# Who should make the final hiring decision, and how? > One named person should make the call, after everyone has rated the finalists independently against the same criteria. Who decides, who advises, why ratings come before discussion, and what to do with two great finalists or none. David at SortTheStack, October 19, 2026. Original: https://sortthestack.com/blog/making-the-final-hiring-decision **One named person should make the final hiring decision, usually the person the new hire will report to. They should make it after everyone involved has rated the finalists independently against the same criteria, the ratings have been combined by a rule agreed in advance, and the group has discussed only where it disagrees.** Consensus by conversation feels fairer, but it tends to favour the loudest view and the information everyone already shared. This guide covers who decides, how to run the decision so it uses everything your interviewers learned, and what to do with two equally strong finalists, or with none. For the mechanics of scoring finalists row by row, see [how to compare job candidates side by side](https://sortthestack.com/blog/how-to-compare-candidates). ## Who decides, and who advises Most hiring decisions go wrong on roles before they go wrong on judgement: three people each think someone else is deciding, or everyone believes they have a veto. Settle it before the first interview. This is a split we'd recommend for a small business: | Role | Usually | What they do | | ------------ | ---------------------------------------- | ---------------------------------------------------------------------------------------------- | | **Decides** | The person the hire will report to | Makes the final choice and owns it | | **Rates** | Everyone who interviewed or saw the work | Rates each finalist on the agreed criteria, independently, with evidence | | **Checks** | A recruiter, HR or a second manager | Confirms the process was followed: same criteria, same questions, ratings in before discussion | | **Approves** | The owner or whoever holds the budget | Approves the offer terms, not the choice of person, unless agreed otherwise | Two rules make this work: - **Vetoes are for facts, not feelings.** Anyone can stop a hire by showing a must-have isn't met, or a fact that changes the picture, such as a licence that has lapsed. Nobody can stop one with "I just didn't get a good feeling." If that feeling has a reason, it can be written down against a criterion. - **The person who decides should not be the only person who rated.** One person's ratings carry one person's blind spots. ## Collect ratings before anyone talks The order matters more than it seems. If interviewers discuss candidates before recording their ratings, the first confident opinion tends to anchor everyone else, and the group spends its time on what everyone already knows. A classic experiment by [Stasser and Titus (1985)](https://doi.org/10.1037/0022-3514.48.6.1467) showed how. Groups chose between candidates for a student election. When the facts favouring the best candidate were spread across members rather than known to everyone, groups mostly failed to choose that candidate, because discussion dwelt on the information members already had in common. The setting was a laboratory, not an employer, but the pattern has been studied many times since and it maps directly onto a hiring panel: each interviewer heard something the others didn't. So: 1. **Each person rates alone, in writing, before the meeting.** Same criteria, same scale, with a line of evidence for each rating. The [U.S. Office of Personnel Management](https://www.opm.gov/policy-data-oversight/assessment-and-selection/structured-interviews/) describes this as part of what makes structured interviews work: "All responses are evaluated using the same rating scale and standards for acceptable answers." 2. **Someone collects the ratings and puts them side by side** before the meeting starts. Nobody changes their ratings after seeing everyone else's. 3. **The discussion starts with the rows where ratings differ most.** Where everyone agrees, there's nothing to discuss. ## Combine by a rule, then use judgement on what's left Add up the ratings the way you agreed before the first interview: equal weights per criterion is a sensible default, or must-haves counted twice. This isn't bureaucracy. A meta-analysis by [Kuncel, Klieger, Connelly and Ones (2013)](https://doi.org/10.1037/a0034156) compared combining candidate information by a fixed rule with combining it by overall judgement, and found "consistent and substantial loss of validity when data were combined holistically—even by experts who are knowledgeable about the jobs and organizations in question". In predicting job performance, the difference "translated into an improvement in prediction of more than 50%" for the mechanical methods. People resist this. Industrial psychologist [Scott Highhouse (2008)](https://doi.org/10.1111/j.1754-9434.2008.00058.x) argued that two implicit beliefs get in the way: that near-perfect prediction of job performance is possible, and that experience makes people better at predicting behaviour. He argues that neither holds up, and that together they lead people to trust their own impression over a decision aid. The rule doesn't make the decision for you. It tells you where the evidence points, and makes you say out loud why you'd go against it. Sometimes there's a good reason: an unclear rating that the meeting resolves, or new information from a reference. Write that reason down. ## The decision meeting Thirty minutes is usually enough for two or three finalists: 1. **Ratings on screen**, already combined. No re-rating. 2. **Disagreements first.** For each row where ratings differ, each rater gives their evidence. Not their conclusion: the evidence. 3. **Unclear items.** List anything nobody could rate. Decide whether it matters enough to find out before deciding. 4. **The decider decides**, in the meeting if possible, and says which criteria drove it. 5. **Record it** in a sentence or two, tied to the criteria. ## Two great finalists When two candidates are genuinely close on the evidence, that's useful information: either would probably do the job well. Before you reach for a tie-breaker: - **Resolve the unclear cells.** A tie with open questions isn't a tie yet. - **Look at where each is strong.** Strength concentrated in must-haves beats strength in nice-to-haves. - **Ask for one more piece of evidence** on the criterion that matters most, such as a short follow-up exercise or a reference check on a specific fact. - **Check whether you have more than one job.** Sometimes the honest answer is a second role, or a call to the runner-up the next time one opens. Avoid tie-breakers that have nothing to do with the job, such as who you'd rather spend a Friday afternoon with. That's how "fit" turns into hiring people like yourself; our article on [culture fit and other hiring assumptions](https://sortthestack.com/blog/gen-z-job-hopping-culture-fit) covers the research. ## No good finalists If nobody meets the must-haves, the least bad option is still a miss. Hiring someone who can't do the essentials usually costs more than another few weeks of searching. Look at why the pool came up short: the requirements, the posting, or where you advertised. Our guide to [finding qualified candidates](https://sortthestack.com/blog/finding-qualified-candidates) has five things to check. ## After the decision - **Make the offer quickly.** Strong finalists often have other options. - **Tell the others.** Applicants who interviewed deserve an answer. In Ontario, most employers with 25 or more employees must tell interviewed applicants within 45 days of their last interview whether a decision has been made; see [what a job posting must disclose](https://sortthestack.com/blog/job-posting-disclosure-rules). - **Keep the record.** The ratings, the evidence and the one-line reason are what you'll want in six months, when you look back at how the hire went. Our article on [hiring metrics](https://sortthestack.com/blog/hiring-metrics) suggests a simple way to do that. ## Where SortTheStack fits SortTheStack doesn't make hiring decisions. It reads each application against your requirements and shows the evidence, so everyone on your team starts from the same facts. Finalists can be compared side by side, team members can comment on each candidate, and when someone changes the product's recommendation they add a reason, which is saved to the candidate's decision record. That record can be exported. [See how the screening works](https://sortthestack.com/product). ## Sources - Highhouse, S. (2008). Stubborn reliance on intuition and subjectivity in employee selection. _Industrial and Organizational Psychology, 1_(3), 333–342. [doi.org/10.1111/j.1754-9434.2008.00058.x](https://doi.org/10.1111/j.1754-9434.2008.00058.x) - Kuncel, N. R., Klieger, D. M., Connelly, B. S., & Ones, D. S. (2013). Mechanical versus clinical data combination in selection and admissions decisions: A meta-analysis. _Journal of Applied Psychology, 98_(6), 1060–1072. [doi.org/10.1037/a0034156](https://doi.org/10.1037/a0034156) - Stasser, G., & Titus, W. (1985). Pooling of unshared information in group decision making: Biased information sampling during discussion. _Journal of Personality and Social Psychology, 48_(6), 1467–1478. [doi.org/10.1037/0022-3514.48.6.1467](https://doi.org/10.1037/0022-3514.48.6.1467) - U.S. Office of Personnel Management, [Structured interviews](https://www.opm.gov/policy-data-oversight/assessment-and-selection/structured-interviews/), accessed 19 October 2026.