputs the requirement, the verdict and the quoted line leaves a reviewable trail.
How it works
Paste the applications
The resume text for the whole pile, plus what the role genuinely requires as opposed to the posting.
River reads each one
Every candidate checked against every requirement, with the supporting line quoted for each verdict.
Get the shortlist
A filterable row per candidate, a reasoned shortlist, and the borderline set with questions to ask.
Argue in chat
Reopen a candidate you disagree with, or drop a requirement and see who comes back into range.
What you get
- One row per candidate, with every stated requirement marked met, missed, or unclear from the resume
- The actual sentence from each resume that supports each call, so you can overrule any of them
- A shortlist with a written reason per name, not a rank order to take on faith
- The borderline set kept separate, each with the one question that would resolve them
- Candidates whose resume is weak but whose experience is not, flagged rather than dropped
- The requirements that almost nobody met, which usually means the requirement is wrong
Common questions
Why not just give me a score? It would be faster to sort.
Sorting is fast, but a score gives you nothing to check and nothing to defend. The rows are still sortable by how many requirements each candidate met, so you keep the ordering. What you also get is the reason, which matters the first time a hiring manager disagrees with a cut or someone asks why a candidate was rejected.
Does this decide who to reject?
No. It produces evidence and a recommendation, and a person makes every call. That distinction is doing real work rather than covering anyone: the moment a screen becomes automatic, you own its outcomes without ever having seen them. Every verdict here is designed to be overruled in one click, and the borderline set exists specifically to be argued with.
How does it handle someone whose resume undersells them?
That case gets its own flag. A short or badly formatted resume that still contains evidence of the hard requirements is marked as a weak document rather than a weak candidate, and lands in the borderline set with a question. This is the single most common way a good hire gets lost in a large pile, so it is treated as a finding, not a miss.
Can it look at anything besides the resume?
Only what you paste. If you include cover letters, portfolio links or application form answers, those get read alongside the resume and quoted the same way. It will not go and search for a candidate, which keeps the screen to the same material for everyone and avoids pulling in information that has no business in a hiring decision.
What if hardly anyone meets the requirements?
It says so plainly and names which requirement is doing the damage. A requirement that almost no applicant meets is usually a drafting problem rather than a market problem, and seeing the pass rate per requirement is often more useful than the shortlist. Once the shortlist turns into a panel of interviews, interview scorecard synthesis is what turns their separate notes into one decision.
How many applications can I paste at once?
Work in batches you can still review. Thirty to fifty resumes per pass keeps the evidence quotes accurate and the output readable, and you can run several passes against identical requirements and combine them. Running the same criteria across every batch matters more than doing it in one go, because inconsistent criteria between batches is what makes a screen indefensible.
Resume Screening and Shortlist Report
Fill in the form and your workspace opens with the work already underway.