How to Improve Hiring at Startups

Posted on Mar 27, 2010 | 125 comments


This is part of my ongoing series with Startup Advice.

job interviewOK, hiring is a loooong topic and I couldn’t do it justice in single post.  But I thought it might be useful to do a headline view of the key components and then come back and do the individual topics over time if people seem interested.  I’ve been recruiting for 20 years so I’ve learned a few things along the way.  I think this can also be a great resource for others to chip in with other suggestions since I clearly don’t corner the market on recruiting advice.

BEFORE YOU START

1. Define criteria for judging – I find that many people I know go into interviews without thinking about what is important to them first.  In fact, many people just go into interviews and “wing it.”  If that’s you then vow never to do it again.  You’ll never have a great outcome without some planning.  I recommend that you design a standardized form broken down into high level criteria such as, “Intellect, Work Experience, Personality, Attributes Required for this Job, etc.” and then sub categories underneath.  Example: Work Experience can be broken down into: has managed a team, has led direct sales efforts, has worked for a startup before, etc.  And on this form score everybody 1-10.

Once the form is created make sure to check with anybody else who is responsible for hiring or managing the person going forward before you finalize the evaluation sheet.  Make sure to denote which sub categories are the most important by making them bold, italic, shaded in yellow, whatever.  But you need to know (and agree if multiple people involved) what your most important 4-5 criteria will be.  These will form the basis of your deep dive in the interview.

2.Create a process- Create a process that you’ll use to recuit people before you start.  You’ll want to include a plan such as which databases you’ll use (or which recruiters), make sure you have a timeline of when the deadline for submission is, how people apply, when you’ll review applications, when you’ll give notice to those not selected, when you’ll interview candidates and when you hope to select somebody.  When I run the process without a recruiter I set up a gmail account specifically for the job.  I set up an auto-responder that tells people what the process is when they email us (on some occasions I did this manually).  It set up separate folders for “rejected, in process, phone call, in-person interview, finalist.”

The reason a process is so important is because if you get too far ahead on one candidate it’s hard to slow them down while you wait for more to come through the process.  Also, if people start applying and you don’t get back to them for 3 weeks while you wait for resumes to come in that pisses people off.  Communication with people is key.  On the basis that you’re not going to hire 99.9% of the people you apply – how you handle this can matter to your brand.  I’m not a big process guy, but on recruiting I become a process machine!  You have to be.

3.Have a good pipeline of candidates – I find that too many people who recruit candidates only see a couple of candidates (sometimes only one!) when they want to hire somebody.  I don’t understand this.  Yes, I know you’re in a rush and wanted the person yesterday.  You’re dying without them.  But the person you hire is hopefully going to be an important contributor to the company so making sure you’ve seen multiple candidates for each role will help you benchmark whether you’re hiring the best that you can for that job.  The only time to sole source is if it’s somebody you know from a previous job or a very trusted referral.

4. Use referrals where possible – Obviously if you knew lots of people who wanted to have the job you’re recruiting for you would have just hired them and saved yourself all the hassle.  But I mention this for two reasons: 1) referrals are often the best employees.  It’s far better to hire somebody that a trusted person has told you that they worked with in the past and they can confirm that he/she was a star than it is to hire the perfect resume + interview.  I’d take the former any day of the week.  2) the other reason I mention it is that a large part of your process ought to be outbound emails/FB/LinkedIn/Twitter messages to friends asking for their help.  Dedicate enough time to this.  Also, I think most companies should offer bounties for people who refer other employees provided they join and stay for a minimum period of time.  I’d far rather give $3,000-5,000 to an employee for a well-known referral than to pay $10-20k+ for a recruiter to go through a process.  Obviously you need rules.  But don’t be a cheap bastard.  Pay your staff to bring the best people they know to the table.

5. Pre interview testing – There are some jobs where you can test people before having to spend time with them.  An obvious example is programming.  We always used to do this and there are some good online tools for doing programming tests.  If it’s a job that requires writing you could ask them to submit a sample in advance.  I’m sure it’s true of other professions as well.  One other thing that I do.  If I’m hiring in LA, for example, and the person doesn’t live here (but they seem very qualified) I call them to understand why they’re applying here.  Sometimes you’ll find out that they have a spouse who just got accepted to med school or have family members in your town.  I do often rule people out if they have no reason for being in my town (I’m hugely against relo’s for startups with the exception of Silicon Valley).  I’ll cover that another day.

THE INTERVIEW

1. Read their freakin’ resume before the meeting – How many times have you been in an interview situation where the person is reading your resume right in front of you and clearly hasn’t read it in advance?  It’s too common.  It’s totally disrespectful of the person and a waste of your valuable interview time.  I’d far rather you go in 5 minutes late to the meeting but have read their resume in advance.  Seriously, do yourself the favor of not being this person.

2. Don’t do a “world tour” of resume – Here is the biggest mistake interviewers make.  They allow the candidate to do the “world tour” of their resume for 2-30.  What a waste!  You’ll learn very little.  This is the practiced version of what this person wants to tell you.  You need to drive the interview process, not them.  I do this two ways.  Either I say, I want the resume history in under 5 minutes (I usually do this to be polite since I know they’re dying to tell me what they’ve practiced).  But I ask them to really honor the 5 minute rule.  Or I just say, “I don’t want to take this in order, do you mind if I just dig in on the key jobs that I find most relevant?”  Depends on my mood.

3. Do deep dive in key criteria areas – OK, you’ll never be as psycho as I am here but maybe you’ll feel comfortable going part way.  I like to do really deep dives around specific topics and on specific job aspects.  Example: if I’m interviewing a sales candidate I’ll ask them to name a sales campaign that they worked on.  I’ll say “name one that you remember really well.  Something you’re proud of.”

I then perform a proctology exam.  I’ll ask how they first heard about the lead.  How they first contacted the customer.  ”Did you send an email first or call?  Did you get introduced or cold call?  Who introduced you?  Who came to the first meeting?  Who were you competing against?  Why do you think you won the deal? What was your original price quote? What was your final price?  What was the name of your champion?  What was their title?”  And on, and on.  I tell them politely before that I might get a little detailed on this question. I apologize afterward for the uncomfortable exam.

The reason that the deep dive is so important is that you 1) really understand if they did what their resume says they did.  Nobody can fudge details that perfectly on a proctology exam.  and 2) it shows you much more about how they think about things.  You can ask them questions about how they approached certain situation like setbacks in the sales process or how they handled the price negotiation.  Why they lowered price by 20%.  Honestly, about 50-60% of interviewee’s do poorly on the deep dive.  It’s a great divider in my interview process.  It’s the most important part of my interview.

4. Ask  consistent questions – If you prepare in advance for an interview and know your criteria then asking consistent questions is easy.  Strange to me that so few people do this.  It’s the only real way to get apples-to-apples comparisons in how people think.  Again, sticking to my sales interview examples, I often have a sheet of questions such as, “do you prefer to call high and get passed down or stay below the radar and win pilot projects before you see more senior people?” and “How long does it take to know whether a sales employee is going to be successful or not?” and “What’s your best secret for getting past the assistant of senior executives?”  I usually have 10-12 questions like this.  It is the great equalizer to hear different perspectives on questions where there is no clear right or wrong answer.

5. Don’t do all the talking! Another really common problem in interviews is “the talker.”  Too nervous to sit there and politely interrogate the prospective employee they fall back on talking.  They find it easier to tell you about the company and their job.  I know it sounds crazy to many of you but I promise it’s not that uncommon.  Your job isn’t to socialize with the person, become their best friend or tell them your life story.  You can (and should) start the meeting with the appropriate amount of banter to build rapport.  You should leave time at the end for questions.  But the rest of the time it’s over to you to ask questions.  Not only is it the right thing to do in order to maximize your understanding of the person but it is what they expect and want also!  They didn’t come here to hear you speak.  They came to get the job.  So they’ll want as much time as possible to sell you on why they’re so great.  And they can’t do that if you don’t zip it.

6. Save time for questions – OK, so I just advised you not to do all the talking.  I usually say right before I start interviewing (e.g. after the banter) that I plan to ask question for X minutes and will save 10-15 minutes at the end so that they can ask me anything they want to.  If they don’t ask you anything don’t hire them.  I’m not saying that because they should have prepared the standard 3 questions that we were all taught after college.  I’m saying that because if they’re not curious enough to want to know anything about you, your company, the job, the role, the expectations or whatever, they’re clearly not worth hiring.  Ever.  I’m always shocked when people say, “I don’t really have any questions.  I read your website and talked to so-and-so.  I feel like I understand it pretty well.”  No.  Curiosity.  No. Job.

7. Score immediately when done – I actually try to score many meetings while it’s going.  Especially when I’m running a big process (e.g. seeing 6-8 candidates in one day).  In that situation there’s no time to write it up afterward.  If you have your scoring form you can simply leave it under their resume and take notes on it and circle the score as they speak.  They can’t see – don’t worry.  If the interviews are more spread out you can do it afterward.  This is really important.  An hour after an interview you’ve forgotten key facts.  And if you see 5 people through the process TRUST ME they’ll all start to blend together.  Get the salient facts, circle your scores.  You won’t regret it.

AFTERWARD

1. Multiple meetings with top 2 candidates – Don’t make a hiring decision from a single meeting.  When you have your final two candidates make sure you have 2-3 meetings with each.  I know it’s a big time commitment but if you’re a small company and this is an important hire then you owe it to yourself to be sure.  I never recommend “sole sourcing.”  If you only have 1 final candidate your in trouble.  What are you going to do if they drop out?  What if you start to have doubts? (you’ll just hire them anyways out of fear or fatigue) Keep a spare in the back pocket.  At least one of every 6-8 hires I end up with the backup candidate.

2. Thorough reference / background checks – Yes, you need to call their references.  Expect them to sing their praises.  If a person doesn’t list the most positive references to begin with then you know they’re not worth hiring.  But you have to find a way to call people not on their list.  You need to be careful and respectful of them because it’s possible that their boss doesn’t know they’re interviewing.  But you have to find a way to get some info.  Often you can find people if you learn to become a LinkedIn Ninja.  Also, I highly recommend spending a small amount of money for a background check.  You never know.  There are many cases where prominent startups have hired people with a record.  You may still hire them, but better that you’re aware of it going in.

3. Make sure good cultural fit – I said you needed multiple meetings.  I always try to make one of the last meetings social and over food.  Cultural fit matters a lot to me.  So I want to be sure that we really get along.  Obviously if you’re hiring 10 programmers on your team you can’t follow this entire process and do many meetings and lunches, but do as much as you can.  When we used to do a ton of recruiting at Accenture a lunch meeting with junior staff was always part of the process.  Not only do you learn about cultural fit but people open up a lot more when they’re in a social and non interview environment.  Funny, that.

4. Move fast – Just because you have multiple meetings doesn’t mean you should move slowly.  Changing jobs (if they already have one) is a very emotional decision.  If they’ve decided that they’re excited about a position with you, you need to capitalize on this positive momentum.  All too often candidates drop out of the process because it goes on too long.

5. Show the love – When you decide to hire somebody really show them the love.  You’re most vulnerable right after you make the offer and they accept.  Keep up the pressure and love.  Schedule a dinner.  Have everyone who spoke with them call and tell them how excited they are.  Start having calls to talk about their new role.  And, as I’ve said before, help shepherd them out of their former employer.

  • http://bothsidesofthetable.com msuster

    That's great. I've seen people try but so far no mass adoption. Good luck with your eventual US entry.

  • http://bothsidesofthetable.com msuster

    re: scenarios from a current project that the candidate needs to solve / answer – I do that all the time. It's important. Thanks for mentioning.

    re: letting people who drop out of the process through attrition go – I agree. Better that you know that they're not committed before they join.

  • http://bothsidesofthetable.com msuster

    Recommendations on LinkedIn are a total waste. I use LinkedIn to find out who else worked at a company so I can find people to call to reference check. Example: if you were the CEO (but not founder) of the company I might go to LinkedIn to find out who the previous CEO's were. Or if you were VP Marketing maybe I'll find some of your peers (rather than the ONE guy you gave me who is presumably a friend). You have to be careful about doing this because often people haven't told their employers that they're thinking about leaving. I never do this without the candidates consent (although I don't tell them whom I am going to call).

  • http://bothsidesofthetable.com msuster

    If I knew one I'd certainly use it. Anyone?

  • garydpdx

    Thanks, Mark! It's a topic near and dear to my heart, I will look forward to it.

    As someone who has lived and worked in eight different cities, this issue is always a concern for me and I try to mention in my cover letter that a) I have been to their city before on business or vacation, and am interested in living there, or b) I have lived in that city before and would welcome the chance to return. In my mind, skill and talent should match first and when (e.g.) a Silicon Valley fabless semi start-up states “local candidates only”, it may block out or discourage someone who matches what they need but is working in Florida (or awaiting their next opportunity from there).

    And I have been on the other side before, in the 90's, appealing to alumni of a three letter multinational who are brought to Chicago, pass the interview with flying colors, but turn down the offer because they're too tied to their upstate New York town (or spouse can't move). I haven't been involved with hiring in the 2000's but can imagine that such a candidate would introduce a telecommute wrinkle in that situation before talks end. Anyways, relocation might deserve at least one article, if not more! Thanks, Mark.

  • http://arnoldwaldstein.com awaldstein

    Item #3 “Do deep dive in key criteria areas” is a critical one.

    Separating what 'we' did from what an individual does is critical especially when your hire comes from a larger organization.

    Really useful post Mark.

  • http://bothsidesofthetable.com msuster

    Yeah, the “deep dive” is my patented style in interviews. I don't understand why more people don't do it. And “the talker” is much more common than people think. I think some interviewers don't even realize they're doing it

  • http://twitter.com/beth_tk Beth Holmes

    We've found it's very revealing to get the applicants to play with the product then give us feedback.

    We were hiring a knowledge engineer at True Knowledge so asked them to sign up for an account and give it a test drive. Amazingly many didn't bother at all, even those who looked really good on paper. Obviously that works best with consumer websites like ours, but I'm sure most startups could do something equivalent.

    Then in the interviews they all gave the standard line about how they wanted to join us because 'True Knowledge is so exciting'. So I'd ask them what features they'd most like to see added to the product and what direction was most exciting for it.

    Some really did get excited – “Ooh it should do this and this and this…” but most had no vision of anything. The most memorable massively qualified but total no-hoper was “Well I can't really imagine why people would use a question answering service at all…”

  • pinghelbrecht

    Loads of great information and there's a lot in here which I religiously practice myself. Thanks for the post and many comments.

    My (personally) most successful hiring practice has been a homework assignment. A 2nd/3rd round interview should *always* require the candidate to come in with a deliverable (e.g. an engineer with code, a lawyer with a contract redline, a sales person with a fake sales pitch, a product manager with the 12m roadmap etc). I typically structure it in a way that it would take 2hrs to prepare (=reasonable). Apart from being able to judge the quality of the person's work (read: 50%+ of my candidates bomb it and are out), it also shows their commitment to the job and work ethic (sense the difference between the person who doesn't want to do it vs. the person who spend 10+ hours on it?).

  • http://twitter.com/ravikannan Ravi Kannan

    toivotuo, Check out JobsbyRef.com – it addresses the initial process part as outlined by Mark. I am the co-founder of this portal. You can reach me on ravi at jobsbyref dot com.

  • http://twitter.com/ravikannan Ravi Kannan

    So true.

  • http://twitter.com/ravikannan Ravi Kannan

    Thanks Mark. Will like to ping you with details and get your thoughts. Can I?

  • pinghelbrecht

    Loads of great information and there's a lot in here which I religiously practice myself. Thanks for the post and many comments.

    My (personally) most successful hiring practice has been a homework assignment. A 2nd/3rd round interview should *always* require the candidate to come in with a deliverable (e.g. an engineer with code, a lawyer with a contract redline, a sales person with a fake sales pitch, a product manager with the 12m roadmap etc). I typically structure it in a way that it would take 2hrs to prepare (=reasonable). Apart from being able to judge the quality of the person's work (read: 50%+ of my candidates bomb it and are out), it also shows their commitment to the job and their work ethic (sense the difference between the person who doesn't want to do it vs. the person who spend 10+ hours on it?).

  • http://twitter.com/ravikannan Ravi Kannan

    toivotuo, Check out JobsbyRef.com – it addresses the initial process part as outlined by Mark. I am the co-founder of this portal. You can reach me on ravi at jobsbyref dot com.

  • http://twitter.com/ravikannan Ravi Kannan

    So true.

  • http://twitter.com/ravikannan Ravi Kannan

    Thanks Mark. Will like to ping you with details and get your thoughts. Can I?

  • philsugar

    The only thing I expect to get out of a reference is a referral to another reference…..second level references really show the true picture.

    Totally agree about the process (why piss people off) and the deep dive (amazing how many people will take credit for something but not be able to remember the gory details)

  • philsugar

    The only thing I expect to get out of a reference is a referral to another reference…..second level references really show the true picture.

    Totally agree about the process (why piss people off) and the deep dive (amazing how many people will take credit for something but not be able to remember the gory details)

  • Phillip Morelock

    I like the thought here, but the “big-company-trained-HR” meter in my head is going off. What if you meet the wife/husband and that ends up being the reason you don't hire the candidate? Is this legal? Is it appropriate? And, considering I always try to be really frank with people who make it far in the process but don't get the job — what do you tell them??

    This is very interesting and I admit this angle is something I never would have considered. Would appreciate guidance on the sticky legal/personal front.

  • Phillip Morelock

    I absolutely agree. Hiring is the beginning of a relationship, and you don't want to have other people negotiate and position on your behalf at the beginning of your relationship. Take the time to guide the candidate, make the offer yourself, and close the deal all the way to start day. It pays off.

    I've had more than one “turf” fight with recruiters / HR about this point, but it absolutely matters.

  • Phillip Morelock

    I like the thought here, but the “big-company-trained-HR” meter in my head is going off. What if you meet the wife/husband and that ends up being the reason you don't hire the candidate? Is this legal? Is it appropriate? And, considering I always try to be really frank with people who make it far in the process but don't get the job — what do you tell them??

    This is very interesting and I admit this angle is something I never would have considered. Would appreciate guidance on the sticky legal/personal front.

  • Phillip Morelock

    I absolutely agree. Hiring is the beginning of a relationship, and you don't want to have other people negotiate and position on your behalf at the beginning of your relationship. Take the time to guide the candidate, make the offer yourself, and close the deal all the way to start day. It pays off.

    I've had more than one “turf” fight with recruiters / HR about this point, but it absolutely matters.

  • William

    Excellent post. Two other things that are important for me:

    I will always pepper them with questions about topics they should know something about, but where it's impossible to know everything. I'll then dig into one of their responses, seeing how deep their knowledge goes. This is partly to judge smarts and experience, but mainly to tell what happens when they don't know something. If they can't comfortably say, “I don't know,” then I never hire them.

    The other is to, if possible, make them do some actual work. For developers, my last step is a 3-hour pair programming session working on some toy problem that I use on all the final candidates. This not only lets you eliminate the bluffers, but it lets you find the occasional rock star with poor self-promotion skills.

  • William

    Excellent post. Two other things that are important for me:

    I will always pepper them with questions about topics they should know something about, but where it's impossible to know everything. I'll then dig into one of their responses, seeing how deep their knowledge goes. This is partly to judge smarts and experience, but mainly to tell what happens when they don't know something. If they can't comfortably say, “I don't know,” then I never hire them.

    The other is to, if possible, make them do some actual work. For developers, my last step is a 3-hour pair programming session working on some toy problem that I use on all the final candidates. This not only lets you eliminate the bluffers, but it lets you find the occasional rock star with poor self-promotion skills.

  • William

    Excellent post. Two other things that are important for me:

    I will always pepper them with questions about topics they should know something about, but where it's impossible to know everything. I'll then dig into one of their responses, seeing how deep their knowledge goes. This is partly to judge smarts and experience, but mainly to tell what happens when they don't know something. If they can't comfortably say, “I don't know,” then I never hire them.

    The other is to, if possible, make them do some actual work. For developers, my last step is a 3-hour pair programming session working on some toy problem that I use on all the final candidates. This not only lets you eliminate the bluffers, but it lets you find the occasional rock star with poor self-promotion skills.