Online Assessment Software for Pre-Employment Testing

What does online assessment software do at the screening stage?

It replaces a judgement call with a measurement, early, before anybody has spent an hour on a call. In Pitch N Hire, assessments live under Hiring Operations. You create one, choose the multiple-choice type, give it a title and define how many questions it contains. A test carries three settings that appear in the assessments list: a time limit, a passing score expressed as the percentage a candidate must reach, and the number of questions. Answers are scored automatically against that pass mark, and a submissions count shows how many people have completed it. Open the assessment and you can review individual submissions and their scores. That is the whole loop, and its value is entirely in the word automatically. A test that a human has to mark is a test a busy team stops using in the third week. Automatic scoring keeps the step alive under volume.

Why does unstructured phone screening waste the most recruiter hours?

Because it spends the most expensive resource on the least discriminating question. A fifteen-minute call to establish whether someone has actually worked with a named technology costs a recruiter a quarter of an hour plus the scheduling around it, and it produces an answer the candidate could have given in four minutes unattended. Multiply that by forty applicants and a week disappears into confirming basics. The calls also drift, because two recruiters ask different questions and neither writes down a comparable answer, so the shortlist reflects who interviewed rather than who is strongest. A short, scored test moves that question earlier and makes it identical for everyone. The recruiter's time then goes where judgement is genuinely required, which is the conversation about motivation, context and constraints that no test can hold. That conversation is worth protecting, and it is the one that decides offers.

How do you build a test that is about the job rather than about tests?

Start from the work, not from a question bank. Write down the three or four things somebody must already know on day one, then build questions that check exactly those. In Pitch N Hire you can add questions by hand or generate them with AI by choosing a topic and a difficulty level, then regenerate or remove any that miss what you are hiring for. That last step is the one to take seriously. Generated questions are a draft, and reviewing them is where the test becomes yours. Cut anything that tests recall of a fact a competent person would look up. Cut anything a candidate could answer correctly by recognising the phrasing rather than knowing the answer. Save as a draft while you iterate and publish when it is ready, because only a published assessment can be sent. A draft cannot reach a candidate, by design.

How should you set the time limit and the pass mark?

Set the time limit by asking how long the test takes someone who knows the material, then adding room to read carefully. Documented options run from around ten minutes to an hour, and the shorter end is usually right for a screening step, because a working candidate can finish it in a break rather than scheduling their evening around your process. The pass mark is harder and deserves an honest answer: it is a decision you have to be able to justify, not a number to borrow. Take the test yourself with two people who already do the job well. If they do not comfortably clear your threshold, the threshold or the questions are wrong. Write down why you chose the figure, because the first time a strong candidate falls a single point below it, somebody will ask. Answering that question is the point of writing it down.

How do you send tests without losing half the pipeline?

Send from the pipeline, explain the reason, and keep the ask proportionate. Tests are distributed from a job's pipeline by email, Microsoft Teams or WhatsApp, and the assessment must be published rather than sitting in draft. Which channel works depends entirely on who you are hiring, and for many roles in India a message reaches somebody faster than an inbox does. What lifts completion is not the channel though. It is context. Tell the candidate how long it takes, what it covers and what happens after they pass, in the same message as the link. A test that arrives with no explanation reads as a hurdle, and strong candidates who have other options simply do not start it. Keep the message in a template so every candidate gets the same framing. Consistency in that message is also a fairness question.

What can a multiple-choice test tell you, and what can it not?

It can tell you reliably whether somebody holds a body of knowledge, and it can do that for two hundred applicants without a recruiter reading anything. That is a genuine and underrated result. What it cannot tell you is how the person works: whether they can structure an unfamiliar problem, communicate a decision, or produce something usable under real constraints. Treat the score as a gate, not a ranking. Two candidates on the same score are not equivalent, and the person one point below your threshold is not measurably worse than the person one point above. Unsupervised tests also carry an obvious integrity question, so decide in advance how much weight a remote score can bear before a human has seen the work. Decide that before the first test goes out. When you need capability rather than knowledge, that is candidate assessment software territory.

How do you evaluate assessment tooling in a demo?

Build something in the session instead of watching a prepared test. Ask the vendor to create an assessment from scratch, add a few questions manually, generate a few more on a topic you name, then delete the ones that miss. Publish it, send it to your own address from a pipeline and take it on a phone, because that is what most candidates will do. Then look at the results view: individual submissions, scores against the pass mark, and how many people completed. Ask what happens to a partial attempt and whether scores can leave the system for your own analysis. Finally ask about supervision honestly, since remote testing and proctoring are separate capabilities and you should confirm what is included rather than assume. Then compare the outcome with the rest of your applicant tracking system requirements and pricing. Do that before the contract, not after.

Test settings and what each one decides