How to Hire Better
A couple weeks back, I wrote about how to negotiate a job offer, based on my years of hiring at Google. I thought I’d turn the tables and share my thoughts on hiring instead. I’m in the camp of people that think a founder’s number one job is to build an incredible team – but it’s far easier said than done.
Much has been written about the herculean task of getting a job offer from Google. Far less is understood about how difficult it is to actually hire somebody once you’re there. Back in 2006, I developed a presentation for my managers about how to do just that. While much of my advice was specific to the culture and quirks of Google, there’s useful ideas that I suspect could help hiring managers from all flavor of organizations.
Rather than tell you what type of person you should hire, I’ll focus on the tactics that can help you hire those people more effectively.
Tapping your own network should always be the first and best way to find talented team members. But that only takes you so far. If, like most hiring managers, you have a recruiter feeding you resumes of potential candidates, you probably want to vomit on 90 percent of them. But it’s important that you take time to tell your recruiter exactly why each resume is vomit-inspiring before you toss it in the shredder. In fact, you should markup each resume and go through them one by one with the recruiter, so they know exactly what you didn’t like. If your recruiter is good at his job, he’ll quickly learn what types of resumes don’t work for you – and your vomit ratio will decline fast.
When it comes to administering that initial phone screen, pretty much anyone can handle it, right? Wrong. The screener is often the first live person from your company that a candidate will meet – so think very carefully about the first impression you want to make. If you don’t have a screener who can serve as a strong proxy to you, both in terms of understanding what you’re screening for, and also representing your company to candidates, then you’re better off doing it yourself.
In 2002, a couple of years before I joined Google, I had a disastrously bad day of interviewing at Microsoft. After flying to Seattle, it took all of 30 minutes to know this wasn’t going to be a fit either way. But they used an interview technique that made it worse than it needed to be. After completing my first interview, the recruiter stalled and made small talk with me, enabling the first interviewer to provide feedback about me in real-time, so it could be reviewed by the second interviewer before I walked into his office. I suppose the theory is that they could build on the knowledge obtained by the first interviewer – and refine their questions based on prior feedback.
This is a really bad idea. If you’re going to commit the time to have several members of your team interview a candidate, at least let each interviewer express his or her unbiased and independent opinion about the candidate. It’s smart to plan in advance what each interviewer is testing for (at Google, we had four broad categories of assessment, and each interviewer was assigned to just one). But having feedback from earlier interviews shared down the line introduces a horrendous confirmation bias that can only lead to bad hiring decisions in both directions.
Each interviewer should independently report his feedback and whether he recommends a hire or no hire decision. Ideally, interviewers should contribute their feedback without visibility into feedback from others. Each interviewer should indicate level of conviction about the hire or no hire recommendation with a simple number scale. And while it’s unrealistic to always expect a unanimous recommendation for hire, any more than a single “no” vote is probably too much. Furthermore, all votes aren’t equal: a “no” vote from your rockstar team member shouldn’t be overruled.
By the time you get to checking references, you’re probably worn down from the process, and giddy that your colleagues actually recommended a candidate. Welcome to the hiring dead zone – where most mistakes are made! Most managers have already made up their mind to hire somebody before they check a single reference. If that’s the case, don’t waste your time. It’s popular sentiment that references provided by the candidate aren’t worth the time of day. Who can’t find two or three people that will say wonderful things about you? While there’s some truth to this, and backdoor references are generally more useful, I’ve found that you can, in fact, learn a lot from candidate-provided references, but you have to ask the right questions. But more importantly, you have to keep your mind open that this might not be the right candidate.
Basic questions like “what did you think of Beth’s performance?” aren’t going to provide any sort of enlightenment. Asking something more quantitative is marginally better – ie is Beth top 50% of people you’ve worked with? top 25%? top 5%? This is a chance for the person to decide how enthusiastic they really are, and more truth will surface.
It’s also useful to pose questions that will begin to reveal areas of strength or weakness for the candidate. Rather than just asking the age-old “strengths and weaknesses” question, which will result in the reference citing pseudo-weaknesses such as “too much of a perfectionist”, “ works too hard” etc, it’s better to pose a question that will inherently force the reference to reveal more. Simple example: “We’re considering Beth for two different roles. One is very internally focused and process oriented. The other is customer-facing and requires a lot of interpersonal skills. Which do you think fits her skills better?” Asking a simple question like that will reveal more about Beth’s skills, weaknesses, interests, than asking a candidate to list her strengths and weaknesses.
In the end, always think first where the loyalty of the reference lies. If his or her loyalty is unquestionably with the candidate, you have to work harder to tease out the truth. Backdoor references are generally better, because their loyalties are more ambiguous. Fortunately, the vast majority of people are loyal to the truth, but, as a hiring manager, you have ask yourself: can you handle the truth?
To make an offer or not
Making a hiring decision is perhaps the most important thing you do as a founder or CEO. Every viable candidate has a balance sheet of obvious strengths and potential weaknesses. As a hiring manager, you will never have perfect information – nor will you have the luxury of time in today’s job market (at least in tech!)
The only advice I offer is that it’s a really bad idea to allow your desperation to fill a role to affect your decision about a candidate. In my first year at Google, I made the mistake of trying to push a candidate through, based on how critical the role was to meeting some goals I had signed up for. I can still remember the email I received from Larry, telling me I was working against my own objective with this argument. Larry is a very smart guy, in case you didn’t know.
Making an offer
Once you’ve decided to make an offer, you need to have a plan to win the candidate. You should never make an offer without having as much information as possible about the candidate’s situation: current salary & other compensation, preferences and expectations with respect to compensation, status of other opportunities the candidate may be considering. If you don’t know these things, you’re flying blind.
When you make an offer, it’s best to have the senior-most person whom the candidate met make the offer – it shows you value the candidate and the role. While I’m an email-centric (anti-phone) person for the most part, job offers are the one time that I always make a phone call. Or better yet, meet them in person. I was invited to Google so they could make the offer to me, and it made a huge impression.
You should leave modest room in your offer for negotiation. Taking a hardline on compensation sends a bad signal to a candidate about your company, hinting at inflexibility and lack of appreciation for employees. If the candidate responds with multiple asks (salary, equity etc), make sure they’re all on the table before you respond. And don’t make a better offer until she’s made it clear that she’d accept it if you do.
Closing the deal
It’s reasonable to give the candidate a few days to consider a job offer, but rarely should you give them more. My experience suggests that candidates who ask for a week or more to consider a job offer are very likely to decline. They are simply waiting to see if something better comes along, and that’s not a position you want to be in. You should always politely tell the candidate that you have others in the pipeline, and that you can’t reasonably leave a job offer out there for more than a few days. Occasionally, there can be reasons to give somebody more time (eg they are checking into the hospital for surgery), but these should be few and far between. Needing more time to think about it or talk to somebody about it isn’t on that list.
You should always have a plan to win candidates after the offer is made – and as they say, hope is not a plan. Some companies will send a small gift of some sort, such as a t-shirt, that shows their thoughtfulness. You should proactively make yourself available if the candidate comes up with more questions. Too much communication about the great things happening at your company can feel gratuitous, so go easy on that. And lastly, remember that the job isn’t done until the candidate has actually started at your company. Never assume that the deal is done until she’s happily acclimated to your company and embedded in your team!
What are your favorite hiring tactics? Join the conversation on Twitter @DaveGirouard