Seven signals. One number you can defend.
Your evidence on a finalist is scattered across panel scorecards, a take-home, an AI score, a video answer and two references. The evaluation report puts them on one page, weighted the way your team actually decides — and shows exactly how much each one moved the result.
Arithmetic you can show a candidate.
Every point traceable to the signal that produced it.
Seven signals, one page
Panel scorecards, assessment results, the AI Fit Score, the video evaluation, references, your team rating and any score you add by hand from a tool Hireall does not run.
Weights that match the role
Set the default once for the company, then override it for the job where the take-home matters more than the panel. Engineering and sales stop pretending they hire the same way.
Coverage and confidence, stated
A candidate with three signals is not the same as one with six. The report says which of the seven it had, what share of the weight that covers, and how much confidence that earns.
A written read, on request
Ask for strengths, gaps, opportunities, risks, the contradictions between your own signals and the questions still worth asking. Two credits, and only when someone clicks.
Visible before you open anything
The score rides along on the pipeline card and the applications list, so a shortlist can be ordered without opening seven tabs per candidate.
Usable as an automation condition
Route on the combined score instead of a single signal. Rules wait until at least two signals exist rather than judging someone on a thin record.
The number encodes your hiring bar
A combined score is only honest if it weighs what your team actually weighs. Set the split in settings, override it on the job that is different, and the report recalculates the moment you save. Missing a signal never costs a candidate points — its weight moves to the signals you do have, and the coverage figure tells the reader what that cost in certainty.
The model writes prose, never the number
The score is arithmetic our servers do. The optional written summary is the only part a model touches, and it is fenced in: no score, no percentage, no ranking, no hire or do-not-hire advice. Names, addresses and phone numbers are never sent — your evaluators appear as E1 and E2, referees as R1 and R2 — so the text is about the evidence rather than the person.
Debriefs that end in a decision
Panels, shortlists and the conversation with the hiring manager.
Five interviewers, one view
The debrief stops being an argument about whose impression counts and starts with a number everyone can take apart.
- Panel disagreement surfaced, not buried
- Blind review still applies inside the report
- Print or export for the meeting
Ordering forty finalists
The list sorts on evidence you already collected instead of on who interviewed most recently.
- Score on the card and in the list
- Coverage shows who is judged on thin data
- Automation can route on the combined score
The hire you have to justify
When a board or a regulator asks why this person, the answer is a document with the weights, the signals and the date on it.
- Every contribution traceable to a signal
- Saved reports show when they went stale
- Exportable as a document
How the score is made
The questions people ask before they trust a combined number.
No, and this is the part worth being pedantic about. The score is a weighted average computed on our servers from signals your team already produced. No model reads it, writes it or adjusts it. The only thing a model does here is write the optional summary text, and that text is forbidden from containing a number.
Its weight is redistributed across the signals that do exist, so a missing assessment never reads as a bad assessment. What changes is the coverage figure and the confidence label, which tell you how much of your intended evidence the score actually rests on.
Yes. There is a company default in settings and an override on any individual job. Values are whole numbers and the report refuses a profile where everything is zero.
Yes. A reviewer who is not allowed to see scorecards does not see them here either; that signal is marked hidden and the score is computed without it, for that reader.
No. The combined score counts as an AI-assisted field, so any rule using it hands a rejection, an offer or a hire to a person for one-click confirmation. A rule also waits rather than acting when fewer than two signals exist.
One score, and the arithmetic behind it
See a finalist scored end to end in a 30-minute demo.
Talk to our team
Tell us about your roles and team size — we'll map Hireall to your hiring process and the plan that fits.
Contact sales Priced per company, not per seat