Hiring Part III: The candidate’s visit – job talk and interview

The candidate’s visit to your lab is the most important step in the hiring process. Yet, it’s all but easy to be the interviewer, and it can feel just as daunting to be on the hiring side as to be the applicant. In today’s post, the third of four on hiring, we’ll look at each part of the candidate visit: talk, interview, lab tour, and meeting the team.

hiring interview

This post is part of the open draft for the Research Group Leader Book
[about] [read more]


  • Post I of the Hiring series covered getting all your stuff together, deciding who you need, and getting out the ad.
  • Post II went into how to decide whom to invite, and how to communicate with applicants.
  • This post will discuss job talk and interview in detail.
  • Post IV will look at how to decide on a candidate, inquire with references, and what to do when you just can’t find the right person.

In many cases, the candidate’s visit to the lab will be the only time you meet in person before you make a hiring decision. Because you’d like to get to know the candidate, it’s good to have more than just an interview. I suggest the following structure:

  • Candidate gives a talk about his recent work (e.g. his Master thesis, her PhD work)
  • Interview
  • Tour through the lab and, maybe, the university
  • Candidate meets several members of your research group, preferably those he will have to interact with most later on

Let’s look at each of these items.

The job talk

Presenting your work is one of the most important aspects of a career in science. By asking candidates to give a talk, you get the opportunity to learn about several important aspects about both the candidate herself as well as her work. What exactly you look out for might differ from field to field; here are some points that I’ve used in the past:

  • How does the candidate present herself? Can you imagine her presenting work for your group at a conference?
  • Is the candidate able to find a balance between scientific detail (to show off her qualification and specialization) and generalization and abstraction (to be able to reach the audience that, usually, will not be particularly invested in the presented work?
  • Did the candidate choose a good story line? Does one part of the talk lead to the next?
  • Can the candidate handle questions? Does she understand them? Can she give a satisfying reply? If not, can she gracefully save herself? Does she allow other viewpoints? Does she stand firm on her own?
  • Do you get the impression that the scientific skills the candidate used for the presented work are what you are looking for?

The second reason a talk is a great tool for getting to know candidates is that your team and, potentially, others around your group can all meet the candidate and get an impression. Before the talk, instruct at least some of your team to ask questions.

After the visit, make sure to ask as many attendees as possible about their opinion. I’ve often found surprisingly converging opinions, but I’ve also had candidates about whom the feedback of the group was dichotomous. This is important feedback you shouldn’t overlook and ignore.

And third, be aware that it’s also the other way around: the talk is a great way for the candidate to meet everyone and get an impression about the environment he would be getting into.

Therefore, make sure your group makes a good impression! Have a drink ready for the speaker. Clear up beforehand whether questions should be asked during the talk or afterwards. But most importantly, what will make the most positive impact is when your group is curious and involves the guest in an interesting discussion.

The interview

The interview is the core of the application process. All the more you may be surprised that the correlation between what most interviewers get out of an interview and the later success of the interviewed candidate is zero.

This is because interviewers ask the wrong questions.

The best predictor for future behavior is past behavior. Therefore, the biggest mistake you can make is to just freely ask candidates about how they do something, or how they think about something. Instead, your questions should focus on uncovering how the candidate has acted in the past, and specifically what outcomes he has produced.

I recommend the book Who: The A Method for Hiring if you want to go deeper into outcome-related questions (advisable!). However, the book suggests a series of four interviews, which I suspect is overkill for most university / research labs. In addition, most of the positions we fill are connected to qualification – a PhD, a Postdoc –, and I think that it’s not sufficient to ask about performance in this context. I suggest the following blocks of questions, partly extracted from Who, for a single long interview:

  1. Of course, you want to know about the goals and interests of your candidate. You can ask questions like
    • What are your career goals?
    • Why are you interested in this position?
    • What are your expectations?
    • What do you want to have achieved when the contract ends?
  2. Find out what strengths the candidate has. You can ask
    • What are you really good at (professionally)? Try to get a good number of strengths, about 10. To learn about past behavior, ask for specific examples for each strength. Don’t settle for general replies such as what the candidate “usually” does.
    • What accomplishments are you proud of?
  3. Find out about weaknesses. Ask
    • What are you not good at, or not interested in (professionally)? As for strengths, try to get a large number of responses. This is difficult. Many candidates come prepared for a question about weaknesses, and follow the common advice to name strengths disguised as weaknesses, such as working too hard, being too persistent etc.
      Those aren’t what you are looking for. Aim at being able to delineate a profile of your candidate – this consists of what he can, but also of what he cannot do.
    • You can guide your quest by asking things like Tell me about a project that didn’t go well, and why; and What do you think your boss / collaborator / colleague will say when I ask him about your role and performance in this project? These questions prompt the candidate to think about herself in a third person perspective; apparently, we are more realistic about ourselves then (similar to when you answer questions about yourself while looking in a mirror…). Write down the replies to such questions so that you can check them with references later.
  4. If they haven’t come up, ask about specific areas that are important for the job. For instance, writing is a topic that a lot of PhDs and Postdocs have difficulty with. Again, ask about specific examples.
  5. Find out about culture fit. Relevant aspects are
    • what are the candidate’s expectations about guidance and mentorship? Do they fit with what you offer?
    • How will the candidate contribute to group life? Think of teaching methods to others; helping with statistics or programming; participating in group meetings etc. As before, inquiring about past behavior and concrete examples will prompt more reliable answers.
    • Another point that is often relevant is expectations about and willingness to teach.

If you conduct the interview together with another interviewer, it can help to agree on forehand on who will ask about what topics.

With all these questions, don’t forget to give the candidate the chance to ask her own questions, and have a short but complete introduction about your scientific project and / or the position ready.

Lab tour

If your research group does lab or experimental work, then I guess every candidate will want to see the rooms and equipment he would be working with. Besides that, the lab tour is great for two additional reasons.

First, it gives you the opportunity to probe the candidate more. I have often found that the conversation gets easier, more informal while walking around.
* You can use things in the lab as triggers to ask new questions, such as about experience with certain methods etc.
* Observe what kind of questions the candidate asks about the lab. For instance, if you are showing the equipment for a method the candidate claims to be fluent with, this often shows in detailed and competent questions. If such questions don’t come from the candidate, bring them up yourself: “How did you solve technical problem X in your past work?”. I’ve had some candidates tell me that, oh, that was usually the task of the lab technician – not the greatest proof for expertise, is it…
* Depending on the candidate’s background, the lab tour is a great opportunity to talk about specific work, such as experiments you’re running.

Second, the lab tour is also somewhat of a sales tour. Make sure to show the candidate everything that is potentially interesting for her. Speak about possibilities, about the funds you have (if you do) to add equipment if needed. Ask what the candidate thinks may be missing from the lab.

So, use the lab tour to sell your position, and to get to know the candidate more! If you feel you’ve learned enough during the interview, you can also hand the lab tour off to a team member. If you do, it’s good to discuss with that person beforehand what you find important.

Meeting other group members

Meeting future colleagues can be a key point of a candidate’s visit. It gives her the chance to ask questions she might not dare to ask you. By talking to group members, the candidate can get an impression of the group’s culture: Are the team members enthusiastic about their work and the team? Are they bored?

This part of the candidate’s visit is pretty much out of your hand. If your team doesn’t like you, it will breathe through. But even if you think that your team thinks you and the group are great, it’s helpful to discuss prior to the visit what you’d like for your team members to talk about.

  • Encourage your team members to openly talk about the group culture and climate, and to answer questions about how you are as a boss. Any new team member should know what she’s getting into, positive and negative.
  • Encourage your team members to ask their own questions of the candidate, so that they can add to their impression from the talk.
  • You can delegate certain topics to team members, such as the lab tour or presentation of a scientific project. Just always remember that the team member you delegate to can’t read your mind: Tell them what you find important, and what points she should not forget to bring up.
I’m curious to hear all about how you structure job candidate’s visits to your research group. And I’m especially interested in questions you’ve found useful in the interview. Write a comment!


Who: The A Method for Hiring by Geoff Smart and Randy Street is very helpful for getting a grasp on the kinds of questions that are useful in interviews.

The Professor Is In: The Essential Guide To Turning Your Ph.D. Into a Job is a book about getting tenure in the U.S. The interview part is interesting, but won’t help you so much to construct useful questions because it focuses on getting the candidate through partly weird questions of a committee. But the chapter on the job talk (Ch. 34) is good to look at when you think about the criteria for a good talk. It is also worth scanning over the chapter on outraging questions (Ch. 37) – the questions you shouldn’t ask.

The Manager Tools podcast has some episodes on hiring, such as Setting the bar high (but google for others). This podcast is always very detailed and step-by-step. Sometimes it all takes a bit longer than you wish it did, but you really get in-depth info.


This post is part of the open draft for the Research Group Leader Book [about] [read more].

Photo credit: dollen / Foter / CC BY

Leave a comment...