Prototyping a new hiring process

Weekly Ship #2 | Hacking a better process with Google Sheets, bots, and IFTT

We are prototyping a tool for teams to make the hiring process more human. We value transparency, so we share our progress here each week. If you like what you read, please subscribe, and if you like it, don’t forget to hit 👏 .

Last year at For the People, we realised we weren’t hiring and interviewing well. We knew there was room for improvement. We wanted to test structured interviews, consistent rubrics, and team-based interviewing. But…without involving too much work on our part (#realtalk). Essentially, we needed to prototype a lean new hiring process. How do you prototype a new process? The purpose of a prototype is to test some assumptions.

Our problems and assumptions

Our problems and assumptions.

Nerding out & learning from the pros

We geeked out on academic articles, research, and other companies who’ve researched their hiring process.

We’re all for learning from the best. We had a look around at others who’ve done hiring well. We’ve looked at companies like Google, Nordstrom, and Spotify, and I’ve had the (admittedly nerdy) pleasure of picking the brain of an organisational psychologist and reading a slew of academic papers on the best predictors of performance in employees (if you’re interested in reading these, get in touch). Google has been incredibly transparent with what they’ve learned about hiring, and Laszlo Bock, Google’s former head of people operations has published a book called Work Rules! That Will Transform How you Live and Lead, which I highly recommend reading if you want to dig into culture, hiring and performance. His key takeaway?

“Most people simply aren’t very good at interviewing. We think we are hiring the best because, after all, aren’t we great judges of character?…We all think we are great at it, but we never go back to check if we are, and so we never get better.” — Lazslo Bock, Work Rules!

Here are some of the highlights that we were thinking about as we prototyped our process.

Work Rules! is a great book on culture, hiring and performance by Google’s former head of people operations.

• Collecting, using, and sharing data on what works improves your hiring machine’s performance “Just as our products can always get better, so can our hiring machine. We constantly review and work to balance our speed, error rate, and quality of experience for candidates and Googlers.” — Work Rules!
• The best predictor of how someone will perform in a job is a work sample test, closely followed by general cognitive ability tests, and structured interviews. Unstructured interviews (i.e. a casual chat) are actually pretty bad at predicting how someone would perform once hired (they can explain only 14% of an employees performance). They’re more meaningful than reference checks (explaining 7% of performance) and years of experience (3%), but work samples take the cake in actually predicting success. Second best to work sample are tests of general cognitive ability (26%), like an IQ test. In third place, however, are structured interviews — where candidates are asked a consistent set of questions with clear criteria to assess the quality of responses. More on this at another time.
• Be on the hunt. “The top performers in most industries aren’t actually looking for work, precisely because they are top performers who are enjoying their success right where they are. So your odds of hiring a great person based on inbound applications are low.” — Work Rules!

I like big research papers and I cannot lie.

• Make recruiting part of everyone’s job. “The first step to building a recruiting machine is to turn every employee into a recruiter by soliciting referrals.” — Work Rules!
• Use a concise hiring rubric. “Think about the last five people you interviewed for a similar job. Did you give them similar questions or did each person get different questions? Did you cover everything you needed to with each of them, or did you run out of time? A concise hiring rubric addresses all these issue because it distills messy, vague, and complicated work situations down to measurable, comparable results.” — Work Rules!

Prototyping our Process

Even though we’re not Google, a lot of Lazslo’s book resonated with us.Even without a people operations team (or a human resources person!) there was a lot of room for improvement in our process.

Here’s what we did to start prototyping a smarter, more objective process:

We clarified our hiring criteria and interview questions based on our company values (see below). Too often, hiring leans on our own bias — do we like the person we’re interviewing? Are they like us? Not a good way to hire. So, we got clear on our criteria — we looked at For the People’s values, which were defined by the team in our first annual retreat. Our values provided a good rule of thumb for assessing potential team members. We also used it to generate questions that would be helpful to ask in an interview.

We created an internal tool to help ourFor the People interviewers think objectively about what we were hiring for and how we would assess that in an interview question.

We set up a simple form to make it easy for our team to add potential candidates to the talent pipeline. We wanted to get in the habit of being on the ‘hunt’ for people, so we created a simple ‘add candidate to pipeline’ Typeform and added it to our talent resources pinned post.

Our interview feedback form: a quick and dirty Typeform.

We automated parts of our recruiting and interviewing process. Automation is a wonderful thing. Playing around with Zapier, Google Sheets, and IFTT, we set up a very simple bot to report to the talent channel when a new candidate had been added to our pipeline.

We leaned on automation to help reduce the ‘death by admin’ problem that we faced, as well as keeping our process transparent for the team. Once a teammate filled out a feedback form, a bot notified our talent channel that a candidate had been interviewed.

We codified our process and made it part of our weekly rhythm. This meant outlining talent resources, roles and responsibilities for interviewers and team members in a single source of truth. We introduced this in an all hands conversation on a Friday afternoon, and a pinned Slack post in our talent channel containing our hiring criteria and feedback form links. The whole team was onboard and wanted to create a great candidate experience for everyone who walked through our doors. As part of this, we agreed upon a few rules for ourselves:

• Champions for all. We wanted to make sure every candidate was ‘championed’ throughout the process of interviewing, so we assigned a champion when a candidate entered our pipeline.
• 30 minute limit. No more 1 hour interviews. We want to be respectful of people’s time — both candidates and our interviewers. If you’re efficient, prepared, and focused, you can find out a lot in 30 minutes.
• 2 interviews, max. If it the answer isn’t a ‘hell yes’ by then, it’s a ‘no, thank you!’ (Credit to David Barton for this rewording — we ❤ it!).

In prototyping our process, we identified a gap in tools for team-based hiring.

In short, our goal, to use Google’s words, is to “predict how candidates will perform once they join the team.” At For the People, we’re a small studio with no HR person, so, trust us, we know how hard it can be to get the hiring and interviewing process right — even when you’ve got a top-notch, tech-savvy, forward thinking team of people on the case. As you’ve seen above, we’ve been in the trenches using Trello, Typeforms, Slack and Excel spreadsheets, getting lost in open tabs (we’ll talk about our teams feedback on the process in our next post — but in short — we have got a lot of bitty little systems going on at the moment, and it’s rather overwhelming!). We’re now looking at how we can streamline and enhance this process through software.

What does your hiring process look like? We’re talking to small businesses without dedicated HR managers right now. We’d L-O-V-E to ask you a few questions about your process here. Or, check out Weekly Ship #3, here.

Most products launch when once they’re completely polished and perfect. We’re committing to share our work on this bot as we go. We believe more feedback makes everything we do better. We’d love to hear yours. If you’d like to be notified when we launch our beta product, sign up here.