Screen with a scored test, not a hunch

An assessment template holds a set of questions, a time allowance and a scoring rubric. Question types cover multiple choice, single choice, free text, numeric response and file upload for work-sample tasks. Multiple choice and numeric questions score automatically; free text and file uploads go to a human reviewer with the rubric on screen.

Templates are reusable and versioned. A firm typically builds a handful per discipline rather than one per requisition, and attaches the appropriate one when the job is created. Because a template is versioned, changing a question next quarter does not retrospectively alter how an attempt from this quarter was scored.

An assessment is sent from the application rather than from a separate tool, so the invitation, the attempt and the result all sit on the record the recruiter is already looking at. The candidate receives a link, completes the attempt in the browser, and the submission returns to the application.

Reviewers work through attempts from an attempts view with the candidate record beside the answers. Scoring is recorded per question with the rubric visible, and the total attaches to the application so it can be used as a filter on the shortlist. An unscored attempt shows as outstanding rather than sitting silently, which is usually where assessment processes fall over.

An assessment can be attached to the requisition so every applicant receives the same one automatically, or sent by hand at a particular stage where only a shortlist warrants it. High-volume desks tend to attach at the requisition and use the score as a first filter. Retained and executive desks tend to send by hand, later, and treat the result as one more piece of evidence in a long conversation. The mechanics are identical either way, and the choice is a judgement about the role rather than a setting we make for you.

Basic time-locking ships today: an attempt has a time allowance and it closes when the allowance runs out. That is the whole of the integrity story at present.

Camera proctoring and copy-paste blocking are Wave 2. Today, a candidate can open another tab, and nothing in the product will know. If your process depends on supervised conditions, either invigilate the assessment yourself or use a dedicated assessment platform for that stage. We would rather write that sentence than let a feature list imply a controlled environment we do not currently provide.

A score is one input. Treated as a gate that automatically rejects below a threshold, an assessment becomes an automated employment decision, with the notice, record-keeping and audit obligations that attracts in some jurisdictions. Treated as evidence a human weighs alongside the rest of the record, it is a straightforward improvement on guessing.

Surhires records which candidates were assessed, what they scored and who reviewed the attempt, which is the data an independent bias audit would need. Commissioning that audit, issuing candidate notice where it is required, and deciding whether a score may reject anybody are yours. The product supports the record; it does not certify the process.

Surhires is not an assessment platform, and the anti-position on the pricing page says so. Technical coding assessments, psychometrics and validated cognitive testing are specialist products with their own research behind them, and for a role where that depth matters you should use one.

What is built here suits the far commoner case: a short, role-specific screening set that separates the candidates worth a call from the ones who applied to everything. Where you need the specialist tool, the assessment integration pages carry the real status of each connector rather than a logo wall.