March 1, 2022

Candidates Don’t Like Bullsh*t Homework


Episode Highlights

Subscribe to the Talent Insights podcast on Apple Podcasts, Google Podcasts, (recommended for Android users), Amazon Music, or Spotify. Watch us on YouTube—and don’t forget to rate us!

Everyone says pay + remote are the two must-not-screw-up-if-you-want-to-hire issues in this market.

But there’s a 3rd: bullsh*t homework.

We’re looking at you, ‘coding test before talking to anyone.’ You too ‘build me a product roadmap on a whim.’

Yes, assessing skills is crucial. But there’s a time and a place for these types of exercises, which is usually later in the process. Or (gasp) never.

Jeff Smith and James Hornick will put the busy work on blast and pitch a few superior alternatives in The 10 Minute Talent Rant, “Candidates Don’t Like Bullsh*t Homework”

Episode Transcript

The 10 Minute Talent Rant is live. I’m James Hornick joined by Jeff Smith and we are on the clock. The 10 Minute Talent Rant is our ongoing series where we break down things that are broken in the talent acquisition and hiring space, maybe even pitch a solution or two. Before we dig in, all our content could be found at

This week’s topic: candidates don’t like bullshit homework. Yes! I love it. So this all started, we had an idea, really this idea started with coding tests. That’s a problem which we’ll get into in a second. We realized it’s a bigger topic than that because this goes far beyond just kind of technology and software development hiring.

We also want to point out that not necessarily all homework is bullshit. Or maybe I should say not necessarily all assessments are bullshit. Maybe all homework is bullshit still, I don’t know. There are valid and legitimate assessments that need to be done. They just need to be better conceptualized, whatever.

I guess first off Jeff, let me ask you, why is bullshit homework a problem? I mean it’s- everyone hates busy work. Everyone. Yeah. Especially when you have absolutely no idea if there’s going to be any sort of incentive at the end of the tunnel. Look, anyone good is basically, there’s going to be a de-centivization like on the priority list of your role

if they’re looking at other roles, if you’re going to be incorporating homework early on. So job seekers just inherently have a low threshold by in large for doing any sort of this work in any market, let alone this market. Secondarily, asking people to do real work that’s going to benefit your company, then not hire them and not pay them is straight up unethical.

Full stop. So if you’re going to do stuff like this, there is a world where you just basically have to pay somebody to do it. Yeah, no I agree. Yeah. I mean, it might not even be on -it probably is unintentional if you’re looking to just kind of pick their brain. But doing work for you in real time for your company is you have to pay them.

So anyway, in your mind, what is illegitimate? i.e bullshit and what’s legitimate? Let’s go back to where this whole thing started because I was posting about this on LinkedIn. Coding tests before talking to anyone, they’re just not going to do it. It’s just not even feasible. I mean, I talked to a client last week who they had a 50, literally a 50% fall off rate of people who apply for the jobs because their step one was doing a coding assessment.

It’s just- weird. I wonder why. Yeah, they’re just not going to do it. And I think that there’s also other issues with like coding tests in general too. There’s a lot of kind of corporate institutional bias with it, right? People who favor coding one way in a certain style versus another. There’s a lot of places I’ve seen that they say they’re open to any language, they’re open to polyglots language of choice.

Yet, they only give the test in C-sharp and don’t give it in Python when the person’s like a Python coder. So there’s just tons of institutional bias and stuff there too. When you’re talking like take-home tests that are kind of phrased in certain specific ways like that. There’s one. Yeah, absolutely. Stuff that I’ve seen is

like, “Hey, go create a custom product or go to market roadmap at any phase”. Again, going back to my original point, all of this stuff without pay. First of all, everybody by in large in the passive market has a job. You’re literally asking them to do a second job. We had somebody internally bring up that

they were asked to analyze actual data within the company’s ATS and it took him six hours to put a plan together to fix the problem. And they never got back to him. Yeah. So ghosted. So six hours spent on a project using custom data to come up with a custom solution then they’re never even got back to them.

And then what I think is the soup du juor, like asking for literally anything at all, for anyone entry level or a low-paying job, stop it. The point of an entry level job is just giving someone a shot and yes. Exactly. What’s legit? Give us some- there’s a couple of things we said that are legit. What do you got?

Yeah. So real-time pair programming, like set some guidelines. Put the lanes in order, make it fun, make it interactive. Give the interviewer an opportunity and a space to ask questions while real time interacting with the candidate with your code base. It’s okay. Everyone’s code base has warts, like deal with it.

You can show your code base without scaring somebody away. See how they’re going to react with things you encounter day to day in the moment. Makes it fun, it also gives like tons of icebreakers, allows for rapport, all that good stuff. Similarly, there’s other fields where things like that kind of exist.

And there’s a difference between- one of our colleagues on the finance accounting side was bringing this up. So certain complex areas, asking an accountant to do homework is silly because they’re typically reporting on past events. So if they can just show the other things they’ve done in the past previously, it serves the purpose. But that’s different from things that are more future-focused.

So like in the investment banking industry, if you’re like forecasting out and synthesizing and creating something new, modeling techniques, you have to be able to prove you’re able to do it based upon like new parameters or things like that. So there are times where, there it is kind of a subtle difference when you’re looking at past versus future events and things are more customized, but anyway. I think any real time exercise

in any discipline can work so long as it’s well thought out. Yeah. And I like the fact that, especially with talking real-time exercises, you as an interviewer are putting some skin in the game, you know what I mean? And actually getting a sense for how people work. So anyways.

What do we do instead? Because we had a bunch of takeaways for this one. So like, instead of just like dishing out homeworks and assessments, what should we be actually doing? Lead us off on this. Yeah, never, ever, ever, ever, ever, ever- did I say ever? Don’t start your interview process with any homework or a project or anything that even resembles that. Talk to candidates first and assess the fit as a human being. Honestly, enough with the gate keeping bullshit.

Nobody’s going to do it. You referenced it before, the drop-off rate is insanely high for processes that start with projects. Make the homework- number two, make the homework mutually beneficial. So you’re going to get some insights as like your company entity about the candidates work product, but the assignment should inform that candidate about what their day to day work life will look like, hopefully in a positive way.

Again, the whole idea of just being thoughtful about the exercise that you’re putting together for these individuals. All right number three, we’re still going through this- recognize situations where past work like a portfolio actually accomplishes the goal. If they’re allowed to share it, asking them to kind of go through a previous product plan or a roadmap or something like that, but describing why they kind of made the certain decisions they did, more than serves the purpose than just kind of creating something brand new and hypothetical that doesn’t actually like- find out what they actually did in the real world, why they did it, what decisions they made. Right. Number four, recognize when an actual conversation can flush this out.

I mean we talked to, we know a lot of tech leaders, a lot of them were able to very accurately assess someone’s skill level just by talking to them like a human being. And again, if needed doing some sort of paired exercise, which you can kind of mentioned before. But a lot of the things you’re looking for to see if someone- you can just find that out by talking to people if you know what you’re doing.

Yeah. We’ve mentioned it a couple of times. Number five, if you’re counting, in situations where you actually do need to assess forward thinking capability for whatever reason, do it thoughtfully but consider paying them. There are companies that are doing it now. It’s still a vast minority, but- we do it. We’ve done that for people.

Like when we’re getting outside of our lane and we want to know what they’re doing, we’ve actually made it a paid projects. It works and weird, you get a much better product and an action- somebody is going to actually take a full swing at it if they’re getting paid for it.

Yes. Last one. And this is the one like, okay, but we’re flooded. How do we sort this out? Our tech leaders are busy, everyone else, how do I? Get a recruiter who actually knows what they’re doing. That’s the reason why- if you have whether it’s tech, whether it’s any other area, if you actually have someone who’s a specialist recruiter, they don’t need to fully vet them to the point where you need to be able to hire them.

They just need to be able to vet them to the point where they know if you should talk to them or not. That’s the entire point of the function. And if you have a recruiter who can’t do that and everything’s still getting through the cracks, you don’t actually have a recruiter- or a good one. Throwing that out there.

So we are short on clock. That’s a wrap for this week. Thanks again for everyone for tuning into the 10 Minute Talent Rant, part of the Talent Insights series, which is always available for replay on as well as YouTube, Apple podcast, Google podcast, Spotify and Amazon.

Jeff thanks again, as always. Everyone out there, see ya soon.

Episode 90
The 10 Minute Talent Rant, Episode 90 “Gen Z’s 12% Unemployment, Explained” It came as a shock to pretty much everyone that the...

Episode 89
“Recruiting Is Not The Goal. Hiring Is.” – The 10 Minute Talent Rant is LIVE What do you call recruiting that doesn’t result...

Our Shows

Our Latest Blog

4 or 6 day work weeks. Who you got?

2 opposing ideas can be simultaneously true. The 4 day work week and the 6 day work week. Both ideas are great and terrible at the same time. The context: 👉In the UK, the South Cambridgeshire council ran a 4-day work week trial for their ...