Making Sense on Stage
How to speak to your audience’s sense of logic
Before becoming a go-to stock image for philosophic quotes, Rodin’s “The Thinker” was originally part of another merrily named sculpture — “The Gates of Hell”. Those very same gates that Dante and Virgil pass through to take the scenic route through the nine concentric circles of Alighieri’s Divine Comedy.
Once they ‘leave all hope’ and pass through, they enter the vestibule, a place for the ‘uncommitted’—individuals who did not use their intellect to make a choice between ‘good’ and ‘evil’, and as a result were denied entry to Heaven and Hell. In essence, aside from a few colorful details involving nudity and wasps, it isn’t too different from being stuck waiting for a connecting flight — that, and the layover is for all eternity.
It‘s my belief that this ‘special place’ is reserved for most vague speakers.
It’s a place for the headliners who can’t decide what to focus on, because reasons. A slot for moderators of platitudes that end up saying everything — with much ado about nothing. An eternal stage for evangelists of tech that preach the gospel of ‘innovation’, ‘A.I.’,, ‘blockchain’, ‘design thinking’ or many of the 249 words found in the New Devil’s Dictionary. A disruptively innovative platform reserved for speakers who are yet to discover their commitment to specifics, their audience, or preparation. Metaphorically speaking of course.
In a more literal sense, I’d like to encourage those speakers to try a more behaviorally conscious approach. An approach to fight the ‘forgetting curve’ that would likely remove close to 75% of your talk over the next six days from memory. Or overnight, depending on how many bottles of craft beer, wine, or mead your audience managed to rack up.
This is the point-by-point guide I wish I had when I started preparing for my first cringe-inducing talks. Yet before I get to listing the ways how our faulty reasoning can derail meaning in a speech and what we can do about it, there are about 450 words of ‘brain-talk’ ahead. They will take us to the stem of the problem, and to be specific, above the brain stem — to explore the questions of why we rationalize — or even fail at making any goddamn sense, and hopefully lay a path to your audience’s logic.
Cortical Showdown: ‘Limbic system’ vs. the Neocortex
Regardless of the number of patents filed, conferences attended, talks given, papers or emails written—you’re still an animal at the core of your brain. An animal that can learn a barrage of engineering, medical, or business acronyms and stuff them as bullets in one slide or sentence, yet one that still sometimes craves a hug or a drink after presenting all that complexity in front of a few hundred confused faces.
“At his best, man is the noblest of all animals;” — Aristotle
A jargon-spouting beast that tends to rationalize why they needed that third cup of coffee at the speakers’ dinner, or why they should show that impressive-looking market number, or tweak the competitive matrix in their favor, or cherry pick that user statistic. But, hey, always-be-closin’, right?
The same beast that decides to invest in a founder with an unfair advantage because of a ‘fear of missing out’, or because that other ‘brand-name’ investor pitched in. An animal that still needs full 8 hours of sleep, and unless you’re fashionably fasting, an average of 2000–2500 calories a day. One that needs to feel safe at an airport or hotel, to belong, and to look good in front of other animals looking at you present, if Maslow’s theory of needs is of any guidance.
We’re heuristic decision-making machines that are very good at forgetting we’re animals, and whose lower brain regions, such as the amygdala and the hypothalamus, likely evolved with survival and not rationality as purpose. In fact, according but not limited to multiple Stanford studies, such as the ones in 1975, 1979, or the 1980, we’re remarkably good at holding on to our beliefs, even (or especially) if they’re proven factually untrue.
Luckily for us, it seems that alongside our tendency to rationalize away facts, we’ve also gained our ability to reason and admit that we’re wrong (although our anterior cingulate and insular cortices don’t make it easy). Those have arrived with the more recently-developed neocortex, which brings with it the ability to regulate emotions, to ‘self-actualize’, and to read this lengthy listicle on meaning.
Yet, whereas many of the above examples serve to highlight our tendency to feel first and think second, some of us, such as Aristotle, Locke, and Mill also have been able to put their minds to work on how to take charge of our thinking. Such people started by recognizing and explaining fallacies — those instances when there’s a problem with our reasoning. Those times, when the different parts of our brain are engaged in an MMA-style beatdown.
“First step in solving any problem is recognizing there is one.” — Will McAvoy
The bad news is that there are at least two different groups of fallacies — one formal and the other informal, with more than 200–300 of them out there. Also, some are in Latin and require you to wear a tweed jacket with elbow patches. The good news is that I prefer to make jackets out of lighter Italian wool, and will mostly focus on the core fallacies and a few informal ones, with plenty of quotes and examples outside the pages of dusty tomes, as well as a smidgen of calculus.
1. Equivocation
As a fallacy of ambiguity, equivocation happens when a key term or phrase has more than one meaning in an argument and is used in a vague way, such as the words ‘death’ and ‘data’ or the metaphorical use of ‘invention’ and ‘oil’ (see examples below), or any of the words listed at the beginning of this piece. In more colloquial terms, it could lead to ‘bullshit’ and phrases like ‘quantum healing’.
- Steve Jobs said that the best invention of life is death.
Weapons manufacturers are known as the merchants of death.
So, Raytheon, Northrop Grumman, and General Dynamics must have the best inventions. - Data is the new oil.
Uber uses machine learning to mine data.
Therefore, Travis Kalanick and Benchmark Capital are the new Daniel Plainviews (oil baron).
Metaphors and analogies, such as the ones in Alphabet’s Sideways Dictionary, are memorable ways to explain what you do as long as you ground them with facts and specific examples to make sure your audience understands what the hell you are talking about. Unless your goal is to achieve the opposite I’d suggest to seek examples of what not to do in Uber’s newsroom (like the one below) or the claptrap-culling FT column of Lucy Kellaway.
“I remember when I first used Uber as a passenger, even before joining the company, and drivers would tell me how much they loved the service. It was transforming their lives…we’ve underinvested in the driver experience and relationships with many drivers are frayed. We are now re-examining everything we do in order to rebuild that love.” – Rachel Holt, VP and Regional General Manager at Uber
What exactly is ‘love’, ‘transforming’, or ‘underinvested’? Is it hyperbole, an attempt to obfuscate meaning for journalists, deeply ingrained corporate speak, or something else? I guess in the words of the always-eloquent former U.S. President Bill Clinton — “It depends on what the meaning of the word ‘is’ is.”
Reader’s Digest version: Equivocation is about using ambiguous language. This can lead to a hall full of confused faces going through a little personal hell of their own. Jargon, acronyms, hyperboles are not a replacement for facts, examples, or meaningful metaphors.
To-try: Make a list of the most commonly (over)used vague terms in your field. Invite a colleague(s) to do it with you. Split the list into two or more parts among yourselves. Attend a conference or meet-up. While at the conference each of you would need to mark down the number of times each of the words has been used by a speaker(s). The one with the highest word count wins, and now has to do his next talk without using any of the most frequently used words.
2. Fallacy of Division
This fallacy happens partly due to our faulty inference, in other words, lazy guessing by relying on superficial cues as opposed to deeper analysis.
Y Combinator invests into ‘unicorns’ like Dropbox, AirBnB, and Stripe.
Our company has received investment from YC.
Therefore, our company is on track of becoming a ‘unicorn’.
Just because something is true for the whole does not mean it will be true for all or some of its parts. In other words, assuming that just because many of the companies in an investor’s portfolio ended up getting a certain valuation does not presume yours will.
“Those who avoid the sin of intellectual sloth could be called “engaged.” They are more alert, more intellectually active, less willing to be satisfied with superficially attractive answers, more skeptical about their intuitions.” — Daniel Kahneman
To-try: A way to upgrade our default inferential ability is to take a moment to calculate probabilities and payoffs (see examples below) before deciding to apply to an accelerator, pitching a popular client or investor, or buying passes to a conference the size of Web Summit, Mobile World Congress or Slush.
Warning: probability calculus ahead. Feel free to move on directly to the third point if numbers make you yawn more than stale keynotes.
Probabilities: Roulette vs. Startups
Out of all the 1,430 or so investments Y Combinator has made so far, 64 are valued at over $100 million, 10 are valued at over a $1 billion (see data), which leaves you with probabilities of around 4.5% and 0.7% of becoming one of the two respectively. That is if you get accepted in the first place. The odds of getting in are worse than getting into Harvard as one of the
YC partners pointed out to me during our Startup Grind interview. Harvard accepted 5.4% of applicants in 2016, YC accepted 1.9% in the same year.
So getting into YC and then achieving >$100 million status is not too different from betting on one individual and then two neighboring numbers in roulette, and winning twice in a row. The bets are called a ‘straight’ and a ‘split’, and the probabilities of winning them are 2,70% and 5.41% respectively (if you go by European roulette rules). Sure, overall someone has a chance of winning, but what are the chances that it’s going to be you?
Well, using the 1.9% 2016 YC acceptance rate, the 117 number of newly added companies in Jan 2017, and the current 4.5% probability of becoming valued at over $100 million, the probability (P) of you getting accepted into YC and then reaching a $1 bln>x>$100 mln valuation is around 0.079%.
P (of getting accepted into YC, and then becoming valued at $1 bln>x>$100M) = 1.9% x 64/(1430+117) = 1.9% x 4.14% = 0.079%
Your chances of becoming a ‘unicorn’? They’re something close to 0.012%.
3. Fallacy of Composition
Some of Y Combinator’s portfolio companies are valued at over $100 million.
YC must be very good at finding ‘10x’ companies.
Whereas the fallacy of division is about making the wrong assumption that the whole will be true for all or some of its parts, the fallacy of composition happens when one infers that something is true for the whole from the fact that it’s true for some of its parts.
Example of this being the fact that just because some of YC’s companies grow at a ‘10x’ rate, and on average raise at valuations over $10 million does not mean that all of its companies do. YC does not have a Magic 8-Ball to make better investment decisions than anybody else most of the time. They would probably be the first ones to admit it. What could be said with more certainty is that it had 237 company applications in 2005, of which 8 were accepted. In Jan of 2017, it had 7,200, of which 117 were accepted (see data).
With that kind of dealflow growth, and enough ‘spraying and praying’, some of the companies might indeed become worth at least a $100 million. In fact, over the past decade it has found 74 — out of 56,839 applications. That is a success rate of 0.13%. Then again, “In over 12 years of operation Y Combinator found 74 ‘10x’ companies from 56,839 applications.” — wouldn’t make for such an impressive headline.
Reader’s Digest version: Fallacy of composition is about assuming that because Chimpanzees share around 98.8 per cent of their DNA with humans, their great ape cousins — they can also pitch Sam Altman for $120k in funding for 7% of their company. Although Nim Chimpsky and Washloe each learned over a 100 signs of American Sign Language, how hard can it be to memorize an investor pitch, for a small reward.
To-try: Write an up to 140 character headline that could lead to this fallacy. Here’s two examples to get you started. “Fortune 500 Glass Ceiling Endures: Only 5% of CEOs are Women” and “Why are there so few male nurses?: Last year just 11.4% of registered nurses in the UK were male. Continued stereotyping is partly to blame.” — the two make for attention-grabbing headlines, yet because of our “lazy-guessing” heuristics we might conclude that both professions suffer from systemic discrimination, purely based on a single number that’s often presented out of context. More males end up committing suicides, and in prison, does it mean that they are being discriminated against?
4. Hasty Generalization
“You think you have no moves? How about taking your company public with $2 million in trailing revenue and 340 employees, with a plan to do $75 million in revenue the next year? I made that move. I made it in 2001, widely regarded as the worst time ever for a technology company to go public. I made it with six weeks of cash left. There is always a move.”– Ben Horowitz, The Hard Thing About Hard Things
Just because Mr. Horowitz made it does not mean you, I, or any other entrepreneur who read his book would. The same fallacy has been used to conclude that smoking isn’t that bad for you — “My father’s been smoking four packs of Malboros a day since he was eight. He is still alive.” Good for him, yet what is the total number of fathers/technology companies that died as opposed to the ones that survived?
Well, it seems that only 48% of dot-com era companies survived by 2004, which is only slightly better than betting on black in American roulette. Seemingly the data set of 1,018 companies also shows that about 5% of founders who knew or gained a personal intro to a venture capitalist received funding. Companies like ‘Boo.com’, which burned through $135 million in 18 months and the subject of one of the first ‘postmortems’ published online, made up the majority that did not have a lot moves left when the venture funding dried up. Then again, Mr. Horowitz has been publicly quite open about his on-going struggle with statistics.
Reader’s Digest version: Hasty generalization, also known as ‘anecdata’, is about basing your decisions and statements on anecdotal or insufficient evidence.
Warning: more probability calculus ahead. Fast-forward to the fifth point if you already know everything about the ‘37% rule’, and heard all of Jack Ma’s stories.
To-try: Jack Ma was rejected from Harvard 10 times, and now he is worth over $45.4 billion. The man doesn’t know when to quit, and partly because of it tends to inspire others with his own struggle against all odds, but what were these ‘odds’, at least in the case of Harvard applications?
The acceptance rate at the university in 1988 (Jack was 24 at the time) was around 14.6% out of 14,430 applicants. Provided one would apply ten times, and assume that the 14.6% chance stayed the same for each shot — what would be the overall probability of getting accepted? This is where you can try to do it on your own — or, using a binomial probability function, we can estimate that it was around 35.28%.
P (Jack getting accepted with 10 attempts) = 10 x (14.6%)¹ x (100%−14.6%)⁹ = 35.28%
How many attempts would we have needed to optimize our chances of getting in? The answer is 6 attempts to get the chances close to 39.8% (more can be found on the ’37% rule’ in Brian Christian’s post on ‘optimal stopping’).
However, here’s where it gets real-world-complicated. The admission rates have been systematically going down for the past thirty years, to a measly 5.2% in 2017. At the same time, during the past twenty years or so the number of Chinese enrolled in American colleges and universities has grown about sixfold, from 54,466 in 1999 to 350,755 in 2016 (see data). Moreover, the acceptance rate for international applicants at Harvard and MIT in 2012 was around 3.3%, with about 5.9% and 8.9% for everyone else respectively.
To take a step further and spare you three days worth of research and math, out of those 6,315 international Harvard-hopefuls, about a 1,000, like Jack, likely were Chinese nationals, with only a little over a dozen being admitted. That, is an acceptance rate of 1.5%, give or take. With those ‘odds’ per shot you’d need 66 attempts to maximize for 37% overall. In the words of Ms. Kimberly Wilkins — “ain’t nobody got time for that!”.
5. False dichotomy
It’s also known as ‘all-or-nothing’ or ‘false dilemma’ fallacy. The fallacy is committed in user and job interviews, by speakers waxing poetic about their success stories, or group coaching sessions when an audience is being given an illusion of choice.
“I think that the minute that you have a backup plan, you’ve admitted that you’re not going to succeed,”– Elizabeth Holmes, Theranos
The fallacy is about presenting two options as if they are mutually exclusive and all there is.
6. Complex Question Fallacy
This fallacy is committed when a question is asked that suggests its own answer. This is often achieved by an implicit assumption that has been tucked away somewhere between the lines. Some do it consciously, yet more often than not, it’s a result of habit and bias. Luckily habits can be broken, and we can become aware of our biases. When a complex question features controversial assumptions in the premise, it’s known as a loaded question. The fallacy is very much context dependent.
How many of you are tired of being ripped off by the banks?
Well, what if I told you that we have a solution that will put an end to that, would you be interested?
The first questions implies that we all are being ripped off, and it involves a false dichotomy of a yes or no. The second question assumes that the solution has been proven to work, and once again creates a false dichotomy of two options. The natural habitat of these sort of questions is in pitch competitions, marketing copy, and founder-investor ‘questions and answers’ (Q&A) sessions.
In fact, back in November 2016, four researchers from Columbia Business School, Columbia University, and Wharton published a paper where they analyzed 6 years worth of Q&A transcripts from TechCrunch Disrupt, that’s 1,857 questions asked by investors from a total of 189 companies. They found that 67% of the questions asked from male founders were ‘promotion-oriented’, while 66% of those asked from female founders were ‘prevention-oriented’. Those founders who mostly got ‘prevention’ questions went on to raise seven times less funds than the ones who were mostly asked ‘promotion’ questions through 2017. Here’s a few examples from the study, with a few likely assumptions.
- “How much of this are you actually doing in-house?”
Implicit assumption: that some of it’s not being done in-house. - “Do you feel that the infrastructure is mature enough that, for something as mission critical as security, you can build a whole system on it at this stage?” — this particular fallacy is discussed in the 7th point below.
Main implicit assumption: the infrastructure has to be mature enough for building a secure system on it at this stage. - “Is it a defensible business wherein other people can’t come into the space to take [a] share?”
Implicit assumption: this business shouldn’t have competition.
7. Double-barreled question
Do you see that there is a ‘product/market fit’/potential for this and that we should be seeking further investment?
A question that asks more than one thing, yet in reality only allows for one answer, or makes it unclear what exactly is being asked or measured. This one is common in ‘mentoring’ sessions when a founder might just really be asking for a ‘pep’ talk or cash. Yeah, we know, building a company is hard, and you are tired of ‘bootstrapping’ on instant noodles and energy drinks. It’s often rationalized as ‘character-building’.
Anthony Robbins is quoted in saying that “Quality questions create a quality life.” He, as well as some of the other ‘life’ coaches, and ‘motivational’ speakers are also masters of the double-barreled question, hell, some need a ‘semi-automatic’ category of their own. Let’s take the following example from one of Mr. Robbins’ sessions.
How many of you’d like your company focused on outcomes, results, and not activities? Say ‘aye’!
This question is so loaded with points, that I’ll respond to the in-built false dichotomy by invoking the words of Thor — “I say thee nay!”.
This piece, in its current form, wouldn’t be possible without the bare-knuckle editing and ‘guidance’ of Jen Geacone-Cruz. An alternate version of this article was published in the 11th issue of CoFounder Magazine.
