When most investors think of due diligence, they think of sifting through years of financial statements and crunching numbers to better understand a company's potential. Not many people think about interrogating senior executives and employees, and asking the right questions to figure out whether they're telling the truth about the business model. In this episode, we speak with Phil 'Dick' Houston, a veteran CIA officer who's been called the human lie detector. Phil literally wrote the book on how to spot lies, and has been a long-time collaborator of entrepreneur and serial acquirer Brad Jacobs, helping him with due diligence on both senior hires and potential investments. In this episode, Phil explains his strategy for identifying deceptive behavior and how it can be applied to the corporate world. This transcript has been lightly edited for clarity.
Key insights from the pod:
What to look for when detecting deception — 05:43
Myths on eye contact — 12:33
Psychological buckets used to understand deception — 14:49
The ideal environment for lie detection — 22:39
Using verbal and physical cues to facilitate honesty in others — 27:47
An example of lie detection in real time — 34:53
How to ask the questions that most people fail to ask — 37:09
---
Tracy Alloway (00:10):
Hello and welcome to another episode of the Odd Lots podcast. I'm Tracy Alloway.
Joe Weisenthal (00:15):
And I'm Joe Weisenthal.
Tracy (00:16):
Joe, I'm kind of nervous about this episode.
Joe (00:19):
I am too. About 15 minutes ago, the same thought flashed through my mind. I too, am a little nervous about this episode.
Tracy (00:27):
Believe it or not, we do some preparation for our episodes, although maybe it doesn't always show. This particular episode, I haven't done a lot of reading into this topic, and I'm really hoping the guest doesn't notice. But I think my chances are very low because we are literally going to be speaking to someone whose whole career is about basically spotting lies.
Joe (00:52):
I mean, that's essential in the business world. If you really think about it, companies come on their calls, I think they don't typically lie, per se, I don't think executives lie that much, but, you know, they shade the truth. They highlight some things; they don't highlight the other things. I feel like it would be a pretty good skill to have in life to just generally — and especially in business — to just generally know whether someone is being honest with you or not.
Tracy (01:16):
Absolutely. It feels like a module they should have at Wharton if you're doing a business degree, an MBA or something like that.
Joe (01:21):
I don’t know. I've never been in an MBA program.
Tracy (01:24):
Well, the person we're going to be speaking to actually came to us via Brad Jacobs, you might remember him from the episode in which he was talking about his latest acquisition and the new part of his logistics empire. And he's written a new book and we were reading it. And one of the things that we thought was really interesting was there was a mention that he was working with a man called ‘Dick,’ or Phil Houston, an ex-CIA person who was basically a polygraph examiner for a long time.
And Brad and Phil were working together as part of the due diligence process for Brad's many, many, many serial acquisitions. So the idea was he might go in and buy a company and then Phil would interview some of the senior execs and basically try to see whether or not what they were saying about the company was true.
Joe (02:18):
Right. You know, speaking of nice skills to have, there's always that information asymmetry, whether you're just an investor or whether you're buying out a whole company. And the seller's always going to know more, [that’s] sort of a classic problem in economics. But if you knew the result of an answer was going to be 100% true, then that would be pretty great.
Tracy (02:37):
That would be nice. What’s that TV show with the girl who can spot a lie?
Joe (02:40):
Oh. Oh! Poker Face! It's kind of entertaining. It's fun.
Tracy (02:48):
It does seem like a useful skill.
Joe (02:48):
It would be good for poker too.
Tracy (02:49):
Yeah. By the way, another reason I'm slightly terrified of this episode is, I used to know an ex-CIA person who then went into another profession. I don't want to give too much away, but it was finance related. So I used to speak to him about financial things, and I remember I met him once for lunch after a particularly terrible morning where something really, really bad had happened in my professional life.
And I sat down, and I was like ‘Hi, how are you? How's it going? I can't wait to have lunch. We're going to talk about capital markets or whatever.’ And he immediately leaned forward and was like ‘Tracy, tell me what's wrong, something bad happened.’ And I asked him how he knew, and he was like, ‘Oh, you got this little crease in the middle of your forehead.’ And this wasn’t, you know, this was like a casual source meeting. This wasn't someone I knew really, really well. But he knew that I was upset.
Joe (03:40):
He had a good read on you.
Tracy (03:43):
Anyway, so on that note very, very pleased to say that we are going to be speaking to Phil Houston. He is a career CIA officer and also the author of the book, Spy The Lie: Former CIA Officers Teach You How To Detect Deception. So we're going to learn how to spot the BS. Phil, thank you so much for joining Odd Lots.
Phil Houston (04:03):
Tracy, thanks a ton for having me.
Tracy (04:05):
So just to begin with, how do you become a lie detection expert? What is this career path?
Phil (04:11):
When I joined the agency fairly early on, they had a polygraph in the agency and they had spotted the fact that I had developed a lot of information doing background investigations, and they thought that I would be a very good candidate for such a position.
Joe (04:34):
There's so much to get into. Tracy mentioned that we came across you talking to former guest Brad Jacobs, who you apparently have worked with a lot. How do you deploy lie detection skills in the corporate context?
Phil (04:47):
Well, we use them every single day. It's not just in a regimen of due diligence protocol, we're looking day in and day out. It's continuous evaluation, especially with new companies that Brad acquires, because we never know if we've missed something in the due diligence process. So we are always paying attention and Brad has been highly trained now by us and my team in the lie detection model that I developed when I was at the agency.
Tracy (05:24):
What kind of things are you looking out for specifically? So if Brad says he wants to buy a company, I'm sure he gets lots of information, printed information, as part of the due diligence process, and then he has interviews with senior executives. What type of questions would you be asking them?
Phil (05:43):
There are multiple areas of probing that we do. Number one is, first and foremost, are there any problems or circumstances associated with the acquisition that are not a matter of record, that no one has come forth and said ‘Hey, this is a problem or might be a problem going forward.’
And so, to gather that information, usually the first encounter is with a management meeting where we sit down in front of management and ask them questions. Now, granted the topics are fairly traditional, but the way that we word questions sometimes are very different. For example, on the really key issues, we rarely will ask a ‘Did you or do you?’ So for example, we won't say ‘Do you have any concerns in this area?’ We use what we call presumptive questions. And a presumptive question might sound simply like ‘What concerns do you have in this area?’
It presumes that there's potential for a concern. The beauty of that type of question is that, if a person is telling the truth, it's still a very easy and fair question to answer. However, if a person is concealing something and we ask it presumptively, it generally and often plants what we call a Mind Virus.
The Mind Virus is that thing that we've all been victims of. It's when somebody walks in the door in the morning and their coworker comes running up and says ‘Joe, the boss wants to see you.’ And you say ‘Well, what's that about?’ And they say ‘I don't know, but right away,’ and as Joe's walking down there, is he likely thinking more ‘Geez, it's going to be a great day. I'm going to get a bonus, I'm going to get a promotion,’ or does he immediately start thinking ‘What's wrong? Is there a problem?’ And if it is, then he's inventorying mentally what the problems might be and how he's going to deal with that. That Mind Virus triggers if someone's lying, it often triggers the exhibition of deceptive behaviors, both verbal and nonverbal behaviors.
Joe (08:10):
I already love this conversation so much, but let's continue to pursue — I imagine that when you're doing due diligence on the company, it's the unknown unknowns, it's the things that aren't written down, but you don't even know the right question to ask because it's not written down. But as you say, the presumptive approach.
Take the example further with a question like that. What are the types of responses that would trigger warning lights, either verbally or physically, this looks like a lie. What might be gleaned in the response to such a question?
Phil (08:45):
Sure. In working with Brad, we've had all kinds of — when people are lying — all kinds of crazy responses. We've had responses where we’re simply [making] the statement ‘Listen, we’re also going to come in and take a look at your accounting.’ And the person immediately says ‘Oh, there's no need to do that.’ And you'd think that they would realize how obvious that dodge is to spot, but at the moment, they're on the spot and they need it out.
nd so, they're not realizing the kinds of responses they give. They'll also say ‘Well, why do you need to go there? Or why are you asking that? How much do you need to know? Here's what I can tell you.’ And that's an interesting statement we hear and we listen for because if you think of the literal interpretation of that statement — ‘Here's what I can tell you” — it immediately in our world, conjures up, ‘What can't you tell us?’
Tracy (09:45)
Do people try to deflect? Having interviewed many people on this podcast, the most problematic response that we tend to get, hopefully not outright lies, at least not often, but you ask a question and someone will say ‘Oh, that's very, very interesting. Now let me talk about this entirely different point that has nothing to do with what you just asked.’
Phil (10:05):
That could be a product of their media training. But it also could be, if it's accompanied by other deceptive behaviors. And if we talk a little bit later about the model for detecting deception, clusters are an important component of that model. So more than just one deceptive behavior in response to a stimulus, meaning a question or a statement, that's what gets our attention.
And so, when they start that deflection, deceptive people know that they can't just sit there like the proverbial bump on a log and say nothing. They have to say something. And so, what they often do is try to lead us astray by talking about something they can comfortably talk about that might be interesting to you and perhaps even assuage what concerns they think you might have.
Tracy (11:00):
Are lies in the business world different, or do they exhibit different qualities to lies outside of the business world? So I guess in criminal activity or maybe in intelligence gathering, going back to your career at the CIA, are there differences in the way those lies manifest?
Phil (11:19):
The primary differences, Tracy, are the topics, but in reality, the behaviors remain the same. They're intrinsic to human nature. And we have a very specific list of behaviors that we are looking for. The biggest problem in spotting lies if people are not trained is when they see something odd that happens or something different or uncomfortable, their instincts may kick in, but they might not recognize what's going on, and they're only guessing at that point as to whether they've just heard a lie or not.
Joe (12:17):
Why don't we back up and talk about the framework or the model that you first developed at the CIA? And I imagine that over the years it's been refined, you have more examples and you get better and better at applying it. Why don't you bring us back to the core like underpinnings of your approach?
Phil (12:33):
The core is that we have codified the behaviors, so meaning we've picked the behaviors that we know through research and anecdotal evidence are the most reliable indicators. What we learned in the early days is that many of the traditional behaviors that we thought were good were not. And, for example, take eye contact, for example. Eye contact is not nearly as reliable as people would think it to be.
Joe (13:08):
Yeah, no, both me and Tracy have big relief on that.
Phil (13:10):
Yeah, eye contact can be very different, especially in different cultures in different regions. I'm here in New York City, if I were to walk down the street smiling and saying ‘Hi’ to people and so forth that wouldn't look very good, right? Or it probably wouldn't go over very well.
But in the little town where I live, the reverse would be true. If I didn't smile or if I didn't, nod or even verbally greet someone, they'll walk away with a different opinion. And so, the eye contact piece is just simply not as reliable as we need it to be. The anecdotal evidence in our world supports that as well as some major research studies, two in particular, one in the US and one in London strongly suggests that the deceptive person often has better eye contact than the truthful person does, highly likely because they're forcing it.
Tracy (14:09):
You know earlier I was using the term ‘BS’ and ‘lie ‘sort of interchangeably, but they're not exactly the same thing. And in the business world, one thing that tends to happen is you have people telling a good story about their business and the story, it might even be true. That might be the way [a] particularly ambitious executive sees things unfolding. I guess what I'm getting at is, how do you tell the difference between an outright lie — saying something fraudulent about their business — versus someone who's trying to sell the optimistic best-case scenario story?
Phil (14:49):
Okay, first of all, let me go back to the model itself. In the model we're monitoring both their verbal behavior, or what's commonly been referred to as body language. And we're also monitoring their nonverbal behavior. What I said earlier, we're monitoring their verbal behaviors, what they say the nonverbal is, obviously what they don't say.
And those both happen often simultaneously or in conjunction with a single response to a particular stimulus. And we are picking up on both of those. Now, what are we picking up? All of the behaviors that we use fit into five psychological buckets. The first bucket is the evasion bucket. These are behaviors such as failing to answer the question or qualifying the question. They're not giving you some or all of what you want from them.
The second one is this persuasion bucket. And this is where you were leading to, if I heard you correctly. Where someone's trying to tell you something positive or good because they can't tell you the truth because the truth has consequences to it. So they start using convincing statements — ‘Oh, we would never do that. We're a great company. We've been around forever. We've got the best team in the industry.’ And they're trying to convince you that you don't have to worry about, whatever it is or whatever topic that you surface to them.
The third behavior is aggression behavior, sometimes referred to as attack behavior. Sometimes it's very visible. Other times it's more nuanced. So, for example, you ask someone a question and they say, ‘Why are you asking that?’ And/or sometimes they say, ‘Why are you asking ME that? You guys always ask these things. I've gone to five different banks today and it seems like every single person has fixated on this issue.’ And so forth. They're trying to get you to back off.
The fourth behavior is what we call manipulation. This is where they're manipulating the circumstances of the interaction to their favor. A good example of that is when you ask them a question and they repeat your question. Now you might say, well, what value does that bring to them? Well, the value is that we think about 10 times faster than we talk. So, in that second or two to repeat the question could equate to 20 seconds of material to say, or strategy to pursue. And that's how they stay up with this.
Tracy (17:43):
I always thought that was good media training, right? If you need to buy yourself a little bit of time to like formulate the thoughts in your head, you repeat the question or you say something non-descript. “What a great question”.
Joe (17:48):
Or you praise the question.
Tracy (17:49)
Yeah, what a great point, Joe, thank you so much for bringing that up!
Phil (17:57):
A non-answer statement is another one in the manipulation bucket. And for the same purposes, it's buying time. The fifth and final bucket is the reaction bucket. And this is again, the body language. When people respond either as a result of the fight, flight, or freeze response, or there are a couple of other things that aren't caused necessarily by that response.
For example, something we call a verbal/nonverbal disconnect, where you ask someone a question and they're saying ‘No, I wouldn't do that.’ And at the same time, they're nodding their head yes, in response to the question. You see that kind of thing. But two big ones in the reaction bucket are when people have anchor point movements in response to something you ask them or see them.
I remember in the early days when I first started working in the investment world at a hedge fund. And I was sitting there with the hedge fund manager interviewing a management team. And I remember he asked someone a very direct question about what the street was saying was a problem. And the moment he started answering the question, he reached across the table and started making huge sweeping gestures on the table. And we call those grooming gestures or anchor point movements. And it can be as simple as a swivel in the chair, a leaning forward after answering, or in the midst, or in preparation for answering your question, leaning backwards.
Joe (19:35):
So in the case of the sweeping of the table with the hands, what is the respondent doing there implicitly? What makes that a tell?
Phil (19:45):
From a psychological standpoint what he's doing is cleaning up the surroundings. He, in his mind, is making life better at that moment. Because right now life's terrible because he doesn't have a great answer to the question. And that example, by the way, the company went under a few weeks later, for the exact issue that he was doing that sweeping.
Now, we don't rely on any one of these behaviors in response to the question. We're looking for a cluster, two or more deceptive indicators will tell us immediately that we have more work to do. More work simply means not that we're leaping to judgment, but that we're going to ask more questions, we're going to follow up, or we're going to talk to somebody else once they leave, or do research, whatever more work means. But the beauty of it is that we don't get snookered so to speak.
Tracy (20:43):
How often do you get false positives where you're interviewing someone and you think ‘Oh, this person is lying, or there are some red flags here to suggest that maybe he's not being 100% truthful.’ And then you go out, you do additional info gathering as you were just describing, and you find out actually, maybe he's just a weird guy, or socially awkward or something like that?
Joe (21:05):
He just likes to straighten up the table.
Phil (21:07):
Very meticulous individual. More often than not, [we’re] not likely to see, but if we're going to see one, it's likely going to be in a situation where we're dealing with multiple issues. So, for example, for Brad, for years we've done a ton of pre-employment screenings, especially for the senior executives.
And there may be more than one lie or more than one problem that they're worried about, but because of something called Psychological Set, it means that they have a fixation or fear on one particular area of deception, and they're not nearly as worried as much about the other areas. And so, most of their behavior comes out on that particular issue that they're worried about. And as a result, we may miss, which would be a false negative. But if our bias has the Pitchfork Effect, in other words we're thinking, ‘Oh, this guy's not telling the truth about anything,’ then we're starting to see false positives. And it happens.
Joe (22:14):
I know you mentioned that eye contact is not as robust as maybe people imagine, [the] popular belief. But I'm curious, in a world where more and more is done digitally, [via] Zoom, have you had to update your tactics at all? Or approach? Can you talk a little bit about lie detection in the pre- and post-Zoom era?
Phil (22:36):
Sure. We always, if we can, want to be doing the interaction in person and in particular when we're there, whether we can do it overtly or even covertly, we want to a head-to-toe observance of them as we're talking to them because the behavior can leak out anywhere.
For example, we know that the feet are always an anchor point, and if the fight or flight response kicks in, it will often go to the feet first, because that's our primary anchor point in life. And that's what gets us out of trouble. That's the escape mode and so forth. So, we want to be able to see those. In the Zoom, we lose a lot, we lose at least 50% of the reaction bucket. And so, we rely very heavily then, or more heavily even on the other four buckets, because those are all almost completely related to the verbal activity.
Tracy (23:35):
What about eye contact in Zoom? Because I will admit, I have a hard time focusing my eyes on Zoom calls, partially, because I don't want to look at the screen and see myself when I'm talking, it's just very awkward. So I sort of look off in the distance. Does that mean anything?
Phil (23:50):
No, Tracy. There are people that will tell you the opposite. But our research and again, the anecdotal evidence and a lot of your taxpayers’ money, has gone to invalidate some of that. When you're on Zoom, one of the biggest problems you have, both as the interviewer, if you're the person asking the questions or if it's just a conversation, one or the other is likely to experience what we call mental drift.
When you're sitting there, you have all kinds of distractors that are drawing your attention. Whether it's an email that pops up or a dialogue box that pops up, whatever the case may be. Or someone sticks their head in the door or a sound outside or whatever, [it’s] so easy to become distracted. And it happens a lot. And so, we have to be very careful with using eye contact as a lie behavior, so to speak.
Joe (24:49):
So in the CIA context, obviously a lot of people engaged in intelligence gathering, it's good to know if someone is being honest or not, but I also have to imagine that in the CIA context, there are a handful of people for whom the ability to lie is actually a valuable skill, including say, people who are going undercover in some situation, in which case deception is actually part of the job.
Can we learn to be better liars? Can we learn to identify our own leaks and tells and behaviors and sweeps and anchors and pivots so that we come off more trustworthy than we should be in some situations?
Phil (25:23):
Most of the time when people try to avoid lie detection, they're doing two things. Number one, is they try not to do things that everyone believes is obviously a deceptive behavior. The second is, they try to do things and focus on things that mask or carve out the other behaviors that sound better, so to speak.
And we know that because once we developed this methodology in the CIA, I was the principal developer of both the model and the training. We started training federal law enforcement and they loved the training. [Years ago we trained] Homeland Security [and] US Customs down at the Southern California border, we did a lot of training for them. And they did a study, a little mini study to see how well this worked. And it showed that their agents were much better skilled at spotting who has the contraband versus who doesn't.
Tracy (26:47):
I mentioned in the intro that we hadn't done much prep for this particular conversation, that was in fact a lie because I've done a little bit of prep. But one of the things I saw that I thought was really interesting was a bit where you were talking about cognitive dissonance and how that feeds into asking the right questions or interrogation techniques.
And you were basically suggesting that someone who is lying or misrepresenting the facts is dealing with cognitive dissonance because more likely than not, they still think of themselves as a good person, even though they've done or are doing a bad thing. And so, the idea is to allow them space to deal, or sort through that cognitive dissonance out loud to you, and provide a narrative that explains their behavior. Is that a useful technique in business as well? The idea of just providing people space to try to walk through their mindset at a particular time?
Phil (27:47):
If you really want them to be truthful with you, your demeanor is very important, number one. Number two, you need to present yourself as doing your job. Whatever your job, if you're an investment analyst, for example, if you simply, present the questions or the opening monologue, to the meeting saying ‘Hey, listen, let me just say upfront, we're really interested in you guys, but I really need to ask some direct questions. So please, I don't mean to offend you or anything of that nature.’
And then you lower their guard a little bit and it's amazing just simply opening up the conversation that way, and then presenting your question in a non-threatening manner, you can begin to get people to open up. There is another methodology that we use if and when we think that someone is lying and we have the green light from the client to use that methodology. And it is much different than what you would think of as interrogation. It's a persuasion technique that is enormously effective. And sometimes you get the whole story. Most of the time though, you get at least more than they'd given up at that particular point.
Joe (29:12):
Can you explain that a little bit further? What's this approach about?
Phil (29:14):
Sure. I once had a hat in my office. It said ‘If your lips are moving, you're lying.’ And in reality, if I think the person's lying, I want to go for a period of time where their lips are not moving. I want to be the one talking to them, and I'm going to be, not saying random things. I'm going to be using very selective influence techniques that will help me make it easier in their mind at that moment to fess up.
And we've used these techniques on spies, on criminals, on double agents. I know I have three confessions from double agents in my career at the agency. And to get a confession from a double agent is like solving the most major crime you can solve, so to speak, in our world.
Joe (30:13):
So what does that sound like? What are some of the things you say?
Tracy (30:16):
Yeah, do they actually admit it? Do they go ‘Oh, you got me, I'm a double agent.’
Joe (30:19):
What are the things you say to put them in that framework?
Phil (30:23):
That is a longer monologue, so to speak. If you think about the interview and interaction, it's a dialogue. And so, when we go in the monologue mode, we're in that direct elicitation mode. And it sounds very low key, so different than what you see in the movies where somebody starts the conversation by yelling, screaming, ‘You better tell me!’ and they list a series of consequences.
We do the reverse, we lower our voice, we start talking, we start telling them reasons why people have done this and that we understand those reasons. So there are a lot of different categories of things that we can say and do that make it easier over time to talk about it. And often it doesn't come out as the big confession. It often comes out as an admission about something related to it. And then we take that admission and it's like a thread or perhaps peeling the onion to use another metaphor. And you start just peeling or tugging and you begin to get the information.
Tracy (31:46):
That reminds me, actually, this is a little bit media, naval gaze-y, but Joe and I have talked about this before, but every once in a while we will get comments saying like ‘Oh, I wish you'd pushed back on this point, a little bit more in the interview or why didn't you press them on this particular point?’
And it feels like people have an idea in their heads that, if you just ask the right question enough times, eventually you'll get that ‘Gotcha!’ moment from the interviewee and they'll just throw their hands up and, and say ‘Ah, you're right, I'm completely wrong.’ But that hardly ever happens, in my experience. Even the best interviewers in the world, it's difficult to elicit that particular reaction because everyone is so on guard for that major admission. So it's interesting that you go for the smaller admissions that maybe are tangential to that and then try to pull those threads.
Phil (32:44):
In some cases yes, in some cases no. We’re reading them very closely as we're talking to them. If we think that they're prone to confessing the ‘Yes, I killed them, or yes we cooked the books, or yes we did this.’ Then we'll go for that. But if it's clear that they're being cagey and they're trying to dance around the issues and so forth, then we'll go for much smaller chunks of information and it's very effective.
And keep in mind the more questions you ask someone, a human being, what we know from the behavioralist is that every time a person lies their resistance doubles. So if you were keeping score and you were lying to me and I said, ‘Tracy, did you do it?’ and you say, ‘No,’ you get two points for your answer, but I only get one for my question. And then I ask it again in some other way. And now the score's four to two and then eight to three and so on and so forth.
Tracy (33:50):
So I'm sort of building up a wall?
Phil (33:52):
Yeah. Exactly, that's a great metaphor.
Joe (33:56):
Does everyone lie?
Phil (33:58):
The research says that in the western world, meaning in the western hemisphere, that the average person lies at least 10 times every day. Now, in the metrics for that research they include what we call the white lies. I like to call them the smart lies. If I go home from this trip and my wife has a new hairdo and my immediate thought is ‘Yikes!’ and she goes, ‘Hey honey, how do you like my hairdo?’ I'm definitely not going to say ‘Yikes!’ But I'm got to try to soften the blow, at a minimum. And if I can get away with, ‘Oh, that's nice’ she'll pick up on something. She worked for the agency as well.
Joe (34:49):
It's got to be not so easy.
Phil (34:50):
It’s a two-edged sword. You’ve got to be very, very careful.
Tracy (34:53):
Wow. Communication in your household must be both excellent and at times fraught.
But Phil, at times you have been described as the human polygraph, I think. Could we play a little game, a little lie-spotting?
Phil (35:08):
You can, but I will tell you upfront that what causes deceptive behaviors is the fear of detection. And so [in] parlor games, it's difficult, if not impossible to spot the deception because the person doesn't have any fear.
Tracy (35:25):
Because there’s nothing at stake?
Phil (35:26):
They know what the game is. And they know the path to win it.
Tracy (35:30):
Oh, maybe it's not a good idea then.
Joe (35:32):
I want to know what your lie was going to be, Tracy.
Phil (35:34):
I'll give it a shot as well.
Tracy (35:36):
Okay. All right. Three things. Number one, I went on vacation to Brazil and it is my favorite place in the world. Number two, I have a Corgi named Pablo, he just turned four. Number three, I once sang karaoke with Sir Ian McKellen.
Joe (35:52):
I think I know the answer.
Tracy (35:54):
I think you might know some of these, Joe.
Phil (35:58):
You're exhibiting almost zero deceptive behavior.
Tracy (36:03):
Excellent.
Phil (36:04):
If I were forced, I would've picked number one.
Tracy (36:06):
Oh, that's right!
Joe (36:07):
That's right, that's right! Was there something at all in the way she said that?
Phil (36:11):
Absolutely. She went to the persuasion bucket. [For] the others, she didn't. There was nothing persuasive about [the others].
Joe (36:19):
Wait, what did she say?
Phil (36:21):
She said, “I went on vacation to Brazil and that's my FAVORITE place in the whole world.”
Tracy (36:27):
Oh, interesting. I emphasized it. And I don't normally emphasize words.
Joe (36:31):
That's really good. That was really good because, I knew it because I had heard you talk about karaoke, and I know your dog, so I just knew it was Brazil, but I did not pick up on your embellishment of that fact.
Tracy (36:43):
I swear, this is such a frightening episode for people who talk for a living,
Joe (36:48):
Do polygraphs work? You mentioned the human polygraph, but where do we stand in the conventional wisdom on actual polygraphs?
Phil (36:54):
In the hands of a well-trained polygrapher? They're incredible.
Joe (36:59):
Oh, really?
Phil (37:00):
Yeah.
Tracy (37:01):
Maybe just to boil it all down, what is your top tip for spotting lies in the business world?
Phil (37:09):
At a minimum, pay attention to the evasion bucket. So, whether it's in a pre-employment interview you're doing, whether it's in an M&A situation, whether it's in an employee malfeasance situation, if they don't answer your question, and if they don't answer it entirely, don't think of yourself as a human lie detector. But take that as a catalyst for doing more work.
I'll give you a great example. I remember interviewing a very senior executive candidate at a Fortune 200 company. And they loved this guy. Everybody had interviewed him, they loved him. And when I interviewed him after the one-hour interview, I walked back out into the CEO's office. He said ‘What did you think?’ And I said, ‘Well, he's got a lot going for him, but the problem I'm concerned about is he's been either fired or separated involuntarily from his last five jobs.’
Tracy (38:13):
How did you know that?
Phil (38:14):
Because when I ask him the first time, I ask the normal questions, but then I ask a question that no one typically asks. So somebody comes in and says, ‘Oh, we had a restructuring and I was thinking about maybe changing jobs, so I put up my hand and said “Hey, I'll be glad to take one of those,”’ or whatever, and so, and, and people say, ‘Okay, that makes sense. We hear a lot of people that do that.’ But they fail to ask the simple question ‘Could you have stayed if you wanted to?’
Tracy (38:53):
Wow.
Joe (38:54):
I just have one more small question out of curiosity. Have you or [as] anyone approached you, or is there any work on AI machine learning approaches to lie detection? If you have all these videos of interrogations and texts and documents , is there any work being done on that approach?
Phil (39:07):
Joe, I'm so glad you asked. We are working on it furiously and we've already had some success and we're making great progress.
Tracy (39:17):
Exciting stuff. Alright. Phil Houston, thank you so much for coming on Odd Lots and telling us how to spy the lie. Appreciate it.
Phil (39:25):
Thank you guys. It's been fun.
Joe (39:26):
That was really fun. Thank you so much.
Tracy (39:28):
Yeah, it was a blast. Terrifying and fun.
Joe (39:44):
I was super impressed that he caught your lie. Especially because as [he] said, it makes sense that parlor games when the stakes aren't really there, that you're not got to have that same response, but that he caught that embellishment your first question and flagged it and nailed it.
Tracy (40:00):
It's definitely going to make me think about what words I'm emphasizing on this podcast.
Joe (40:04):
You didn't mean to add that point, right?
Tracy (40:06):
No.
Joe (40:07):
You didn't think, ‘Oh I'm going to give a little clue here on this one,’ right?
Tracy (40:10):
No, I did not.
Joe (40:11):
You just involuntarily offered up a clue that you did not intend to.
Tracy (40:15):
Yeah, I mean I tried to make them fairly neutral. Obviously, you have to say something. But I was trying to make each of them roughly similar to the others, but he absolutely nailed it.
I thought that entire conversation was really interesting, especially given that our day job is to interview people and the idea of maybe you don't seek out the big admission if people are cagey, or if they're in PR mode and they have a story that they want to tell, but maybe you try to nibble around the edges at smaller admissions.
Joe (40:47):
And the other thing too is, even if we're not in our jobs, most of our interviews aren't ‘gotcha!’ interviews. But I think that there's something to be gleaned about question structure and what is it that's going to elicit the most fruitful answer? And so even in his first example ‘Are you concerned about X?’ It seems like that's the question we might ask on a range of things. ‘Are you concerned about what AI is going to do to your business? Are you concerned about what the energy transition is going to do to your oil company?’ versus maybe we could frame it as ‘What are your biggest concerns about how AI will affect your business? What are your biggest concerns about the energy transition?’ Even when we're not trying to catch people in a lie, there's probably good lessons in here about higher quality questions, just in general.
Tracy (41:32):
Oh, absolutely. Also asking the questions that other people haven't asked yet. We do try to do that. Sometimes we ask the obvious questions and they're the most interesting ones, but no one has asked them before because they seem so obvious. Well anyway, shall we leave it there?
Joe (41:49):
Let’s leave it there.