Have said this before, will say it again in case it helps. I don't know why this works for me but it manifestly does.
1. Write slides, more than you think you need, and try to have fewer than 1/4 of them include bullets.
2. Print the slides out, one sheet per slide.
3. Tape slides to wall in a grid.
4. Rehearse the talk to the grid of slides on the wall with a sharpie in your hand. Any time you have trouble with a slide or trouble transitioning from one slide to the next, note it (or fix the slide) on the paper slide.
5. Reorganize slides in Keynote (or whatever), print them out again, repeat.
This may just be a bit of a ritual for me, but one thing it definitely does is it creates a context where I don't feel silly rehearsing by myself.
I disagree with one point: don't start with writing out the talk. Otherwise the talk will feel unnatural, stilted... Instead, practice speaking your way through it - and write down nice turns of phrase that you come up with while talking. This will resolve most of the flow issues you'd get with starting with writing, most of the awkward sentences, keep the whole thing flowing naturally, and generally result in a better talk.
Preparing a talk by writing it out may work to some extent, but it's as unnatural as preparing an essay by speaking it out. The two are very different formats, and you should approach them differently.
I've found that the weakest part of many presentations is in the transition from one point or idea to the next.
Shaky transitions are a prime cause of audience disinterest, and poor flow can quickly remove the auditor from the energy and flow the speaker has established, which makes such a time perfect for checking twitter and FB and otherwise disengaging.
The boundaries between ideas should feel natural and smooth, and I think for many it's difficult to arrive at a natural transition without getting very intentional about it, i.e. writing it out.
So I say write it down, edit, speak chunks out loud, and edit more.
In this way, the presenter can craft the most important sections with style while becoming familiar enough with the material that the delivery feels quite natural.
One thing I find useful for transitions is to set up some unrelated points in advance.
So instead of doing :
Thing a, thing b, thing c
I do:
We had to do a few things, then tie them together, First thing a, so we also had to do thing b. finally to get going we had to thing c. They all work together like result.
Another transition method that works for me is to just announce the seeming non-sequitor. "Now I need to talk for a minute about X. I know it seems unrelated, but it turns out to be vital..."
I find that I don't need to write down nice turns of phrase - if they're really nice, I naturally remember them, and can't help but hit them when I get to the subject.
(speaking as someone who literally gave buttloads of presentations all over the world for over a decade)
if you use slides i suggest reading cliff atkinson's blog "beyond powerpoint bullets". in it he lays out ideas, principles and templates for PPTs (and any other preso tools) that tell stories. these are based on cognitive studies, like the number of bullet points you should use based on how much your audience can hold and remember.
secondly, advice from one of my ph.d. advisors: "tell 'em what you're going to be telling them, tell them, and then tell 'em what you told 'em". the take home message is not a punchline, so don't be afraid to "spoil" it up front and repeat it.
don't read your slides, use images (see the first paragraph) and tell a story. talk with passion, enthusiasm, and joy. RAVE: rate, articulation, volume, emotion (from mrs hedaman, my old theater teacher back in high school), remember that.
love what you do, and share that joy with your audience. that's the goal.
"tell 'em what you're going to be telling them, tell them, and then tell 'em what you told 'em"
This is common advice. http://perl.plover.com/yak/presentation/ disagrees emphatically with it. Consider well that mjd both gave a lot of presentations, AND managed to consistently get the top speaker rating at a major conference.
My personal interpretation of the "tell 'em what you're going to be telling them, tell them, and then tell 'em what you told 'em" advice is to never tell people something the same way twice in the same session.
Poor presentations often do something like this:
* You should Foo because of X and Y
* We did Foo because of X and Y
* We've told about Foo. It solves X and Y
More entertaining presentations do something like this:
* Our life was full of problems X & Y - and then we found Foo
* We started doing Foo, which solved X & Y - with bonus Q
* These folk also did Foo - and found bonus Z
Repetition of the topic isn't necessarily bad. Doing it in the same way, or ritually, is.
I was trying to distill some good advice for OP, but you said something that struck a chord, and so I'm just going to reiterate it:
>"tell 'em what you're going to be telling them, tell them, and then tell 'em what you told 'em".
Telling them exactly what you're going to tell them provides an incredibly effective anchor for their attention. It gives them a quick mental road map, and the opportunity for them to think about and place each point you make on that map as you move through the presentation.
I'm curious, how did you et into the 'speaking business'? It seems a lot of these conferences only invite people to speak who have already spoken at large conferences. I did a talk at a local usergroup (5 people attended) and I loved it. I want to give talks at bigger conferences, but my proposals always get rejected. It seems once the ball is rolling, it just keeps rolling. Got any advice?
Keep blogging on the topics you want to speak about. So if you want to speak about X, blog at least once a month about things related to X.
Find out who is doing the speaker selections (even if they are just shortlisting) for the conference. Try and find a way to 'meet' them (this may be online). Find if they have a blog, and comment thoughtfully on their blog. If it's a yearly conference, go to the conference and seek them out and introduce yourself. Go to a session on the topic you want to talk about, and get up, introduce yourself (it feels tacky and awkward, but comes across as very professional) and then ask an intelligent question that will help the presenter cover an aspect they maybe didn't get time to cover. Ideally not a 'gotcha' question.
The aim is that when your submission is on that persons desk, they immediately have a picture in their head of who you are, what you know, and that you can write coherently on the topic at hand.
I followed this prescription and went from nobody wanting me to doing several invited user groups per year and speaking at at least one big conference per year, and also being invited to deliver webinars with quite large audiences.
Finally, don't mess up your big chance when you get it, and beg the attendees to put in evaluation forms if they are there. Hand out candy if you have to.
A call for papers that I read yesterday seemed to sum this up well:
"Your submission will be judged by your ability to communicate succinctly in writing, how good we think the talk sounds, what other people say about you behind your back and whether, frankly, we feel like people are going to be happy they paid money to see you yap for an hour."
It sounds a little harsh, but I think it's true. Getting to know people involved in the community or conference organization will pay off.
It's a big leap to go from a short user group presentation to hundreds of people listening to you for an hour. So you should expect to spend time getting to know people and money on attending other events in order to get there.
Of course, if you have a truly awesome new idea that nobody else has thought of, you can skip that step, but very few people do.
well, for me it helped to have a company, a team, and projects with great data behind me in an area that few people could address at our level. that and a great PR group, although i preferred to drive my own opportunities (less marketing crap, more technical).
i am, however, off the stage and off the road by choice. couldn't be happier.
The most important advice I've ever been given is to time myself talking through my slides. It is incredibly difficult to guess how long a particular presentation will take, which makes things rather awkward when you have conference organizers with schedules; talking through your slides gives me an accurate number which can result in me adding or removing slides. Even when I don't change the slides, knowing if I have plenty of time or need to squeeze helps me determine how much to re-iterate points and let things "sink in" before moving on to the next point.
I've had to speak often lately in front of small and medium sized groups, and the biggest thing that has helped is to just practice my talk several times alone.
Obviously giving the talk in front of others is important, especially when it comes to message and content, but a practice run for other people will obviously be a little rough around the edges from a word-flow perspective (assuming you have a hard time with this like myself).
Lately I try to give my talk to a blank wall at least 3 times before I will talk in front of real people so that I really know my slides and I know how long it will take. Usually you have a fairly strict time limit, so running through the talk with a stopwatch is critical.
I've also tried writing my talk out and simply reading it for practice, but it always sounds awful and is much shorter than an ad-lib version, which is never the same twice. I imagine people who have had drama, or any other acting/standup experience have a great advantage, just because they seem to be able to say almost exactly what they want, two times in a row. Perhaps it just has to do with have done enough speaking, that you become used to saying things in front of people.
This won't apply to most people, but if your talk involves any math at all, use LaTeX + Beamer (www.math.binghamton.edu/erik/beameruserguide.pdf) to prepare pdfs of your slides. They will look much better. The beamer user-guide also has a chapter on presentation preparation. You can export outlines from Emacs org-mode as pdf slides via beamer, which can be (depending on the version, etc) a little fiddly but is much easier than typing the slides up directly in LaTeX.
FWIW, I started feeling much better about presentations once I convinced myself to view the audience as friends who are trying to help me understand the material I'm presenting, rather than a group that I needed to try to impress.
While Beamer is fine, you can also generate HTML-based slides with both Org-mode and Pandoc. You can use MathJax for the math. I personally prefer HTML slides, but it really is a matter of preference. (I also find it easier to customize with CSS than LaTeX.) The real beauty is that you can use the same document to generate both, so it's easy to try each option.
More generally, I find using Org-mode or Markdown far easier than any WYSIWYG program. Life is much easier when you separate the content from its presentation.
Type in your equations/math, hit render, drag the PDF (or many other image formats) to your slide.
LaTeXiT gives you the best of both worlds; wonderful rendering of math, with the bonus of using Keynote (which is the best presentation software on any platform in my experience).
That looks nice; part of the advantage of beamer, though, is that the equations and (especially) inline variables are automatically positioned correctly and consistently -- it looks like that might not be the case for LaTeXiT.
You can also use this to produce PDF slides that have multimedia embedded in the PDF itself - something guaranteed to impress as you seamlessly play a video without having to alt+tab, buffer, or go find an IT guy to work out why the wireless has decided to stop working.
On one hand, keeping slides simple and avoiding "eye charts" clearly makes for better readability, and reduces the risk of the presenter "reading from the slide."
OTOH, a set of slides like that is utterly useless after the fact, unless it's accompanied by a transcript. Both audio and video recordings require serial attention for 15/30/60 minutes that I typically will not have, and I certainly can't see doing that to anyone else.
A question to the assembled: If you make "Presentation Zen"/"Merlin Mann"/etc. low-text high-concept slides for your talk, how do you make sure your presentation is accessible after the fact?
The best way of handling this in my opinion is to publish an annotated version of the slides afterwards. Here's a good example: https://speakerdeck.com/zachinglis/notes
I would agree for a talk you should optimize for the primary medium, being live presenting. Sure if you want to spread the word you can create other content with the same topic name, like a good blog post, to help you get the message out.
There's noting worse than crappy over texted slides where I attend and am thinking, I could have just stayed home and read the slides.
Apparently there was a classic talk by Dr Nic from Engine Yard who also does stand up comedy, the speaker didn't show up for some reason so Dr Nic jumped up and gave the talk cause he could just read the slides like everyone else.
(disclaimer: wasn't there so facts may vary)
I'm actually the opposite: I tend to avoid being paid to speak (one time I was actually paid, but I didn't cash the check), but if you want to record a video of me speaking for your usage on your website, then I consider that something you really need to be saying up front and which you should probably be paying for (and thereby which I often just won't do, as I usually avoid being paid; my standpoint on avoiding being paid will probably change over time, though).
The first issue is that it isn't really "publicity" for me (to respond to your linked point #3 about wanting publicity), it is free and valuable content for the event: to the extent to which I obtain publicity, I get it from the people at the event that I am addressing with my talk, not from the people who run the event; the second one is that recording the video to go with the slides doesn't mean that the content will now be understandable, as I feel truly awesome content is often somewhat interactive (I try to break the ice quickly to allow questions to interrupt me); and, finally, it has a severe cost on the talk... when you are speaking with a smaller circle, you can tailor your talk to directly meet their needs (I make it a point to meet as many people as I can before I give a talk to ask why they are attending the conference, what they wanted to get out of it, what company the work for, etc. and I then retarget my talk to directly serve those interests, use language appropriate to the audience, and sometimes can come up with specific and more useful examples), but when you know that thousands of random people are going to be watching this recorded later you suddenly have to talk to a general audience and have to be more guarded about stories and predictions (as you will certainly be called on them years later in drastically different contexts)... it really just ruins the experience.
(edit:) In essence, it is actually a different medium from a "live" talk: a talk that is being recorded is to a different audience (undermining the goal of having a conference tailored to specific topics and interests) and has different constraints (difficult interactivity with either the audience or the slides, a tradeoff on movement on stage vs. framing of the camera), in the same way that a slidedeck without an associated talk is a different medium from a talk; you aren't really giving a talk to a conference anymore: you are giving an address to the world.
> If you make "Presentation Zen"/"Merlin Mann"/etc. low-text high-concept slides for your talk, how do you make sure your presentation is accessible after the fact?
The problem here is that you try to optimize for two opposite goals - high focus/low information density visuals for talking, and broad-spectrum/high information density visuals for reading afterwards. I'd say, you should go with first, and as for the latter, go and write an article. A video or slides with additional adnotations is an acceptable compromise.
The approach I usually go with is to initially create a full set of notes/slides that contain everything I'll want to cover, and then use a copy of that as the basis for my slides, which I then aggressively strip stuff out of. Then I've got the original notes still to refer to if needed while giving the talk, and to distribute.
Slides for during a presentation and slides/notes for reference without hearing the talk at the same time have two completely different needs to fill, and I don't think there's a one-piece solution to both. Especially since I feel the same way that, outside of a conference setting, referring to an audio/video recording of a presentation is much less efficient than a proper set of notes.
TeMPoraL's comment comes close to my response, but it's not quite what I'd say:
A slide deck is a visual aid. Its only purpose is to be a visual aid. If you want your audience to take a physical memento away with them, don't be lazy: make a separate, physical memento designed to be taken away with them. Your slide deck should be useless after the fact, just like a bullet shell casing is useless after you've fired the bullet.
When you make something, make it with a singular purpose in mind and make it fulfill that purpose well. IIRC, Presentation Zen covered this fairly well; in fact, I'm pretty sure Garr Reynolds is where I got this notion to begin with.
The biggest thing which trips me up is perspective and space. I'm uncomfortable in unfamiliar spaces and rooms, and unfamiliar perspectives within a room. This reduces my ability to perform, interact, and speak to people I don't know.
I'd suggest finding an empty stage somewhere and standing on it for 5 minutes. Just get comfortable. Get bored on stage. Try this a few times in different places with different lighting and furniture.
When you get to the conference, try to sneak onstage for 5 seconds, or even just peek through the curtain. Then visualize the space and try to become comfortable with it.
And try to channel your nervous energy into positive thinking.
1) Your public speaking will be more polished and compelling if you think of your speech as a story.
2) If you have some vital points to make, draw a picture on a notecard that will remind you of what you need to communicate. It doesn't matter if your picture is simple or complex, it just needs to get the message across to you. Then if you freeze up, you can look down and relax--no awkward pauses as you squint at several lines of notes.
I like to write speeches for clarity's sake and then practice speaking them over and over again leading up to the event. At least a day before I shut everything down and let the chips fall where they may. It usually results in me getting all my points across clearly without sounding like a robot.
Record yourself speaking. Once or twice through and you're going to have a very good idea of the parts you need to work on.
Also very helpful to have a "good run" recording saved that you can give a quick listen to if you're asked to do the presentation again. You don't have to find your words all over again.
On the subject of practising: doing a test run in front of real human beings is definitely ideal, but it's not always possible (especially if you've left preparation a little late...) - in the absence of that, it's really, really important that you practice the talk out loud. Even if it's to an empty room, doing it out loud will help ensure your pacing is right and that what you are saying makes sense.
I find that a talk only begins to really take shape the first time I rehearse it out loud, so the earlier in the process I do that the better - leaving plenty of time to adjust the talk to flow better or cover important missing pieces of information.
That's the key step to me. Rehearse and rehearse. When you're done, rehearse again.
Seriously, it is that important. At the beginning I used to think that _"nah, I've already done the slides, I know the subject inside out, and can always improvise. It sounds more natural"_. Wrong. This way you just sound unprepared.
As you practice, the flow gets tighter, you tell the same story with fewer words, paints a more colorful story. You start to see more clearly the holes in your thought process, and even jokes become more natural.
This is particularly important when doing a group talk. When you have multiple speakers, then you definitely need to practice, because it's not just about you anymore, and the quality of the entire talk will be determined by the weakest link.
For really important talks, five is the magic number. 5 full runs start to finish, non-stop, timed with a stopwatch. A good devil's advocate as audience helps (to take notes while you're presenting), but even that is not critical, provided you rehearse enough times.
To prepare for a presentation at Ignite Phoenix (which has its own particular set of rules) I used Screenflow to record myself speaking over the automated slide changes.
I wrote out my talk first, as an essay. I then read it and adjusted it based on how well it matched the automated slide changes and how easy/natural it was for me to say it.
I ended up with a video I could play over and over, listen to, and then recite from memory. I also printed my slides on index cards, along with some notes on what was supposed to be said for that slide, and had them with me since, as I it happened, when i gave my talk I needed some prompts.
I also practiced a lot. I hate it, but it helps, so there you go.
Ignite talks are a different sort of animal, where timing is very important (since you have no control over the slide change speed). But I found this approach of use for more casual presentations; it's still a good idea to have a good idea of how much time you will take and how well it all flows.
What I found was that you can start with something that would take 30 minutes to read to a crowd, edit out the cruft, the joke slides, and the amusing-but-unimportant factoids, and whittle you talk down to something punchy and concise.
Most talks are too long. Modulo topic and subject complexity I'd say they should be no more than 20 minutes.
Even if you're not doing an Ignite talk, plan as if you were and be mindful about how and why you make it longer. Don't give people an excuse to be checking Twitter or HN while you're speaking.
I always start in my head with the goal of the speech, then I sit down and make a quick outline.
I use the outline as a skeletal structure for my slide deck, otherwise I spend hours tweaking font size on the slides instead of understanding the overall goal of the talk. Once the outline is done, I add in a bit of muscle to the talk - one to two sentences per point to the outline. Then I go to the slides and add full-pages images for each slide, with short topic headlines for each part of the talk. Then I memorize the gist of the outline sentences and embed those mentally into the slides (making the slide deck a memory palace). When I get up on stage, the slides remind me of the storyline or topic I'm on, and I simply tell the story.
I have written out some talks before, but I only use that method to practice the talk as if I were speaking it out loud. I write with the same tone of voice that I speak in, and write it to get it out of my system. I don't look back to the write-up afterwards.
Learn your talk as a bullet list not as a script. Know each slide and what it means: which bullet points it matches. Assume that you will ad-lib on the day. Put as little of the real text as you can into the slides themselves. Never read anything: not even a quote.
Have a go at doing your slides backwards (or at least your main slides if you tend to have lots of slides per topic).
Be sure that you are making a clear message at the end. Then have a go midway (pulp fiction style). Your message should still be clear.
You are fully prepared when you can click through your slides while (mostly) looking forward.
On your #3, I prefer mjd's advice to "get to the point". They have already chosen to sit in your talk. Hopefully they know something about who you are, what the talk is supposed to be about, and there is little point in wasting the period when you are most likely to actually have their attention on reminding them of that.
(Note, mjd is the author of the conference presentation judo talk that I linked to above. TPC is The Perl Conference is the predecessor of OSCON. I don't know if he has continued to maintain the top speaker rating that he had, but he did so for a long time. He is worth listening to.)
I do think practicing is helpful and also having an outline of the major points you want to make in your head.
I use the 'memory palace' technique* for important talks so that I walk through a place I know in my head and each location is associated with a major point I need to make. This is important because I don't have the points on my slides (necessarily). The slides will often support the point, not make it.
1. Don't try to write jokes into the presentation. It won't be funny. It will only be funny if the joke comes to you on the spot.
2. Try to write the talk in terms of vague concepts instead of specific wording. WHen practicing, don't worry about how you're going to express a certain concept, just focus on understanding the concept inside and out. Every time I give a talk, the words come out a little bit differently each time. This keeps the talk from sounding robotic.
This is probably going to sound old fashioned but I would suggest looking up and attending the next meeting of your local Toastmasters (http://reports.toastmasters.org/findaclub/). They have a 10 speech series that will take you from terrified speaker to "competent communicator" in a well-structured non-threatening environment (and yes, I am a Toastmaster).
Practice at your local user group. We have several national known experts in the our field at the NTSSUG and they all do their dress rehearsals for larger conferences at meetings and regional conferences. And a majority do virtual meetings, video blogs and teach classes.
1. Write slides, more than you think you need, and try to have fewer than 1/4 of them include bullets.
2. Print the slides out, one sheet per slide.
3. Tape slides to wall in a grid.
4. Rehearse the talk to the grid of slides on the wall with a sharpie in your hand. Any time you have trouble with a slide or trouble transitioning from one slide to the next, note it (or fix the slide) on the paper slide.
5. Reorganize slides in Keynote (or whatever), print them out again, repeat.
This may just be a bit of a ritual for me, but one thing it definitely does is it creates a context where I don't feel silly rehearsing by myself.