How do you structure a job talk?

Oct 28 2014 Published by under [Education&Careers]

Many people applying for academic jobs this season are well into the process. Interviews are happening across the country and more are on the way. If you're putting together a talk for an on campus interview, you'll want to make sure you're communicating to your audience the information they'll need to make a decision. Obviously, it's not all going to come down to your talk, but many faculty will only see you during this time.

The most critical piece, IMO, is figuring out whether you need to include a full Future Directions section in your feature talk. I've seen it done several ways, but what you present on the work you plan to do at the hiring institution is obviously the key. In my current job, the future directions was a separate talk. I still gave a 5 minute compressed version for those who wouldn't see the second talk, but it was just a teaser.

If you're asked to put everything in one talk, I would roughly break it down like:

30 min of highlighting what you have done

20 min of talking about what you will do

10 min for questions

People often say that their completed work is either too much or too diverse to cover in 30 or 40 or 50 minutes. This is good and probably why you are being interviewed. However, it is up to YOU to create a narrative about who you are and what you bring to the table. What led you from one training environment to another*? What is the flow to your science and what you have accomplished? What are you going to take from that and apply to your new awesome lab?

Your audience should come away with a feel for the flow and trajectory of your career. That is your take home for them. You want them to think "Wow, she did all that and is ready to really take off on an original path!"

Look at the structure of your talk and make sure that is front and foremost.




* Or, what is the best possible spin you can apply post hoc?

18 responses so far

  • odyssey says:

    IMO it's also critical to pitch your talk at the right level. You need to be able to bring everyone* in the audience up to speed. That means spending the time to explain things in a way that the people who know the least about your sub-sub-field understand. Remember, these talks are often used to judge your ability to teach as well as your science.

    * At least those that are teachable and/or awake.

  • bashir says:

    Pitch the talk at people *NOT* in your area.

    Don't feel like you need to prove how much stuff you've done. If you've been invited then you've done enough. Now is the time to tell them how awesome a subset of it is. What's your story? Provide details as needed to support the story. Anyone can go on data blast and show 37 figures from 37 experiments, doesn't mean it's interesting. Trust me I've seen plenty of data blast talks bomb.

    Really work on what next steps are. You need to appear to have a plan of awesomeness. This is not a contract or even a promise, so don't feel like you can't say anything unless you're 100% sure you can do it later. Plan big.

  • These are good points.

  • MCA says:

    I've bounced around quite a bit in systems and immediate questions, and I worry that the "big questions" that are big enough to encapsulate my history are "too big", and will be seen as vague or over-ambitious, especially compared to more "I study X in Y" folks.

    Does anyone have any advice on this? Do search committees generally acknowledge that any simple, linear career story is merely a simplification for the sake of presentation?

  • Established PI says:

    Your talk is HUGELY important, perhaps even more than the way in which you interact with people. You already made the cut and were invited, so they know you look good on paper. Now you have to wow them - get them excited and impressed by your accomplishments, by your ability to convey what you have done to a broad audience, and by hinting at bigger things to come if they would only give you your own lab and a big wad of startup money. You also need to be great a fielding questions - a critical element some people forget about until it is too late and they freeze like a deer in the headlights when they get an unexpected question from Professor Gravitas. Practice and refine your talk, get advice from seasoned faculty, bribe your friends with beer and pizza to hear you one more time. Great speakers are made, not born, so get going!

  • Comradde PhysioProffe says:

    Practice, practice, practice! But don't memorize. You need to be relaxed and smooth, yet come across as extemporaneous. And realistically, the best practice by far is giving real jobbe talkes. My first couple were shitte, and I didn't get offers. My last one was at the best institution by far, and that's where I took my first faculty position.

  • DJMH says:

    Is this on a visit where you'll also be giving a "chalk talk"? If so, I would do more like 35-40 min your own research, 10 min future directions.

  • DJMH says:

    MCA, I don't think it's unusual or problematic to have jumped around. You don't need to demonstrate that you are interested in one "big picture" question that encompasses all of your work, ever; you need to demonstrate that you can take what you've learned from these various fields and combine it into unique awesomeness.

  • Just came back from a job talk:
    1) No diagrams or cartoons. All equations with no effort to explain what the equations except verbal sentences (fail)
    2) No indication of how your work is as compared to other work in the field (fail)
    3) Minimal future work talk that only included high level problems (with no explanation of how you would tackle such problems; no saying that you would solve them doesn't mean any thing)
    4) No time for questions ! wtf is wrong with you

    PLS you are right on spot except for may be cut 5 min from future talk and put in questions or what I have done

  • Yes, if you have a separate opportunity to explain future directions, via a talk, chalk talk, or whatever, then you only need a cursory description of your future directions. But if it all has to be in one talk, then make sure you spend adequate time telling your audience what YOU want to do THERE. Borrowing on your training to tackle a new question is an excellent way to frame it.

  • DJMH says:

    "* Or, what is the best possible spin you can apply post hoc?"

    And really, isn't this what all of science is?

  • Dan says:

    > You need to be able to bring everyone* in the audience up to speed.

    Knowing who "everyone" is critical, and an important test of your ability as a communicator.

    In my biology department at a SLAC, some senior undergraduates know more about the area of a candidate than many of the faculty. Those faculty are the majority of the people voting on candidates, so you better be sure that they understand the big picture.

    The non-specialist faculty will also look to the faculty who are specialists in this area of biology for their opinion on the research, so you need to make sure the specialists have their curiosity satisfied.

  • Joatmon says:

    Do NOT try to cover everything you have done but rather focus on telling a story in 35–40 min. Then spend another 10 min to tell people your fundable ideas, why those are important ideas, and why they set you apart from everyone else including your research mentor(s).

  • K99er says:

    I've never seen anyone give 20 minutes of future directions in a job talk. The focus is usually on what was accomplished as a postdoc followed by a few slides with awesome ideas for the future.

  • Do they have a chalk talk as well, K99er? If that's the case, then I agree with a few slides, as stated above.

  • I went through several job interviews last year and I have a report of each one of them on my website. I recommend including diversity in the job talk. I spent about 15 minutes on one project and showed the depth of the work and then spent about 5 to 10 minutes on other projects and then only 10 minutes on future plan. It is important to make a list of some funding agencies and collaborators from across the country and especially from different faculties at the university that you are interviewing at. at the end, what people will remember is how you carried yourself, how prepared you were and how you handled questions. Practice is the key.

  • bashir says:

    Handling questions. That's a big one. I have seen people bomb the Q&A because they don't how how to handle questions properly. Learn how to deal with a question you don't know the answer to (admit it and then speculate). Or how to handle aggressive questioners (be nice but firm). Try to anticipate the types of questions you may get and have an idea of what to say. Not just picky method stuff but high level "why would we even want to do this?" types of questions.

  • cookingwithsolvents says:

    In my field you give two separate talks. 100% your public 'what I did' talk should be accessible and set up your 'what I will do' talk (which is private). In your 'what I will do' talk a slide on funding targets is a good addition to highlight your awareness of the landscape. Be ready for questions like

    'what will you do first?'
    what if that doesn't work?
    what if it DOES?
    how do you envision working with xyz in the department (xyz is in a related area)?
    what size group do you want?
    what kind of resources do you need?
    what can you teach?
    what do you want to teach?
    have you written proposals before? (yes=follow up: was it successful?)
    who will your main competitors be in your area?
    will you collaborate? in the univ? outside?

    Make sure you have enough details on-hand to answer technical questions but also be ready for the BIG PICTURE ones. You are applying to be a PI and your technical ability to get science done is right there in the published papers.

    Final note, if you do a lot of team/collaborative research and multi-author papers it is EXTRA important to say 'i did this, x did that' in your talk on your own work.

Leave a Reply