How to Give a Talk(zapier.com)
zapier.com
How to Give a Talk
https://zapier.com/blog/how-give-talk/
11 comments
I think there is a place for this kind of article. Someone who has just given their first talk may be able to relate to the needs of other soon-to-be first-timers more closely than an experienced speaker. Also, I think an article like this serves the author by forcing self-analysis of his or her procedure and discovering the ways of improving the process for future speaking engagements.
Agreed, but the title should be "Post-mortem of my first talk" or "What I learned from my first talk" rather than "How to give a talk", imho. Otherwise it's just misleading and disrespectful to readers. Maybe I'm oversensitive about these things.
I agree with you, but I guess it is interesting seeing how someone who hasn't done talks before would approach the issue. I certainly imagine that if he does another writeup after doing lots of talks it would be very different.
I'm not sure I agree with the advice to write out your talk in complete sentences. I think you'd be tempted to memorize the talk. In grad school, we gave lots of technical talks. There was one guy who wrote everything out, and he gave by far the worst talks in the department. He sounded like he was reciting something, rather than just speaking. He would pause awkwardly when he forgot where he was in spewing out his text. It was just painful to watch.
I think it is far better to know what you're going to say by practicing your talk out loud, a number of times. You have your main points in your head, but your delivery will be a bit different each time. That way you can sound loose and natural.
I think it is far better to know what you're going to say by practicing your talk out loud, a number of times. You have your main points in your head, but your delivery will be a bit different each time. That way you can sound loose and natural.
The trick is to write out the entire speech, word for word, but don't force yourself to stay 100% on script when you actually give the talk.
All through school as I was growing up, I was always instructed never to write out a speech or presentation in complete sentences, only to plan an outline, write down the main points on notecards, and then practice, practice, practice. The idea was that this strategy would result in a more natural talk.
The result, for me, was basically a recitation of the main points written on the notecards interspersed with some mumbling and fumbling. Every time I would practice, I'd find myself very consciously thinking "ok, now what do I say to get to the next main point?" It would be different every time, and the result was always a stuttery mess. My problem was that, with practice, the main points were the easiest part to memorize. It's the stuff in between that I would forget and stumble over.
Now, I write a presentation as I would write an essay: plan, outline, draft and iterate, and read out lout repeatedly. Writing out the entire talk forces me to think about every supporting detail and every transition. It helps me to analyze the entire talk: I can easily pick out portions that don't flow well or don't make sense - or avoid writing them altogether, since every word of the speech is undergoing a drafting and editing process. I can identify phrases that are confusing and rewrite them, or even better, find phrases that particularly stand out and highlight them for emphasis. I can time the speech to within a very narrow window and find opportunities to expand or reduce the content, both on the paper and on the fly.
Finally, I practice - not until I've memorized it (unless it's relatively short), but until it sounds natural. I take notes on the bits I stumble over repeatedly, and those become my notecards.
This strategy probably isn't effective if you have to give a different talk every day or if you have minimal time to plan. But in my experience, it results in the most well-planned, natural talks with the highest content quality.
All through school as I was growing up, I was always instructed never to write out a speech or presentation in complete sentences, only to plan an outline, write down the main points on notecards, and then practice, practice, practice. The idea was that this strategy would result in a more natural talk.
The result, for me, was basically a recitation of the main points written on the notecards interspersed with some mumbling and fumbling. Every time I would practice, I'd find myself very consciously thinking "ok, now what do I say to get to the next main point?" It would be different every time, and the result was always a stuttery mess. My problem was that, with practice, the main points were the easiest part to memorize. It's the stuff in between that I would forget and stumble over.
Now, I write a presentation as I would write an essay: plan, outline, draft and iterate, and read out lout repeatedly. Writing out the entire talk forces me to think about every supporting detail and every transition. It helps me to analyze the entire talk: I can easily pick out portions that don't flow well or don't make sense - or avoid writing them altogether, since every word of the speech is undergoing a drafting and editing process. I can identify phrases that are confusing and rewrite them, or even better, find phrases that particularly stand out and highlight them for emphasis. I can time the speech to within a very narrow window and find opportunities to expand or reduce the content, both on the paper and on the fly.
Finally, I practice - not until I've memorized it (unless it's relatively short), but until it sounds natural. I take notes on the bits I stumble over repeatedly, and those become my notecards.
This strategy probably isn't effective if you have to give a different talk every day or if you have minimal time to plan. But in my experience, it results in the most well-planned, natural talks with the highest content quality.
Exactly this. Learning things word-by-word is pointless and unrewarding. Being familiar with them, however, is a great idea.
It depends on your preparation time.
The best talks, IMO, are those written out in detail and completely memorized through many practice sessions. If you only do 2-5 practice runs, you're probably better off with an outline-based or extemporaneous approach. If you do more, say 10-25 practice runs, the script-based approach becomes very effective because you can deliver it naturally from memory without pauses or confusion.
It sounds like your guy probably only practiced his talks once or twice, if he practiced at all. It's tempting to view a script as a way to minimize practice, but really it's just the opposite. It takes more practice to refine the script and to present it naturally.
The best talks, IMO, are those written out in detail and completely memorized through many practice sessions. If you only do 2-5 practice runs, you're probably better off with an outline-based or extemporaneous approach. If you do more, say 10-25 practice runs, the script-based approach becomes very effective because you can deliver it naturally from memory without pauses or confusion.
It sounds like your guy probably only practiced his talks once or twice, if he practiced at all. It's tempting to view a script as a way to minimize practice, but really it's just the opposite. It takes more practice to refine the script and to present it naturally.
So I introspected what I did and realized something I didn't include in the post.
I was under a time crunch to memorize my talk (I think I started about 4-6 hours before) so the sentences were a way to speed that along.
The process went like this:
1. Read talk end to end out loud from sentences
2. Start trying to time sentences with slides (with sentences still there)
3. Remove sentences but see all the slides still
4. Only see the current slide
Took me about 2-3 hours to nail down. By step 3 and 4 I was already improvising most of the sentences each time though and only the core point stayed constant.
I was under a time crunch to memorize my talk (I think I started about 4-6 hours before) so the sentences were a way to speed that along.
The process went like this:
1. Read talk end to end out loud from sentences
2. Start trying to time sentences with slides (with sentences still there)
3. Remove sentences but see all the slides still
4. Only see the current slide
Took me about 2-3 hours to nail down. By step 3 and 4 I was already improvising most of the sentences each time though and only the core point stayed constant.
Great refinement mike. I did speech (forensics) throughout all of high school and the trick is exactly what he says at the bottom - you work through the speech as is, with full sentences because it allows you to take advantage of excellent writing and structure. Then, you improvise on top of it so you 1) don't get stilted and disconnected from your audience and 2) you can add content based on what people are responding to, or even other input from the day (previous speeches, etc)
I think writing it down in complete sentences is useful largely if you're a competent writer to begin with. I am, for instance, and when I write it's very similar to how I talk normally, except that the thought process is more complete than when I'm extemporaneously chatting.
What happens when I write down a planned talk is that I marshal all the ideas I have about what I want to say into a format that forces me to check what thoughts flow naturally into others, what doesn't fit anywhere naturally, and so on.
As nlawalker points out, the trick is to not worry about staying on script. The point of a talk is the talking; otherwise you could just hand out copies of your transcript and there wouldn't be any point in putting everyone in the same room for 15 minutes. And talking, rather than reciting, is about the little "mistakes" and variations and spontaneous reactions you have to what's going in the room. Someone looks a little confused and you can choose to drill down on a point even if it's not in there. People look bored and you can cut something dry and unimportant.
What happens when I write down a planned talk is that I marshal all the ideas I have about what I want to say into a format that forces me to check what thoughts flow naturally into others, what doesn't fit anywhere naturally, and so on.
As nlawalker points out, the trick is to not worry about staying on script. The point of a talk is the talking; otherwise you could just hand out copies of your transcript and there wouldn't be any point in putting everyone in the same room for 15 minutes. And talking, rather than reciting, is about the little "mistakes" and variations and spontaneous reactions you have to what's going in the room. Someone looks a little confused and you can choose to drill down on a point even if it's not in there. People look bored and you can cut something dry and unimportant.
People interested in improving their public speaking skills should really consider joining their local Toastmasters club. Becoming a good public speaker takes a lot of learning, practice and good feedback..
Editing to expand on this: Toastmasters is a non-profit organization. In Toastmasters you follow a structured program to improve your public speaking skills, initially working through some canned speaking assignments to focus on specific skills, and over time working on more complex projects. As an example, you may work on structure in one project and something like vocal variety in another.
A big part of Toastmasters is feedback so you will get good feedback on how you're doing and where you can improve and you will also learn how to give others feedback.
Since the local clubs are run by the members there are also many general leadership training opportunities in a Toastmasters club. Someone will be in charge of doing PR (VP PR) or taking care of many other things that are required to have a healthy club (which is not unlike running a mini-business).
There is a minor "cultish" feel to it and a few semi-weird customs (e.g. clapping at the end of a speech) but all in all it's good fun and a great way to improve many skills a lot of us don't get to practice in our day to day engineering life...
Editing to expand on this: Toastmasters is a non-profit organization. In Toastmasters you follow a structured program to improve your public speaking skills, initially working through some canned speaking assignments to focus on specific skills, and over time working on more complex projects. As an example, you may work on structure in one project and something like vocal variety in another.
A big part of Toastmasters is feedback so you will get good feedback on how you're doing and where you can improve and you will also learn how to give others feedback.
Since the local clubs are run by the members there are also many general leadership training opportunities in a Toastmasters club. Someone will be in charge of doing PR (VP PR) or taking care of many other things that are required to have a healthy club (which is not unlike running a mini-business).
There is a minor "cultish" feel to it and a few semi-weird customs (e.g. clapping at the end of a speech) but all in all it's good fun and a great way to improve many skills a lot of us don't get to practice in our day to day engineering life...
+1 on Toastmasters
As a CTM, TM will do the following:
0. Depending on the club, give you recorded video of your speeches along with your assigned evaluator evaluating.
1. Eliminate your verbal (umm, yeah, know) tics and minimize your non-verbal tics (swaying, pacing w/o intent, arms crossed)
2. Assuming it is hosted outside your normal workplace, provide a safe supportive environment for failing and learning
3. Learn how to mentor. Once you get going, you will pass on what you have learned by teaching it one on one
4. Not a benefit directly - but provide an outlet to channel energy towards non-coding goals (getting the CTM designation - which requires 10 speeches is a function of initial speech velocity and consistent effort)
5. Teach you impromptu speaking by forcing you to talk 1-2 min on random topics
As a CTM, TM will do the following:
0. Depending on the club, give you recorded video of your speeches along with your assigned evaluator evaluating.
1. Eliminate your verbal (umm, yeah, know) tics and minimize your non-verbal tics (swaying, pacing w/o intent, arms crossed)
2. Assuming it is hosted outside your normal workplace, provide a safe supportive environment for failing and learning
3. Learn how to mentor. Once you get going, you will pass on what you have learned by teaching it one on one
4. Not a benefit directly - but provide an outlet to channel energy towards non-coding goals (getting the CTM designation - which requires 10 speeches is a function of initial speech velocity and consistent effort)
5. Teach you impromptu speaking by forcing you to talk 1-2 min on random topics
Clapping at the end of a speech doesn't sound weird at all.
This may be cultural and context dependent but we had a company club and we had our meetings in this open seminar area. I know people looking on us from the outside considered the clapping to be a little weird. It works well though as positive feedback to speakers...
The applause is important because you need to congratulate the speaker for getting up there and talking no matter the quality of the talk itself. Because that's what Toastmasters is about.
You get to call it cultish when they hide that stuff. :P
You get to call it cultish when they hide that stuff. :P
I gave my first talk this year, and I find this very backward.
I gave a talk because I had something to talk about that I thought other people would be interested int. I chose the venue/event those people would be at as they would benefit most from my talk.
The 'Learn from your talk' segement is great, but what I wanted to 'learn' from my talk, was not only 'how to give a better talk', but I wanted to know from the audience what they're thoughts were on the topic. That one might be specific to different subjects, but the first two I'd hope would be obvious for any presentation.
I gave a talk because I had something to talk about that I thought other people would be interested int. I chose the venue/event those people would be at as they would benefit most from my talk.
The 'Learn from your talk' segement is great, but what I wanted to 'learn' from my talk, was not only 'how to give a better talk', but I wanted to know from the audience what they're thoughts were on the topic. That one might be specific to different subjects, but the first two I'd hope would be obvious for any presentation.
This is a good way to think about it! I suspect most people get invited to talk and say yes (versus trying to find a venue) because that's what the venue model supports.
Learning how to give a better talk is just a proxy for: how can I better communicate my ideas to the audience so they remember them?
In 2 months no-one is going to remember any of the details about your talk but they'll remember the experience and if you have resources online, they'll seek those out.
Learning how to give a better talk is just a proxy for: how can I better communicate my ideas to the audience so they remember them?
In 2 months no-one is going to remember any of the details about your talk but they'll remember the experience and if you have resources online, they'll seek those out.
Get drunk. Write it, and be complete. Go to bed. Sleep poorly. Get up before alarm. Caffeinate. Move your rousing conclusion to the top and try again, picking off your darlings. You're finished when you can't stand yourself anymore and it's time to shower. Think of something much better during steam. Dragon dictate while dressing. Rehearse en route -- a problem if you're just upstairs in the conference hotel. Email text to friendly teleprompter girl as sulky technician hooks up your mic. Wing it. Humbly accept false adoration. Cash the check. Add line to CV. Iterate.
> Learn from your talk
This is probably the easiest and most underutilized technique to improve at public speaking.
This is probably the easiest and most underutilized technique to improve at public speaking.
This is a resource I put together after giving a talk at IowaConf this year (http://mikeknoop.com/iowaconf-road-to-zapier/)
It's huge (4,000+ words) but if you ever give a talk it could be really useful.
It's huge (4,000+ words) but if you ever give a talk it could be really useful.
What’s with memorization? I give mostly short presentations, but usually I just need few bullet points on a slide and some solid eye brain control to calm down my nerves. Remember to talk slow, don’t wave too much, don’t stand too still, make it about few core ideas, nobody will get more than few ideas from your speech anyway.
Thank you for posting this Mike. Preparation is indeed an act of respect towards the audience.
Patrick Winston's video "How to Speak," available through isites.harvard.edu, emphasizes the same philosophy and promotes similar techniques for effectiveness.
Lot of work. Well done.
Patrick Winston's video "How to Speak," available through isites.harvard.edu, emphasizes the same philosophy and promotes similar techniques for effectiveness.
Lot of work. Well done.
I think you'll do good as long as you don't do what I typically do... put together the talk + code samples + slides two hours before the talk is scheduled to start! :)
That works for some people, not being prepared that is. For others it doesn't. Whatever works for you and makes your audience happy is great :)
I've given quite a number of talks to decent size audiences and there are some giant smoking turds of BS in here that I hope nobody giving a talk ever reads. Take for example this:
"Depending on the length of your talk, the memorization needed, and your speaking time, you can either start practicing the day before or the day of your engagement."
What the hell are you smoking? Practising a talk for the first time the day before you deliver it (let alone the actual day) is just about the worst advice I ever heard. Unless you are a certified egomaniac, this will probably destroy your confidence as you realise you can barely stutter your way through even the first two slides in the time alotted.
I guess I have to apologise to the author for sounding negative about something he has clearly put huge amounts of work into in an attempt to be helpful, but this kind of post just rubs me the wrong way.