The score isn't the finding. The blanks are.
The tool that stops a senior hire being decided in the first twenty minutes and rationalised for three weeks. It names the criteria before the scoring, makes you argue against your favourite, and builds a reference call that produces signal.
Most senior-hire mistakes are made in the first twenty minutes of the first conversation and rationalised for three weeks afterwards. The candidate is impressive, the rapport is good, and from there every piece of evidence gets read as confirmation. The interview stops being an assessment and becomes a courtship.
Hiring Decision interrupts that. It forces the criteria to be named before the scoring, makes you argue the case against your preferred candidate, and structures the reference call so it produces signal instead of a warm character witness. The goal is not to talk you out of the hire. It is to make sure that if you make it, you made it on purpose.
Score the candidate you have already half-decided to hire.
Same finalist, same evidence. Here is the decision, run two ways.
You run a 14-person agency, hiring your first Head of Delivery. One strong finalist: ex-big-consultancy, polished, and you already like them.
You get a thoughtful framework. Define what the role needs, watch for red flags, take references, trust your read but verify. A list of good interview questions, maybe. It mostly agrees they sound strong.
Which is the problem. You walked in liking them, and the answer read every impressive thing as confirmation, the same way you have for three weeks.
It will not discuss the candidate until you name what the role actually needs. Then it scores on evidence, not impression:
- →Build from zero: every example was inside a firm that already had process. Score 2, leaning unknown
- →Respected by the team: no evidence either way. Unknown
- →Director-level client polish: strong and specific. Score 5
The unknowns are the finding. A candidate with three blanks is not strong, they are unassessed. Then it builds a reference script aimed at exactly those blanks, and makes you name the walk-away line before you call: if two references cannot give a real example of them building from scratch, I make the offer a level down or I keep looking.
The cold version told you they sound good. This told you what you hadn't checked.
- ✦Founders making a first senior hire where a wrong call costs months
- ✦Operators weighing two finalists who are hard to separate
- ✦Anyone who can run a good interview but knows that is not the same as a good decision
- →hiring-decision.zip: a native Claude Skill. Upload to Settings → Skills in Claude, or unzip to
~/.claude/skills/ - →hiring-decision-skill.md: plain markdown. Works in ChatGPT Custom GPTs, Gemini, Claude Projects, agent system prompts, anywhere
- →A README inside the zip with install steps for Claude, ChatGPT, Gemini, and the API
What it does with a real question.
I am choosing between two finalists for a Head of Sales role at a 30-person B2B software company. One has a strong enterprise record but has never built a team from scratch. The other has built two small teams but sold at lower deal sizes. References were warm for both. I half-decided on the first before the final interview.
You've already told me which way you're leaning, and the evidence you gave points the other way. At 30 people, the job is to build a repeatable sales motion from near-zero, and only one finalist has any record of doing that. The enterprise candidate's deal sizes are impressive, but they were won inside a company with brand, pipeline, and a team already in place. You don't know whether the results came from the person or the scaffolding. Your instinct fell in love with the polish before it tested the fit.
Warm references on both sides tell you nothing, because candidates choose referees who will be kind. The signal is in what nobody said. Treat the enterprise candidate's missing build-from-scratch evidence as an unknown, not a neutral, and weigh it that way. The team-builder's lower deal sizes are a known gap, but one you can price: either the market moves upmarket later, or you hire an enterprise closer under them.
So my position is that the team-builder is the stronger hire for this stage, unless your next 12 months genuinely depend on landing large accounts immediately.
Next move: before any offer, have a backdoor reference call with someone who reported to the enterprise candidate, and ask what they did in their first 90 days when nothing existed yet. If that answer is thin, your half-decision was a preference, and you can let it go.
The full Tool runs the whole procedure and writes the complete deliverable, not this excerpt.
Written by claude-sonnet-5-5 with this Tool loaded, 2026-10-11. Checked automatically for length and for copied framework text.
No subscription · free updates within v1.x
Works as a Claude Skill (.zip) and as a plain .md for ChatGPT, Gemini & every LLM
Delivered to your email within 60 seconds · Secured by Stripe
Price includes UK VAT where applicable
Before you buy.
How is this different from just asking Claude directly?
Asked cold, Claude tends to agree the candidate sounds strong, because you do. This refuses to discuss the person until the criteria are set, scores only on cited evidence, treats every gap as an unknown rather than an assumption, and tunes the reference call to the specific doubts. It is built to make the hire harder to wave through, not easier.
Will this work with ChatGPT, Gemini, and other LLMs?
Yes. You get a native Claude Skill (.zip) that auto-invokes in Claude, plus a plain .md you can paste into a ChatGPT Custom GPT, a Gemini system prompt, or any LLM. The procedure travels with the file.
Can I share this with my team?
It's licensed for your own personal and professional use. The output it produces is yours to circulate. To put the Tool itself in the hands of a whole team, email hello@authority.md about team licensing.
What if it doesn't work for me?
If the file hasn't been delivered, or arrives defective or empty, we refund in full: see our refund policy.
Hiring Decision is a structured judgement tool. Not employment-law advice, not a substitute for proper references and due diligence.