Quality of hire examples: six worked scores for different roles

A quality of hire score combines a few outcomes for one hire into a number from 0 to 100, using components and weights that suit the role. The examples below show six of them worked in full: an account executive, a support agent, a software engineer, a warehouse associate, a hire who left at month seven, and an agency placement. Every name and number is invented. The point is to show how the components change by role and how each score leads to a decision.

The method behind the examples (choosing components, the survey, and the conversion rules) is explained in how to measure quality of hire, and alternative formulas are compared in the quality of hire formula, worked in full.

Conversion rules used in every example

• 1–5 ratings (manager survey, performance review): (rating − 1) ÷ 4 × 100, so 1 = 0, 3 = 50 and 5 = 100.
• Retention: still employed at the checkpoint = 100; left = 0.
• Ramp or productivity milestone: met on or before target = 100; minus 20 per month late (or per week late for roles measured in weeks), floor 0.
• Percent-of-target measures (quota, productivity rate): the percentage, capped at 100.
• Weights add up to 100% and are set per role family before anyone is scored.

The manager ratings come from a short structured survey; the quality of hire survey template has one ready to use.

Example 1: account executive

Sales has an objective output, so it carries the most weight.

What it tells you: a strong hire whose ramp ran a month late. If most AEs from the same period also ramped late, the ramp target or the onboarding is the issue, not the hiring.

Example 2: customer support agent

Support has a quality measure (QA scores on reviewed tickets) and a clear milestone: handling the queue without a buddy.

What it tells you: the objective measures are strong and the manager is lukewarm. Read the survey's free-text answers. If the manager wrote that Sam escalates too readily, that is a judgement competency the interview could test with a scenario question, and it is worth checking whether other support hires show the same gap.

Example 3: software engineer

Engineering output is hard to count fairly, so the 12-month performance review carries the most weight, with a concrete early milestone.

What it tells you: a fast start (90 at 90 days) and an average year (50 at 12 months). That gap is common and is the reason to collect both readings. Ask the manager what changed. If Alex was strong on well-defined tasks and weaker once the work became ambiguous, the interview may be testing execution but not scoping; a design or problem-framing exercise would test the second.

Example 4: warehouse associate

For high-volume hourly roles, use components the site already records for everyone, and make decisions on cohort averages.

The cohort view is what matters here. Suppose 40 associates started in March, 29 were still there at 90 days with an average score of 88, and the 11 who left averaged 30 (retention 0, partial attendance and productivity). The cohort average is (29 × 88 + 11 × 30) ÷ 40 = (2,552 + 330) ÷ 40 = 72.1. Retention drives almost all of the gap between the individual and the cohort, so for this role the lever is early attrition: a realistic job preview, shift expectations stated at the screen, and a first-week check-in. See first-year attrition rate for how to follow those leavers.

Example 5: a hire who left at month seven

Riley, an operations analyst, resigned at month seven, before the 12-month review. The rule, set in advance: retention scores 0, the missing performance component is dropped, and the remaining weights are scaled back up to 100%.

What it tells you: Riley did the work well and still left. The exit interview reason matters more than the score. If Riley said the role was more reactive than described, the job description and the interview's picture of the job need to change, and the next candidate should hear an accurate account of a typical week.

Example 6: an agency placement

An agency rarely sees the client's performance reviews, so it scores what it can observe.

What it tells you: on its own, very little beyond a satisfied client. Across a year of placements, the same two components, averaged by client and by recruiter, show which clients' briefs you understand and which you do not. The 90-day call that collects the rating is also an account management call.

The six side by side, and why not to rank them

The warehouse associate did not outperform the engineer; they were measured on different things. Compare scores within a role family, or compare a family's average with its own history. A table like this one is only useful for seeing how the components differ.

Linking each score back to the interview

The score is worth collecting because it can be checked against what the interview predicted. For each of the examples above, the useful question is the same: which competency explained the outcome, and what did the candidate's interview record say about it? That only works if interview scores were stored per competency with the evidence behind them, as in a structured interview scorecard. Interview Signal writes scorecards where each score quotes what the candidate said, which makes this look-back quicker, but any scorecard that records evidence rather than impressions will do.