I tried Google's new job interview practice tool and I want to cry

Google wants to help you succeed in interviews. Its new tool, though, makes me wonder how interviews are conducted these days.
Written by Chris Matyszczyk, Contributing Writer
Two mature business people congratulate a young professional.

I've come with my lawyer, just in case.

Morsa Images/DigitalVision/Getty Images

I'll admit I'm out of practice.

The last time I was being interviewed for a job was certainly when America was still sane.

Yet I never really prepared for particular questions to be asked. I merely feared that the first question would be: "Tell me a little about yourself."

Because that told me the interviewer likely hadn't read my resumé or, if they had, could not remember anything about it. And didn't care.

Google's stretching exercise

So when I saw that Google had released a new online tool to help you practice for job interviews, my mind was intrigued, and my soul prepared to be saddened.

Google's Interview Warmup promises it'll help you practice key questions, get insights about your answers, and get more comfortable interviewing.

Naturally, this is all achieved via glorious AI, which only wishes it could stare intently into your eyes. That's because this tool doesn't switch your camera on but merely activates your microphone.

At least, that's what I thought.

What I subsequently discovered is that you can actually type your answers rather than have Google listen to them and type them down for you.

What I also discovered is that Interview Warmup lets you practice for interviews in all kinds of areas -- everything from data analytics to e-commerce.

Why this? Please, why this?

For each job type, there are five questions.

For data analytics, the first question was: "Can you please tell me a bit about yourself?"

Ah. Oh.

Has nothing really changed? Interviewers are still as lazy as they always were? And is there any good way to answer this?

"Hullo. I've recently been released from jail, where I spent five years for persistently defaming tech leaders."


"Just seeing if you were paying attention."

No, I wouldn't be a great interviewee. Still, this tool -- because every technology must categorize -- offers insights into three different areas: job-related terms, most-used words and talking points.

Oddly, my answer above gave Google's AI not a single shudder. In fact, the tool told me it had detected no talking points in it.

It also told me I'd used no job-related terms. It appears neither "tech" nor "leaders" are in any way related to a job in data analysis.

I'll leave you to analyze that while I tell you that I also practiced for an interview in e-commerce.

Question: "What do you do when you need to learn something new?"

I knew that "I weep" wouldn't be a good answer, so I tried to pander to Google's AI.

I answered: "It depends what it is. If I need to learn a new dish, I Google a recipe. If I need to learn to swim, I Google 'swimming instructor'. And if I need to learn Serbo-Croat, I get hold of those wonderful new Google translation glasses."

Surely this answer would inspire further conversation. The AI said it wouldn't.

The more I tried, the more I failed to be remotely interesting. To a robot, that is.

Creativity isn't welcome here. Or is it?

I couldn't believe it when I got this question while preparing for an e-commerce interview: "Explain the benefits of e-commerce platforms to someone who is completely unfamiliar with e-commerce."

This had to be a question seeking a creative answer.

So I attempted this: "Hullo. How are things in outer space? Down here, we have this thing called e-commerce. It works on something called the internet. Why am I telling you this? You're from outer space. You already know this. Anyway, we're hoping to send our best e-commerce expert, Jeff Bezos, up to you so that he can show off the finer points and we can be free of him here."

This, apparently, offered no discernible talking points. I was ready to break down.

Until I finally made a breakthrough. One question in my data analytics interview was: "Tell me about a time you had to deliver on multiple competing priorities. What did you do, and what were the results?"

I answered: "There is no such thing as multiple competing priorities. So I pick the one that's going to benefit me the most personally first."

I was hoping this wording might confuse the AI. Instead, it commended me for, oh, giving an example.


Finally, a breakthrough.

Screenshot by Chris Matyszczyk/ZDNet

The AI does try to be helpful. It gives you a list of keywords you might choose to drop in an interview. There are a lot of them.

Then again, it highlights the words that you use more than three times and tells you this is "not necessarily a bad thing."

It may not necessarily be a good thing either, so please, oh robot, can you be a touch more helpful?

Having waded through quite a few questions in different job areas, I wondered just one thing: "Is this what it's like to be interviewed at Google?"

Indeed, this tool was specifically created to help people get Google Career Certificates, by means of which UK citizens could enhance their digital abilities.

It may well be that those who are especially nervous in interviews may enjoy being able to practice by talking to a robot. It may be more incisive than practicing in a mirror or with a friend who's holding a glass of wine.

But this all feels a little like a job interview as a test. Google has certainly enhanced that impression in the past. We're so clever, so we're going to test how clever you are.

I still can't help wondering whether the best job interviews are those that are unstructured when two people begin to talk, find each other interesting and discover the experience extremely pleasurable.

I'm an idealist, I know.

Editorial standards