Tuesday, June 25, 2013

How do you give a scientific talk to people outside of your field?

A reader recently asked this question: how do you give a job talk about your work to scientists and technical people who are not in your field? Specifically, this was a person who was interviewing for a non-bench-research-but-still-technical position, coming out of an academic postdoc.

I have actually done this -- and unsurprisingly, it is harder than it seems. You can get messed up in the details of trying to "teach" too much and get tripped up in the details of your work. I don't claim to have an answer, but here are a few suggestions:
  • Teach your terminology early, and keep it simple. 
  • Have two or three clear stories, with obvious "problem solved" sections. 
  • Encourage your audience to interrupt you to clarify.
Readers, have you faced this situation? What would you recommend?


  1. This is a super-important skill and when you get it down, a lot of it carries over to presentations within your field too. The main thing is to focus on one important point per slide, and if possible, make that point the slide title. Example of a wrong title "UV-Vis titration data" vs a good one "Stronger binding of A vs B confirmed by UV-Vis titration"

  2. Read Guy Kawasaki's "Art of the Start".

    He has a rule called the 10-20-30 rule when presenting to potential investors, partners, or customers. No more than 10 slides, no more than 20 minutes, and no less than 30 pt font. Now this rule isn't hard and fast, in fact, it's more like a goal and you'll probably end up going to more like 18 slides 30 minutes and 24 font, but it's a still a great guide to making presentations to people outside your field.

    He also has a template for what each of those 10 slides should cover. You'll have to obviously change that template for a job interview talk, but really its not all that different. You're trying to sell yourself.

  3. I would add that you should focus on the result, rather than the method you solved the problem. Scientists love to talk process, hardly anyone else cares.

    -- Unstable Isotope

  4. While you're not going to want to spend all of your time teaching background material, you at least need to teach enough such that you can explain to laypeople why your work is important. If the audience can't understand what you did and can't understand why anyone would care about it, it doesn't matter if you spend all or none of your talk speaking about your specific results.

  5. If there is any possible way to relate it conceptually to something in their field, do so. Also understand their value system, as it were: e.g. biologists like explanations to be straightforward and elegant, uncomplicated--they feel that complicated math-y explanations are used to hide weak data. Engineers tend to feel that if an explanation isn't super-complicated, it's not very good--they value complexity and consideration of about a zillion variables (even if you're only saying, "we considered this and it was not relevant because").


looks like Blogger doesn't work with anonymous comments from Chrome browsers at the moment - works in Microsoft Edge, or from Chrome with a Blogger account - sorry! CJ 3/21/20