What attitude should a speaker have to going over time?

Question

From John Baez's Advice for a Young Scientist:

Your talks should be clear, concise, fun, exciting, and never ever run over time. For each extra minute your talk runs over, 10% more of the audience will decide you are a jerk and start fantasizing about you falling down a trap door.

After 11 minutes, 110% of your audience will start hating you.

That makes perfect sense, because that includes the people waiting for the room to open up for the next talk to begin.

But some people at my college like to take the opposite view. One lecturer never releases his class until he's run at least five minutes overtime. In the mandatory presentation class, the teacher maintained that going five minutes over the twenty minute talk period time was ok, but when one student went one minute undertime, she insisted that he repeat the assignment.

My question is: how should the speaker treat overtime?

Answer 1

Speaking for longer than the allotted time is unprofessional, and it can be very damaging and offensive in some circumstances. For example, if a conference schedules talks back to back, then it's not acceptable for one speaker to try to use part of the next speaker's time.

It may not be quite as bad in other cases, but it's still disrespectful to the audience. It tells the audience that the speaker believes his/her final words are more important than whatever else the audience might need to do, and it forces anyone with other obligations to risk drawing attention by leaving before the end of the talk.

Classes are something of a special case for two reasons. One is that the professor has a certain degree of power over students in the classroom, unlike a typical seminar speaker; another is that it's easy for professors to rationalize that they are going over time for their students' own good. Going over time is still problematic, but some people don't feel bad about it.

As for how to handle it, this problem should never even come up if the speaker is on top of things. It's important to keep track of the remaining time and adjust the presentation to omit details as needed to finish on time. If you screw this up, then you are out of luck if anyone is scheduled to use the room after you. Otherwise, you could apologize and offer an extremely brief summary of your remaining points. (I.e., if you suddenly discover that your time is up, that's when you should switch to the 30 or 60 second summary of your conclusions, rather than continuing with the seven minutes remaining in your talk as planned.)

Answer 2

I think the presentations teacher who is forcing students to repeat talks if they run too short is not working in reality. Her students will be very ill-prepared for real conferences, where such behavior is not likely to be appreciated.

As someone who has chaired a lot of sessions at conferences, I have to admit that I have minimal tolerance for people who try to run over their allotted slots. If your slot is 20 minutes, I will warn you when you have five minutes left and again at two minutes. If at 20 minutes, you're not on your conclusion slide, I will cut you off and ask the next speaker to set up their talk.

However, a lot of that issue comes up when one of two situations occur:

• The organizers of the event where the talk takes place do a poor job of managing the session and ensuring speakers stick to their time slots.
• The speaker is clearly unprepared for the time slot they have been allotted.

I recently attended a talk which should have been about 45 minutes long. When the speaker stopped, after nearly 75 minutes, he had only completed two parts out of seven in his outline. He had way too many slides, talked about each one for far too long, and generally ticked off the audience in the process.

As for lectures, as a student, I would tolerate a minimal overrun to "tie things up"—if the lecture is 90 minutes, for instance, a one- to two-minute overrun would be OK. But an extra 10 or 15 minutes would be unacceptable. As an instructor, I would rather end the class five minutes early and leave some material uncovered rather than run that far overtime.

Answer 3

Amongst the answers presented so far, I see a lot of strong sentiment, but am missing two things that I think are extremely important: 1) context-sensitivity, and 2) how a speaker should manage timing.

To the first point, the degree to which a speaker should be concerned with running over (or under) time is highly sensitive to the precision with which their talk is scheduled. Contrary to what most of the answers have said thus far, this can vary highly depending on venue.

• Lecturing a class is the most rigid context, as students often have other classes that they must leave to go to immediately at the end. You need to stop on time, period.
• Conference talks are typically designed for a five minute question period, which provides a cushion for going a couple of minutes over: you just get less questions.
• In settings aimed at discussion, such as workshops, there is often a looser schedule with more flexibility and buffer times built in. Here, one can feel comfortable running over somewhat longer as long as the extra time is primarily caused by interaction with the audience, as that is the purpose.
• In informal settings, such as an invited presentation in a group meeting, you can go over as long as the discussion takes you. In this type of setting, I have had a 15 minute talk turn into a 90 minute talk, because the people I was presenting to wanted to dig deep into discussion with every slide.

To the second point, for the more tightly constrained talks, it is important to also have a way of thinking about time management in order to ensure that one can end on time. A very useful method that I learned from one of the best speakers I know is to include an "accordion section" toward the end of the talk, containing material that is enriching but not strictly necessary. If you are running ahead of schedule, you can dawdle in the accordion section, explaining all of the lovely details. If you are running behind schedule, you can skim through.

For example, I might give a talk that ends with a couple of case studies. The first case study will get an in-depth treatment no matter what, while the second case study is there to show the generality of what I have been talking about. If I have enough time, the second case study can be presented in the same detail as the first. If not, then I can instead say something shorted, even to as short as, "X is fairly general, as can be seen by the fact that is works just as well in case study Y as well."

Thus, you should always plan your talk to be precisely on time, and include an accordion section that simplifies the task of controlling your timing. In settings where there is flexibility in the schedule, however, you should feel free to allow the audience to extend the time of your talk within the bounds of schedule flexibility.

Answer 4

The speaker should follow the guidelines given. Granted, normally one should limit a presentation to exactly 20 minutes. A range is much more common, and much more reasonable. Indeed, I would say the limit that was given was really 20-25 minutes (though not given in a very clear way).

It's like speeding. The posted speed limit is 65, so how should you consider yourself when you go over 65? You should slow down and fit within the posted limits.

In presentations, when you cannot fit your material within the allotted time, it shows your lack of preparation. If you just keep blabbing away, it shows your inability to follow your own structure. If you simply do not think the limit is important then you are showing a great disregard for the audience.

This is true of lecturers, too. If they do not care about the time limit they are simply being disrespectful. A philosopher would say being disrespectful to anyone is really showing your own disrespect to yourself.

In the end, people who do not follow the limits are saying they are special and the rules do not apply to them. Sadly, this attitude is all too common. Trying to change others is futile but we should do what we can to set a good example for others, regardless of our position.

Answer 5

Keeping to time is just basic respect, for your audience and for any fellow speakers.

Admittedly, it varies somewhat with context - the only presentation at a group meeting going over probably doesn't have many knock-on effects, whereas at a conference there's coffee going cold outside, parallel sessions getting out of sync, and generally far more going on and more people to annoy.

That said, a rule of thumb: Is your time worth more than that of everyone in the audience combined? No? Then don't finish late - even when they all filed in five minutes late (another pet hate of mine). Yes? No, it isn't.

At the courses and conferences where I've presented, I've always been the last speaker before lunch, where you're already struggling to keep the audience's attention before you start. With a bad chairman, half of that last slot can disappear easily; that speaker is then faced with either making everyone late for lunch or mutilating his/her presentation. Bad chair or not, it's disrespectful for earlier speakers to put other speakers in that position. Admittedly, my experience is that you get massive brownie points for getting back on track and saving the lunch break, and you can get some interesting conversations in the lunch queue as a result, but you shouldn't have to.

Within the department, where we're being kept from actual productive work, I like to count the people in the room, calculate a rough figure for their combined hourly cost, and hence work out how much of our hard-earned funding is being wasted every minute that the speaker goes over. Needless to say, I'm not paying attention while doing this.

The worst example of going overtime that I've experienced was at a conference with a number of lunchtime sessions. Someone whose name started with Sir was assigned one of the first ones, and he was still talking (and his audience still hungry) when the second sessions were over and we were meant to be back in the main auditorium. As it happens, that was the same room where we were booked to give the second session; we ended up giving a software demo standing in the hallway, with one of us holding a laptop for the other and people pressing in to see what should have been projected on the wall.

So, having been on the receiving end of it both as audience member and speaker, my personal tolerance for it is very low indeed. From discussions with many colleagues over many years, I haven't found one yet who's OK with it, even when they find the topic interesting.

Answer 6

It's very different to require a 20 minute presentation. It would follow that speaking more than 20 minutes is okay; perhaps the recommended time is 20-25 minutes but the professor is very strict about the lower limit to make sure you have that much material. Perhaps not the best didactic method but analogous to a minimum page length in writing.

You are trying to see a generalization where there shouldn't be one. There's a culture of being late in America just like there is a culture of going over on talks. It's problematic though, especially because some prefers eschew the boundary entirely and run well into your next class. So yes, that is a thing, and it is problematic, but you should not treat assignments as the same category.

Answer 7

Going overtime in a conference is a nightmare for the organising committee. Usually, there are parallel sessions focusing on different topics, and people try to make their own collision-free schedule to make the most of the conference, if one talk gets shifted, the people changing rooms will be affected. Also, it is common to have a few keynote speakers for the whole conference, but if one session is very delayed, you are either forcing the whole conference to wait for you (and then you will get the hate of 10x the audience of your talk), or make people miss the keynote (that is presumably of particular interest).

Exact timing of a presentation is difficult, but perfectly doable. You should rehearse it until you are confident. You can always take an extra minute from the questions, but then be aware that you are depriving another person from speaking.

Answer 8

Don't do it. If you do, don't expect attention from anyone after your allotted time. The busier people in the audience may very well need to leave at the scheduled end-time in order to make their next meeting.

Answer 9

I think the amount of time leeway in how long you can talk is proportional to how long you are talking. If you are supposed to give a 4 minute summary and it takes 6 minutes, that is too long. Similarly, giving a 4 minute summary in 2 minutes is too short. Alternatively, I think it is perfectly reasonable if a 60 minute lecture takes anywhere between 58 and 62 minutes. I think audiences will generally give you up to a 10% margin for error. Many talks, both research and teaching, also include a small period of time for questions at the end which provides a nice buffer.

For a typical teaching scenario a 1 hour time slot is often only 50 minutes of teaching and you might allocated your self 3 minutes for questions at the end. This means you should be aiming for your class to take between a 43 and 51 minutes. If you hit 51 minutes, you apologize and tell the students they can stop by your office to ask questions and that there will be a chance for questions at the next session and you let them out 1 minute late. If you finish early you can ask for questions. If you do not get questions you can recap the syllabus and schedule for a few minutes and let them go 5 minutes early.

Answer 10

There are many contexts where it is extremely important that talks fit the allocated time as precisely as possible. Some other contexts benefit from more flexibility. A speaker may be able to precisely predict how long it will take to present a certain corpus of material, but it's much harder to predict how long it will take to present the amount of material the speaker has which the audience will be interested in. In some contexts it may be better to have the speakers guess what the audience will want to hear and present a fixed corpus of information, but in others it may be better to have speakers adjust the lengths of their presentations according to the audiences' levels of interest. The latter approach would likely be better most of the time but for the fact that speakers and audience members may have conflicting time obligations elsewhere. The problems created by conflicting time obligations, however, often outweigh the benefits of such flexibility.

If in a certain context a speaker could go overtime without creating difficulties for himself, the audience, or anyone else who would want to use the space, then it may be good for the speaker to adjust the length of his talk according to what the audience wants to hear. Such situations are not the norm, however. If a particular conference or convention has many activities which people will be able to do at "any time", and most visitors will be expected to want to spend a lot of time on such activities, then it may make sense to have talks which might benefit from going overtime be followed by "free time".