Imagine an accounts clerk who puts ‘advanced Excel’ on her CV because she did a course with that word in the title. Her new boss finds out in week two that she can’t build a pivot table. Nobody really lied. The difference between what people think they can do and what they can do under pressure is common, and a technical skills assessment that NZ employers can organise before the offer stage is a practical way to find out early.

Most of these tests don’t ask what a button is for. They put the candidate in a simulated version of the software and ask them to do something: format a report, fix a formula, and find an email from last Tuesday. You can’t bluff your way out of that. You know where things are, or you don’t.

The first thing you have to get right is levels. Microsoft Office exam versions are usually offered in beginner, intermediate, and advanced versions. Choosing the wrong version will distort the whole result. Give a front-desk applicant an advanced Excel test, and they’ll look hopeless, which says more about your test than it does about them. Think about what the person will actually do in their first month, and test there.

Timing is the next decision. A timed test shows how fast someone works, which is important if you are entering data or doing invoicing. A non-timed one is nicer for those who are accurate but careful, or who have been out of work for a while. Neither is superior. It’s a matter of whether the job rewards speed or rewards getting it right.

Ask for a sample report before you purchase anything. The good ones divide a score into areas, say formulas and formatting, so you can see that someone is strong on one and shaky on another. One out of 100 doesn’t tell you much.

Small businesses sometimes assume this is something only big corporates with HR departments bother with. It’s closer to the opposite. A ten-person firm that hires a weak administrator feels it immediately, with no one else to pick up the slack. A technical skills assessment that NZ providers deliver online can matter more there than anywhere. The candidate receives a link, spends about half an hour on it, and then receives a report.

Plenty of employers get timing wrong. They test the one person they’ve already made up their minds about, as a formality, and then feel awkward when the result is bad. Test at shortlist stage, before anyone has fallen in love with the candidate. It’s a lot easier to be objective when you haven’t already pictured them at the desk.

The other trap is pass marks. If you set the bar higher than the job requires, you will scare away people who would have been perfectly good at it. Look at the real job. Can anybody in that role ever write a nested formula? If not, don’t test for it.

A result just below the benchmark does not automatically mean no. If someone is quick to learn and honest about gaps, they can be trained, and the report tells you exactly which features to cover. Someone who needs the skill on day one and scores far below it is another matter, and it is far better to know that before the contract is signed.

Let candidates know what they can expect. A quick email explaining how long the test takes, that it needs a quiet spot and a decent connection, and why you’re using it, costs nothing, and makes people calmer. Keep the results to those making the hiring decision, as they are personal information.

We also tend to overlook basic computer literacy. Hybrid working has shown how much these skills can vary, from saving files in sensible places to recognising suspicious links. A short basic computing test can reveal more about a candidate’s suitability for an entry-level role than a lengthy technical exam. A technical skills assessment in NZ should be reliable, relevant to the job, fair to candidates, and provide results that a busy manager can understand in five minutes.

Specialist software is a different thing. Generic office tests are not going to tell you if someone can use your accounting package or your specific CRM. Many providers cover a wide range of applications and will discuss others if you ask, so describe what the role really uses day to day. It’s a short email, and it saves you buying a test that measures the incorrect thing.

One thing to watch out for is the temptation to make the result a score race. Two candidates separated by a few points are not meaningfully different, and treating them as if they were gives a false sense of precision. Put people into the bucket of clearly meeting the requirement, those close, and those well short. Let the interview decide among the first bucket.

Conclusion

Testing skills is a small, unglamorous step, but it saves a lot of painful conversations in month two. Choose a level that reflects the real work. Test early and read the results and everything else you know about the person. Then move on to the harder side of hiring – whether you’ll enjoy working with them.