Good morning, good afternoon, good evening, good whatever time it is, wherever you are in this amazing planet of ours. I am Malcolm Hawker. I am your host today on the CDO Matters podcast. I’m also the chief data officer of Profisee.

I am thrilled to be joined by our returning champion.

Our returning champion?

If this was Jeopardy mouth.

By by the way, if you were ever on Jeopardy, my money’s on you, my friend. This this is Mark’s this is Mark’s Stouse. Our our returning champion who would be amazing at Jeopardy, and I’m sure tell the truth, Mark. Do you annoy your family when when Jeopardy’s on?

Like, are you are you do you I actually watch it.

I hardly watch any TV.

Oh, you don’t watch it?

Okay.

Yeah. It’s kinda it’s you know, other than the odd historical drama, it’s kind of a waste of time.

Oh, okay. Okay. Okay. I I get it. I do I don’t watch a lot of TV, but every now and then, I love a good jeopardy.

And and I tend to annoy the heck out of my family as well because I’m the one screaming at the TV.

And One thing that is very true about my family and was a true about the family I grew up in as well, right, is that the dinner table is a is a debating club, and and you had better bring your best game.

It’s it’s interesting you should mention that. I actually talk about that in my upcoming book. I’m not trying to plug, but I am plugging. I actually tell a couple of stories about growing up, about how debate, including political debate, was a normal course of action at our family dinners. It was it was normal, and you’re exactly right. You better bring it because if you if you don’t, uncle Dave’s gonna have your lunch.

So Yeah.

And my my dad my dad was a a triple engineer CEO, and my mom was a theologian.

So Heart and mind.

Heart and mind. There you go.

Just imagine how those debates went. Right? I mean, it it was right.

But but I’ll tell you, though, there there’s something there’s something interesting there because when when you in my opinion, when you push science to, like, the boundary. Right? Like, I I fell down like a quantum mechanics rabbit hole a few years ago, and I just like, I started to consume all these books about string theory and quantum mechanics and relativity and all of this stuff. And when you and when you kind of, you know, push Schrodinger’s cat all the way to the edge, you you you reach this boundary where beyond it kinda starts to look look a lot like faith.

Yeah. I have to agree with you.

Right?

Like and I don’t I don’t see I don’t see actually a conflict between what most people call science and what most people call faith.

Right? The sheer volume I mean, what what defines science?

Well, right.

It’s not it’s not what we know. It’s that which we know we don’t know or that we don’t know that we don’t know. Right?

So, I mean, there’s a that’s the push. Right? We, as scientists, we push into that sphere of unknownness.

It’s it’s interesting you should say that. Like, there there are there obviously, there’s a lot of perspectives here. And, you know, some of the smartest people I’ve ever known all say the same thing. I’ll say what you just say. Right? Which there is a there is a place to have a conversation about the interplay and the interdependencies and between between faith and science, and that there is there is not necessarily this overt conflict that many of us would have us that many would have us believe that there is. So that’s not why we’re here to talk today, although that sounds like a heck of a conversation.

Yeah. That’s a cool conversation.

That that sounds to me, that like, that that that would be That’s two that’s two fingers of whiskey conversation.

Oh, that’s at least oh, yeah. That’s that’s at least two fingers of whiskey for sure. I may I may have Canadian whiskey, instead of Irish whiskey. I I do love my, my rye.

It’s been a while. Anyway, okay. We’re not gonna talk about, religion or philosophy or string theory, but we are gonna talk a little bit about data. And Mark’s gonna talk today about something that is very near and dear to his heart that he knows a lot about, which is a recent ruling out of a Delaware state court state. Yes. Delaware state. Okay.

Related court. Court. Yeah.

Okay. Okay. Chancery. That sounds that sounds fancy.

That’s the that’s the court that that does all business litigation in Delaware and that because of Delaware’s overall position in the corporate firmament Yeah.

Nationally and internationally, what it says kind of ripples throughout the country.

So Okay.

I get it. Kinda like, you know, emissions laws in California. Exactly.

That is a really great Alright.

Which I am going to shamelessly steal.

Take it take it, repurpose as as you wish. So I I think it is required at this point. I’m sorry. I’m gonna do it. Neither Mark nor I, to my knowledge, are lawyers. I know I’m not a lawyer. You’re not a lawyer.

No. Okay. No. I’ve been accused of being one by my lawyers.

But but I am not. No. Yeah. I mean, I I am a member of the eight you know, the American Bar Association, all that kind of stuff. But I do that as a, you know, for reasons. I’m I’m a I’m an expert witness.

So Oh, interesting. Oh, okay. That’s interesting. Okay.

Alright. But I’m not giving legal advice, but I can speak to it operationally. How’s that? Right? Yeah.

This is our safe harbor statement.

Council can’t Can’t.

This is our safe harbor statement, Mark. Just run with it. Right? Just just we we’re not we’re not lawyers and all else being equal.

Seek counsel from your lawyer if you’re looking for real legal advice. That’s not what we’re here to do. We are here to share what we know about this recent ruling related to data and the obligation of people who are in the business of care and feeding of data. So with that, Mark, why don’t you spend a few minutes just describing this ruling, describing, you know, what what it said, what it what it found, describe and describe it at a high level its implications, if if you don’t mind.

Yeah. So this is all about fiduciary duty, which historically has been a very rarefied idea, attached to primarily the board, of a company and the CEO.

That’s usually the limits of it.

And it also has historically required a very high bar, around proving bad intent, basically, to even have a shot at suing for breach of fiduciary duty.

So there was a case that was filed in twenty twenty two, by a group of McDonald’s, you know, like the Ambers shareholders. And it was all about the fact that there was this huge mess around sexual harassment and really bad behavior inside of McDonald’s.

And so that’s a that was identified as a breach of fiduciary duty, failure to for the the board and the CEO to exercise oversight.

And at the last minute, a young associate on the plaintiff side evidently added the name of the CHRO to the suit.

Didn’t have any kind of like legal basis to do so at that time.

But this guy was fairly egregious and and just felt like he deserved to be on the suit. I guess. So so then what happened was is keeping keeping the conversation focused, is that, in a great surprise landmark ruling coming out of this case, Delaware chance report vice chancellor, took the opportunity so like a big big time judge. Right?

Took the opportunity to reset the bar, reset a lot of things about fiduciary duty in in any company domiciled in Delaware, but now it’s kind of spreading all over the place. So the odds of everybody that’s hearing me being impacted have gone way up in the last two years. So in January of twenty twenty three, this judgment came out. And it’s the judgment basically says, you no longer have to prove malice or bad intent or anything like that.

Pure, just bad faith and redefining bad faith as negligence, unwillingness to oversee something that’s really important.

That’s enough. And also for the first time, we are now, spreading this responsibility to all corporate officers.

Okay?

So you gotta look at it from so the premise here was the judge said, modern business has gotten too so complicated that it’s there’s no way that the board and the CEO can discharge their fiduciary duty to shareholders without the active and integrated support of all officers, because that’s where they’re getting their information from.

What this means is that now all corporate officers have individual liability for failure to identify and remediate risks and bad performance in their area.

So it’s not about I wanna be very clear. This is not about you didn’t make your sales goal, you get sued. Okay? That’s not what this is.

This is I mean, putting it if I were to go back into a large company as like a chief marketing officer, I would sit down in an off-site with my senior team and I’d say, yeah, we’re gonna do a SWOT analysis, but we’re, you know, we’re gonna really focus on our team, And we’re gonna really focus on our risk factors.

And we’re gonna document those risk factors nine different ways from Sunday.

Right? And we’re gonna document what we think is gonna be necessary in terms of time, money, and effort to remediate those risks, and then we’re gonna file all that upstairs with the general counsel and the CFO.

And, hopefully, they’ll we’ll we’ll get what we need to mitigate those risks. But now they’re identified, and we can show that we did our part. Right?

So on the data side, right, what we’re seeing right now is that as shareholders and particularly activist shareholders, and there’s, you know, all kinds of flavors of activist shareholder, all kinds of sizes of them, you know, it’s not just Elliot and people like that. Right? So they have figured out that virtually every business in America today has a horrendous data situation.

If you look at kind of recent surveys and reports and and all that kind of stuff. Right? It’s somewhere, you know, low nineties to mid nineties percentile of data is crap inside of most corporations.

It’s either crap in terms of quality and accuracy or it’s crap in terms of accessibility, you know. In other words, it’s just, one way to kind of talk about it more conversationally is, if this data can this data support enterprise AI?

And the answer is usually not a chance. Right?

Shareholders see that as profound risk, competitively speaking, and lots of other ways. Right? And so the lawsuits are starting to accelerate fairly dramatically.

And it’s really focusing in this area.

So in so interesting. Let’s let’s put a few disclaimers out there. Well, first of all, I should have mentioned this at the at the top by the way. Mark is the CEO and founder of Proof Analytics dot ai. They’re a causal AI company. We’ll talk about that probably a little bit more because there’s a connection obviously here to what you do day in and day out. Just wanted to get that out there.

Delaware.

So this is a state ruling, but Delaware is important because so many companies, particularly big companies, are actually physically well, not physically, are legally, I should say, domiciled in Delaware. If you are operating all over the place, you are General Motors, you’re Ford, you’re Tesla, you’re whatever. I don’t know if those are actually domiciled in Delaware. But do you know, Mark, like, what what percent of of US corporations are are are Yeah.

In Delaware? Like the global one thousand or something like that. Right? Roughly two thirds are domiciled in Delaware. If you’re talking about PE backed companies, so, obviously, smaller for the most part, It’s ninety percent.

Wow. Okay.

Right?

Most VCs, for example, mandate that the company be domiciled in Delaware because the case law and everything else is so robust and, you know, they know what they’re getting into. Whereas, you know, you put a company domiciled in Wyoming, that’s a whole different deal.

Well, right. Yeah. I’ve got one domiciled in I’ve had one in Texas, and I have one now in Florida.

And companies choose have historically chosen Delaware, among other reasons, because of the perception of business and commerce friendly judiciary.

Right?

That that’s one of the that’s one of the I would say I would actually correct that slightly, and I would say it’s fairly shareholder friendly.

Shareholder friendly. Okay. Interesting. Yeah.

Since alright. Texas would business friendly.

Okay. Fair fair enough. Fair enough. Well, that that that chicken has come home to roost in the case of this ruling.

Because if you got shareholders saying, my shareholder value has been negatively impacted because you didn’t manage something properly, and I deserve compensation for that. That’s basically what the ruling is saying. McDonald’s, you screwed up. You knew that there were shenanigans going on.

You should have had more controls in place. You didn’t.

And that extends beyond just the c suite. Now that gets into officers.

Do we know what an officer means?

Yeah. Basically, it’s, you’re either named as one or you if you have that word in your title.

So Okay.

CRO officer.

Whatever. Right?

But also, there’s a lot of them that don’t they’re not they’re non obvious officers, but they are carried as one in the corporation.

And there are actually so one of the more interesting things about this situation is two things, actually.

One is we live in probably the most litigious society in the world, and yet nobody has challenged this ruling in two years, more than two years. No one. Right?

And so I was I was talking to some lawyers about this, and I said, might have any thoughts on this, like, why that would be? And they’re like, oh, that’s easy.

CEOs and boards love this because now all the liability is spread out across the company. All these officers share this liability.

Right? And I’m like, that’s yeah. Okay. That’s pretty interesting. And and indeed, you know, what happened ultimately with with McDonald’s was they pinned it all on the officer, the CHRO in this case, and let the board and the CEO off the hook.

And so Okay. So there’s there’s gonna be more and more and more of this. Right?

And then, you know, there’ll be probably be some tweaks to it, and there’ll be a rebalancing of it and all that kind of stuff. But the personal accountability piece is not going away.

Well, so you raised something really interesting. I mean, you you you started talking about data quality.

And it immediately hearkens me back to my halcyon days of my youth in the data and analytics space. And the first the first time I I ever landed my own consulting SOW to help a to help a company figure out their data, and I was hired to answer one question.

This was this is I won’t name any names, but it was a publicly traded company. Yep. I was hired by a company to answer the question of how many customers do we have.

I remember calling my mother that night, my mother and father, and giving them the the update. You know, this is the this is the son who’s flown the coupe and is now, you know, in a different country and doing these cool things. And I said, hey. It’s a big day.

I’ve I got basically my first sale and my you know, I because I’ve got my own consulting gig. And and I said, you know and they said, oh, well, that’s great. What are you gonna be doing? I said, well, I’m gonna help this company this customer this company understand how many customers they have.

They’re like, what?

Yeah. Yeah. I’m gonna help them understand how many customers. What do you mean they don’t know how many customers they have?

They don’t know how many customers they have. Right? And and, like, but that that idea in and of itself right? This may sound mundane and it may sound silly, but if you are a publicly traded company and you’re reporting revenue to the street Yep.

That is that is not necessarily backed by data or has fudgy data or the data is very the data quality to your point is very low, where it’s possible that that number isn’t accurate because your data quality is so low. What this ruling is suggesting and what you suggested previously is that if you didn’t have the right systems in place, the right processes in place, or that you were not doing a sufficient job to manage for data quality, that you could be held reliable reliable. You could be held liable for that in a court of law, at least in the state of Delaware.

Yeah. Actually, I was, it went settlement so fast that I never actually got to play in the game.

But, there was a Meaning as an expert witness?

You’d mentioned you were an expert. Okay.

Yeah. The the, but there was a case, California company, pretty pretty large ish, not like, you know, huge, but but pretty damn bit.

Who was a Delaware domiciled company and were sued for breach of fiduciary duty around one thing, and that was their CRM data quality.

Bum bum bum.

And and I was like, oh, they are so toast. Right?

Because I mean, if you’ve been in the And I have heck at all.

Right? You know, this is like a meme.

I mean Yeah.

Everyone knows that it’s screwed up. Right? And so the plaintiffs got the judge to authorize the install of a piece of software that that’s used to it’s pattern patterning type software used to find patterns fraud. It’s used a lot in Fiserv companies.

And, basically, two thirds, a little more than two thirds of fourteen years worth of CRM data came back as, I think that the polite term was engineered.

Like like Theranos engineered? Like, that level of because that gets to intent. Right? Like, that’s Yeah.

That’s that’s correct. So like a lot of them, this company was really well known in the sales area for doing spiffs, spiff programs.

Okay. Okay.

To mainly accelerate, you know, to incentivize deals to move faster to sale you know, sales guys to get with the program. And so and they because of the rules, they found a lot of evidence that deals were not declared when they were actually should have been declared.

And so Okay. They were triggering these extra payments illegitimately.

So now this company probably I don’t know this for a fact, obviously, but it’s probably sitting there going, okay. So, like, we have, like, a bunch of people that we probably need to go claw money back from, but is that realistic? And do we really wanna go do that? And, you know, and all that kind of stuff. But it was bad. And they they were they got a sealed deal, sealed settlement.

Right?

Which is why you haven’t heard about it.

Right. Exactly.

And it’s a huge amount of money.

So this is this is fascinating. So you’re a former CMO of a Fortune five hundred company.

You are an expert in AI data and AI. So you are the expert witnesses getting brought into these cases. Right? You’re you’re you’re getting to see, you know, these these the the kind of sausage being made on the fly, at least in the perspective of litigation around some of this stuff. You just gave one example of that.

You talked about CRM data, and, yeah, it is a meme. Right? It it is a meme, but there are so many things in this that we could be talking about here. The list is just it’s just so long. I mean, how many sales organizations have you worked with where there wasn’t tomfoolery going around from a forecasting and a pipeline perspective? Right?

Everybody Yeah.

That would be a the only answer would be zero.

Exactly. Like, gamesmanship around and I’m not I’m I’m not casting aspersion, and I’m not casting judgment. It’s just it just is what it is. Gamesmanship around sales pipelines and forecasts. Right? Salespeople don’t wanna or they or they sandbag or they, you know like, all of the gamesmanship around how deals are being reported and managed in a CRM.

What you just said now this was a little bit different. This is like paying sales spiffs on deals that hadn’t actually happened, potentially.

But I could easily foresee a situation where that if you as the CDO or e or the CMO, either, or the CIO, pick, go ahead, depending on the organization, didn’t have enough processes in place to make sure and didn’t have enough controls in place to make sure that that the inaccuracies in that world weren’t flowing into what was being reported on a ten k or ten q, that is a huge problem.

That’s a very huge problem. I think that for a longest so, like, if Delaware had had passed this ruling ten years ago, right, it would have been marginally enforceable.

Right.

Right? What is really happening today is it’s incredibly enforceable.

Right? AI in general, not just causal AI, other kinds of pattern match AI, right, is able to rapidly figure this out. Right? And so it’s a it’s a very, very interesting conundrum.

We have all, all of us, me, you. Right? We’ve all operated within what I call a zone of ambiguity for most of our lives.

That and that zone of ambiguity was actually quite large.

I define zone of ambiguity as where my representation may carry just as much weight as yours, kinda depends almost on my ability to present well or present better than you and all that kind of stuff.

That is shrinking substantially.

I mean, one of the things that AI, particularly causal AI, is going to be in the future is a giant bullshit detector on all kinds of things.

And and it’s just gonna be a lot harder to fade the heat.

Fade fade the heat or or or claim ignorance.

Yeah. Because everything will be known or knowable.

Right.

Okay. That’s that’s fascinating. So as you’re listening to this, if you if if if you find this interesting, I would invite you to check out next week’s or two weeks from now’s episode where I made reviewing a Barr Moses. Barr is the CEO of a company called Monte Carlo.

And Monte Carlo is what’s called data observability where what they’re focused on is, for lack of a better word, forensics around data and understanding some of these patterns that Mark was just talking about. Right? Like, patterns that would lead you to that would lead anybody to naturally conclude there’s fraud happening here or patterns that would lead you to conclude that there is some massive data quality issues that could otherwise have been stopped through better controls or better process or better governance.

So a lot of these issues are kind of come coming and coalescing together, but what you’re saying is there’s increasingly no more plausible deniability.

There’s no more sticking their hand in the sand.

And I’ll tell you this too. Right? I mean, the natural human brain functions as a correlative machine. Right?

Mhmm.

Paddle match. The problem is is that as causal becomes more and more accessible, then you have to worry about the delta.

So I actually did some work, like, mathematically, like, really kind of, like, got into it and had, you know, four different tools double check me and had a mathematician who’s, you know, wouldn’t want me to use his name, but he’s he’s, like, one of the very best in the world. He double checked it.

So you ask let’s say you pose five different questions, and you ask correlative math for an answer and you ask causal math for an answer, everybody, I think, that with a brain cell knows that there’s going to be some difference.

But it is actually really staggering how much difference.

It’s between three and three thousand and seven thousand basis points of difference.

That’s just the norm.

Right? So that’s thirty percent to seventy percent delta on the two on the two answers.

So the ability to rely on correlation outside of certain clearly indicated use cases and all that kind of stuff, right, is is just not there. Right?

And so you’re gonna have to really like, one of the one of the things that’s already being being asked in these in these depositions is they get really technical. They’re saying, how do you know?

What what technique are you using to understand this?

And if it’s the wrong one, then you’re kinda cooked.

Mean meaning okay. So the wrong one being, in this case, some idea of a correlation. Meaning, I’m a CDO, and I’m under I’m trying to understand buyer behavior, for example. I’m I’m trying to understand buyer behavior because I report expected future sales maybe to the street as a part as a part of some sort of filing exercise, safe harbor agreement you know, safe harbor notwithstanding, forward looking statements notwithstanding.

I’m still as a CEO or CFO, I’m still gonna probably say, you know, we are bullish on the next quarter.

Right.

And if in a court of law somebody asks you, okay. How did you conclude that you’re gonna be bullish?

Right? How did you conclude that you are optimistic about the sale of this product?

And you said, well, I I did a correlation between a and b, and we know a is gonna happen, and we rely on b for our product sales. That’s how I came up my data.

That court is liable to say, okay. Not good enough.

Yeah.

Absolutely not good enough because If they didn’t become true The world is not a one on one relationship.

Right?

Right. Okay.

And so it’s it’s a now what I do think is a legitimate answer is, you know, we used different kinds of correlative techniques. We used causal inference text techniques, and we did, like, staticistic techniques, right, or something like that. Right? I mean, so we were we we looked at it through a variety of different lenses, and we triangulate it, and that’s where we came out. Right?

I think that that right there is the kind of answer, excuse me, in the context of the ruling that would be almost impossible for a, a plaintiff lawyer to get past.

Okay.

So So kind of belt suspenders.

Right?

Right.

So before we get into, our our last discussion, which will be around what do we do, acknowledging there’s probably no one answer here. Before we get into the kind of the what would what would we do, what should CDOs do, I I do wanna just kind of focus on, you know, something we had talked about before we started recording, which is this idea that Delaware tends to be a bit of a bellwether for the rest of the country.

You know, this idea of California emissions. Right? And and what happens in California from an emissions perspective almost ends up happening everywhere else. You know, There are a lot of our listeners who are maybe listening from outside the United States, including Canada and Europe, who are like, well, wait a minute.

Why does one court in Delaware even matter? Right? Why does one state court in Delaware? Well, it matters because there’s a lot of companies, and it also matters because other other states will will start to adopt these rules over time.

Is there a possibility, though? Of course. I mean, I already know the answer to this, but let’s talk this is very briefly about, you know, legislation in and of itself. Right?

We we talked about Sarbanes Oxley and, you know, how that was a legislation, and it took a long time to kinda dribble out. Took a number of years to kind of really take foot take a foothold. Are we seeing the same thing here? Would we expect that maybe the legislative branch will take some action either to harden this or maybe undercut it?

What are you what are you hearing out there?

Yeah. I don’t think so. I I think that in the case of Delaware’s own legislature, you know, they’re actually trying to go a little bit the other way and and make Delaware because there are some big tech firms that are making noise about leaving Delaware for the Wyoming or the freedom of Texas or whatever. Yeah.

I think that that that the legislature, for that matter, legislatures in general, really prefer the courts to be the iron asses of policy. Right?

And so they’re gonna totally let the chance record in Delaware do that.

However, Texas has already started to cite the Delaware ruling in its cases.

Oklahoma, I mean, these are clearly redder than red states, right, are going this way.

California, New York, so those are two blue states. So I think that there’s a a a general view that this is good, not bad.

And and and I think that, you know, one of the things that’s gonna be a little bit awkward, I mean, let’s just kinda call it what it is, is when AI really takes hold in the decision making and vetting processes of corporations, we’re all gonna find out, you know, kinda like how wrong we were, right, on a whole bunch of stuff. Yep. And and that’s not and that’s not going to be there’s gonna be no respecter of persons, no respecter of rank. No I mean, nothing. Right? I mean, it’s totally different situation, but this is something LinkedIn is is grappling with.

You know, they seventy percent according to them of the content on the platform is bullshit.

Yeah. Yeah. AI generated.

Yeah. Yeah.

And and it’s just but even if it’s not, it’s still it’s Right.

Human baloney versus machine baloney. Yes. Yeah.

You know, you you just kinda sit there and say, there’s a and people kinda, like, at some level understand that that’s true. Right? And so it devalues the the platform, and and they’re kinda wondering how they can best fight that.

And I may have said something like, well, you’re just gonna have to put a little bullshit meter on people’s pages. Right?

That’s tough.

Of course, they they don’t really wanna do that, and I understand why they don’t wanna do that, but but it really does kinda come down to that.

So this this is another tangent and another rabbit hole that we could fall down. Right? Like, because, you know, getting back to the idea of proof, but a bump.

You and I have been in the business world and been in data, been in senior leadership roles. You know, we we can generally smell this stuff a mile away. Mhmm.

Smelling it and inferring it and deriving it is very different than proving it at scale from the perspective of the link of a LinkedIn. Right? Improving it and what I mean by scale, like, in an automated way. Right? So that’s a really hard problem to solve, you know, especially since there there are gradations.

You you and I have actually had some very interesting kind of back and forth on this in terms of, like, what is true and what is not true and some of the rigor that you apply to some of your posts, which is, like, incredibly laudable, by the way. Like, if guys, if you ever read anything from Mark Stuso on LinkedIn, just just know that it’s been vetted and that Mark actually has, like, a a structured process for vetting his own content and feeding it through multiple systems to to make sure that he’s not saying anything that’s that’s total crap.

So so I sleep I sleep so much better at night.

Right. Well, yeah. And and and you should be lauded for that, and and I I absolutely, you know, I’m I’m two thumbs up entirely on that. But I also but I also acknowledge that, you know, this this this notion of truth can be contextually bound, and there are objectives and objectives truths, and there are subjective truths at times.

And and that’s actually one reason why I think it’s really important to stay away from that word.

What? Truth?

Yeah.

Oh, yeah.

It’s loaded. Loaded.

Yeah. I keep it. I I speak about facts.

Yes. You know, but the argumentarian in me, you know, facts can change over time, and they always do.

Our our understanding of some facts changes over time. Right?

This this this this is this is why LinkedIn remains actually a really wonderful place to hang out and meet smart people like Mark Stuess, And and and I love it. Anyway, we we digress. Let’s get back to okay. I’m the CDO, and I know I got some problems. Because I don’t know any CEOs that doesn’t that don’t have a lot of problems when it comes to data quality. When it comes to data management, data governance, you name it.

This sounds a little scary to me. Sounds a little daunting to me. Maybe I just might wanna go open that winery instead or or go do the passion play instead. But let’s let’s assume that I’m passionate about data, and I wanna keep doing what I’m doing.

Is is this is this a a five alarm fire, a three alarm fire, a one alarm fire?

Is actually a five alarm fire for a lot of reasons.

I mean, I’m actually in a situation right now with a with a client where they’re being asked why they have so much data.

And so I was in the room and I said, well, can you give a little more color on why I mean, why do you care about data volume?

Well, it costs a lot to collect it, secure it, manage it, etcetera. And if you’re not getting active value, decision making value from the a lot of this data, then why do you have it? Why are we having to pay for it?

Right?

And I said, that’s actually a completely legitimate answer, which my client didn’t particularly like that I said that, but it’s the truth. So I believe in in giving that when whenever I can. Right? It helps in the end.

And in this case, it did. Right? So I think you have to if you’re a CDO, number one, you have to realize that you’re on a nexus of accountability with other officers. So for example, the CRM situation.

Yeah.

Right? There were three officers named.

And the more I kinda thought about it, the more this is sort of like a personalization of the people process technology triangle. Right?

So you had the CRO who was clearly the, you know, had people governance, you know, making sure that his sales guys didn’t go off the reservation.

You had the CIO, which probably of the three had the least liability, because all he did was own the the tool and the pipes. Right?

And then you had the CDAO who was radically exposed because there was no policy around data quality that was discoverable at all. Right?

And so and each one of them could have easily gotten zapped. Right? And I think you also have to realize this is not just about your financial personal financial liability, which would typically be covered by an insurance policy of some sort. Right? It’s about your reputational hit.

Yeah.

Right? Sure.

And and so, I mean, it’s it’s talking about putting your money where your mouth is. So I think I think that you have to say, and I think this is where the data is oil thing actually does work better than some people think.

So data is crude oil.

If you if you put it in your engine, it’s not gonna take you where you wanna go and it’s and it’s it’s gonna wreck your engine. Right? So that’s just. But if you refine it, right, into gasoline, benzene, and all the other distillates that come off that column, right, then you’ve got something of real value, which by the way, the refined value is what implicitly gives the raw material its value.

So I think that you have to say that until you pass your data through analytics, you don’t really know the value of what you have because you don’t really know how much better the decision making will be as a result of all of this working. Right? And if we stop and think about it for a second, using kind of we go way back into just basic multivariable regression. Okay? Gives you a multiplier number. So it happens in causal AI as well. Right?

That multiplier number is really important. You could argue, particularly for nonlinear contributions, right, that is the number. That’s the score. Right?

So if you are a data science team, if you’re a CDAO, you would want to know, for example, that because you and your team exist, that the business teams are making decisions that are I don’t know. I’m pulling this out of my ear here.

Five times, six times, seven times, eight times better than they were before they had access to data and analytics. Right? You wanna talk about justifying your position in the company? Holy crap, man. That’s it.

That’s it.

So that’s where I’d go with it. Right? I mean, I’m I’m trying to address it quickly and from a real high level. Right? But I think that if you go bottoms up in it, you’re not you’re gonna end up in the same place that every person ever ends up.

And if you go tops down, which is essentially leading with analytics, what sort of analytics what sort of business questions are there?

What kind of analytics do we need to answer those questions?

What kind what does that model look like? What’s the punch list of the data types needed to feed that model on a regular basis?

Right? That’s kind of the deconstruction that has to happen. And, again, I’m making it really conversational here.

But So I completely agree with you.

So to paraphrase, forget about any sort of bottoms up assessment here. Right? It’s not gonna happen. Any framework driven, bottom up driven, inventory driven assessment of, like, man, I gotta go look at all of my data and all of my processes and all of my systems and all of my people and all of these things, Never gonna happen. You’ll spend the next twenty years trying to get to the point of, like, reasonable understanding of of how all these things are connected. What Mark is saying that a reasonable first step at this addressing this five alarm fire and maybe taking it to a four or three alarm fire is to understand a few key things. One would be drivers of value from a data perspective, and another one would be major drivers of risk.

Right? Getting back to the kind of the risk component. You can quantify those things. Right? Risk from the from the perspective of, you know, accuracy of your financials.

Yeah.

Risk from the you name it.

Like, supply chain risk. You name it. Like, you’d be able to identify what some of the biggest drivers of risk are. Focus on the big picture things. Focus on the things that are going on to your any any of your c level reports and work backwards from there. That’s what you’re saying. Right, Mark?

Yeah. And I’ll I’ll add this other thing because it’s a personal human thing. It’s an organizational thing.

Right? So right now, largely because of some of the stuff that we’re talking about, right, data science teams, CDAO organizations, are being subordinated to finance at a greater and greater and greater rate. Okay?

And of course, the data scientists are pissed.

Well, there’s nothing creative about financials, but anyway, go on.

In the same in the same way that twenty five years ago, IT organizations were pissed when they were subordinated to finance. Right?

What’s going on here is they’re not saying that they know how to do your job better than you.

What they’re saying is, we’re gonna get you far more aligned with the business.

Right? Your business acumen is gonna improve. We’re gonna make sure that it improves so that the business gets the value that it needs. Right?

And so and also, I would just like to point this one little thing out.

If you’re in data science and you’re in a big company and you need data from fill in the blank organization, right, and you go to them, send them an email and say, hey, I need this data, The likelihood, unfortunately, is is that you’ll get brushed off or just told no or whatever. Right?

But when you work for finance, who in the world ever says no to a data request from finance?

Bingo.

Nobody.

Yep. Nobody.

Right? You can lose your job for telling finance no.

So that’s I would rejoice in it. In fact, if I were a CDAO, I’d be lobbying to become a part of the CFO organization.

It’s it’s funny you should mention that because that’s that was one thing I was waiting to actually say.

As as we wind up the podcast, I was gonna say, hey, you know, there’s good news here and there’s bad news. The bad news is there may be a five alarm fire. The the the good news here is that compliance and audit and regulatory risk or legal risk drives a lot of investment in business.

It always it always has. Right? It always has.

And that could end up actually being a good thing in your favor. And that’s exactly what Mark just said, which is, hey. Don’t be afraid of closer aligning yourself to the CFO, and don’t necessarily be afraid of using this as a lever.

Right? As as a lever.

Now you’re gonna You need to.

You you it is a lever. Right. You need to pull on the lever. Right?

Yep. Now don’t you know, fear can only take you so far. I would argue that if if the only value story you’re telling is I can help you not get sued, well, that’s that’s you’ll get the minimum required to not get sued. So, you know, I think we should aspire higher.

And Mark just talked about that, like, from the perspective of, like, understanding what some of your key value drivers are. These are all great things. Mark, you and I could probably, you know, talk for days. We do need to find a way to to do the two to four fingers of whiskey sometime, some way.

That would be great.

Can I just say one last thing?

Of course. You’re the you’re the guest. It’s it’s really short. Yeah. Go for it.

Guys, particularly those of you who came up in data science in the last, say, fifteen years, okay, where machine learning it was machine learning, machine learning, machine learning.

Machine learning is great.

Okay? It does a lot of things really well.

It is not data science.

Data science is much bigger than machine learning.

Right? So if you if you kind of say I mean, like in causal AI right now, you see some people trying to do causal ML, double ML, all this kind of stuff. Okay?

And it’s like they totally missed the key message in high school or college stat class that cause out or or correlation does not imply causality. Right?

You can’t get there from here.

Right?

And so you just need to kinda really be thinking about that, particularly in the context of business, where almost everything is a causal question.

Well right. And if you can do this, and you can do this well, and you can articulate to companies, I suspect you could pretty much write your own ticket from a wage perspective.

So Gotcha.

That that being said, thank you to everybody for listening to another episode of CDM Matters podcast. Hey. FYI, if we stick to the current schedule, my book, The Data Hero Playbook, will publish next week, June or July third, in which I talk actually a lot about these subjects. I don’t talk about the Delaware ruling per se, but I talk a lot about aligning yourself closer to the business. I talk a lot about accountability.

So if you are looking to understand how to be a better data leader to about how to use data as a lever of innovation of your organization, check out the Data Hero Playbook available Amazon dot com, July third next week. Mark Stouse, thank you. You are brilliant, my friend. Thanks for the insight.

We need to stay in touch. Thanks for coming out. To everybody else, we will see you on another episode of CDO Matters sometime very soon. Thanks all.

Bye for now.