Interview Management

The feedback module exists to solve a familiar enterprise hiring problem: interview evidence is often scattered across chat, email, spreadsheets, and memory. That makes final decisions harder to justify, creates disagreement with no audit trail, and weakens recruiter visibility into what each interviewer actually assessed.

Ovii moves that evidence into the candidate record itself. Instead of relying on unstructured opinion, the system captures role-specific section ratings, overall recommendation, narrative feedback, and restricted recruiter notes inside the same workflow where candidates are staged, assigned, and reviewed.

The feedback module is intentionally permission-aware. Ovii does not treat every user the same because interviewer evidence, private notes, and recommendation overrides have different governance needs.

In practice, recruiters, HR users, and assigned interviewers all interact with the same feedback record differently. That separation is what keeps the workflow collaborative without making sensitive commentary too broad.

Note

Private notes are filtered per feedback entry. Even if a user can open the feedback list, they do not automatically see every private note unless their role or explicit viewer access allows it.

Every feedback submission is shaped by the job’s active evaluation template. Ovii first looks for a job-specific feedback form and only falls back to the broader category template when a custom job form does not exist. That lets the team tailor evaluation criteria for one role without forcing the same rubric onto every job in the category.

The default form is intentionally generic and safe. It gives teams a usable starting point, but it will not always reflect what matters for a specialized role. For example, a platform engineering role might need system tradeoff analysis and scalability judgment, while a customer-success role may need stakeholder communication and escalation handling instead.

Note

The backend excludes rejection records when deciding whether the template should lock for interview-form editing. The lock is meant to protect live interview evaluation, not incidental rejection comments.

The top of the form captures the summary decision for the round: overall score, recommendation, and overall feedback. This is the layer recruiters and panel reviewers read first, so it should explain the hiring signal clearly even before they expand section-level details.

In the current implementation, the overall score is a 0-to-5 star rating, not a 10-point scale. The recommendation is chosen from Strong Hire, Hire, Maybe, or No Hire, and the overall feedback field is required because the system expects a written decision rationale, not just a score.

Note

The form autosaves a local browser draft while the interviewer types. That helps prevent lost work, but the official record is only created after the user submits successfully.

Below the summary block, Ovii expects section-level evidence. Each section represents a competency area the hiring team has decided matters for this role. The interviewer can expand the evaluation criteria, give a star rating for that section, mark it not applicable, and add comments that explain what they observed.

This is where the module becomes more useful than generic notes. Section-level scoring makes it possible to compare interviewers on the same dimensions, see whether the candidate is strong in some areas but weak in others, and avoid reducing the entire interview to one vague paragraph.

Private notes are not the same as overall feedback. They exist for information that should stay on the recruiter side of the process, such as compensation concerns, internal coordination risks, escalation context, or sensitive interview observations that are not meant to be broadly visible.

Privileged users can choose the visibility scope for private notes. Non-privileged interviewers do not get to widen that audience; their notes are restricted to HR-only handling by design.

Feedback in Ovii is stage-aware. An interviewer is expected to submit against the stage they were assigned to, even if the candidate has already moved forward. That is important because enterprise panels do not always submit at the same pace, and the system needs to preserve which round the evidence belongs to.

The recruiter-facing feedback surface also supports stage coordination. Privileged users can change the round, assign interviewers by email, and keep assignments aligned with the actual round being run. Automated stages such as Assessment and Async Video are treated differently because those stages do not use manual interviewer feedback.

Note

When feedback is submitted for a past assigned stage, Ovii records the evidence but does not treat it as a new stage transition. That prevents late panel input from silently rewriting current pipeline state.

Submitting feedback creates the official interview record for that interviewer and round. It becomes available in the candidate’s feedback history, where reviewers can expand the entry to inspect overall rating, recommendation, section-by-section comments, and private notes if their visibility level allows it.

What does not happen automatically is just as important: the candidate is not auto-promoted merely because someone submitted a recommendation. Ovii keeps stage progression as a deliberate recruiter decision, which is safer when multiple panelists submit at different times or when recommendations conflict.

The feedback module works best when the team treats it as the source of truth for interview evidence rather than as a formality after the real decision has already been made elsewhere. The more disciplined the team is about using the structured form, the easier it becomes to compare candidates, justify decisions, and defend process quality later.

The biggest operational mistakes are usually procedural: interviewers are assigned informally, templates are changed too late, private notes are used for information that should have been part of the shared evaluation, or recommendations are mistaken for automatic workflow actions. The module is designed to prevent those errors, but only if the team uses it intentionally.