Interview Scorecard & Candidate Comparison

About the Interview Scorecard & Candidate Comparison

A structured interview asks every candidate the same questions and rates the answers against the same anchors, so a panel compares people on evidence instead of impressions. Build the interview kit for a role — four to eight job-related competencies with weights, the questions to ask and what a 1, a 3 and a 5 look like — or start from one of five kits written for this tool (any role, software engineer, sales, customer support, people manager) and edit it.

Each interviewer then fills in their own scorecard for each candidate: a rating from 1 to 5 and the evidence for every competency, and a recommendation. The panel comparison combines only the submitted scorecards, flags the competencies where interviewers disagree by two points or more, and orders the candidates by the panel’s weighted ratings with the reasons beside each place. It never decides, never reads a résumé and never rates anything the panel did not; rating fields for protected characteristics such as age, sex or religion are refused. Everything stays in your browser.

Common uses

• Hiring managers running a panel interview for one role with three or four interviewers.
• Small companies that want consistent, documented interviews without a hiring system.
• Recruiters preparing interview guides for hiring managers to print and fill in by hand.
• Teams that want to see where the panel disagrees before the debrief.

Why structured interviews

In a large re-analysis of the research on selection methods, structured interviews came out as the top-ranked procedure for predicting job performance (Sackett, Zhang, Berry and Lievens, Journal of Applied Psychology), and the classic summary of the field by Schmidt and Hunter in Psychological Bulletin compares the validity of nineteen methods. A review of the structured-interview literature by Levashina, Hartwell, Morgeson and Campion in Personnel Psychology covers what structure means, how it reduces bias, situational versus past-behaviour questions and how to build rating scales.

The practical parts this tool puts in place: the same questions for every candidate, behavioural anchors for the ratings, evidence written down for every rating, each interviewer rating on their own before the panel meets, and a combination rule decided before anyone is scored.

How the comparison is worked out

• Competency mean: the average of the ratings in the submitted scorecards (a competency marked N/A by an interviewer is left out of that interviewer’s count).
• Weighted rating = Σ (weight × competency mean) ÷ Σ weight, over the competencies with at least one rating, on the 1–5 scale.
• Weight rated: the share of the kit’s weight that has ratings; under 100% means part of the kit was not assessed, and the reasons say so.
• Disagreement: the highest minus the lowest rating on a competency, flagged at two points or more by default (one or three in step 2).
• Place: candidates in order of the weighted rating; equal ratings share a place. Candidates without a submitted scorecard come last, unrated.
• Strongest / Lowest: up to two competencies each, with means of 4 or more and of 2.5 or less.

Protected characteristics

Rating a candidate on age, disability, gender reassignment, marriage or civil partnership, pregnancy or maternity, race, religion or belief, sex or sexual orientation — the protected characteristics of the Equality Act 2010, section 4 — is discrimination, and the US EEOC makes it illegal to discriminate against an applicant because of race, colour, religion, sex (including transgender status, sexual orientation and pregnancy), national origin, age (40 or older), disability or genetic information. India’s laws and many others protect similar grounds, including caste.

So a competency that rates one of them is refused — “Age”, “Marital status”, “Ethnic background”, “Young and energetic”, “Native speaker” — and a question that mentions one (“Are you married?”, “Do you plan to have children?”) is flagged. When the word is part of the work itself — fault diagnosis for a technician, medication rounds for a nurse, race conditions in software, working with children in a nursery — the competency is kept and flagged as a warning, so you can check it is about the job and not the candidate. The tool has no fields for photos or dates of birth, and blind mode replaces names with codes in scoring, the comparison and every export.

Several interviewers on several devices

Everything is kept in the browser you use. For interviewers who score on their own laptops, set up the kit and the panel, press Save to file and send the file. Each interviewer opens it with Open file, scores, and saves it again. Back on your device, Merge scorecards reads their files: people are matched by the ids in the file, and for each candidate and interviewer the newer scorecard wins — a submitted scorecard is never replaced by an older draft. Files scored with a different kit are refused, so ratings are never mixed across competencies.

Good evidence

Write what the candidate said or did, not your conclusion: “described how she cut the build time from 40 to 12 minutes by caching dependencies” rather than “strong engineer”. Rate against the anchors, not against the other candidates, and do not discuss candidates with the other interviewers until your scorecard is submitted.

Limitations

• The order is arithmetic on the panel’s own ratings. It is not a prediction and not a decision; the people on the panel decide, after discussing the evidence.
• Kits written for this tool are starting points: adapt the competencies and questions to the job, and keep them the same for every candidate of that role.
• Data is kept in this browser and in the files you save. Clearing the browser’s site data removes it, so save a file for anything you need to keep.
• The PDFs draw Latin-script text; names in other scripts show as “?” in the PDF (the workbook keeps them).
• Check your own legal obligations for recruitment records, such as how long you keep interview notes and who may see them.

Privacy

Everything happens in your browser. Kits, candidates, scores and notes are kept in this browser’s local storage and in files you choose to save; MySmartCoPilot never receives them.