{PODCAST} In-Ear Insights: Artificial Intelligence and Corporate Culture

{PODCAST} In-Ear Insights: Artificial Intelligence and Corporate Culture

In this week’s In-Ear Insights, Katie and Chris discuss how artificial intelligence impacts corporate culture and vice versa. Learn about the two different types of organizations, use cases for AI in corporate management, and the hidden danger of AI and institutional knowledge in your corporation.

[podcastsponsor]

Watch the video here:

{PODCAST} In-Ear Insights: Artificial Intelligence and Corporate Culture

Can’t see anything? Watch it on YouTube here.

Listen to the audio here:

Download the MP3 audio here.

Machine-Generated Transcript

What follows is an AI-generated transcript. The transcript may contain errors and is not a substitute for listening to the episode.

Christopher Penn 0:17

In this week’s In-Ear Insights, let’s talk about humans and machines, specifically, how corporate culture enables or disables artificial intelligence, we’ve talked in the past about how machines are taking tasks away, either freeing up more time for people to do higher value work, or in less progressive companies, resulting in the loss of physical bodies and seats, because you need fewer people to do the remaining tasks.

But a big part of the adoption of AI and how it’s implemented, as dictated by that culture is, is your company’s culture a progressive one, where you’re trying to uplift it up level your people? Or is it a regressive one, we’re like, yeah, we’re just gonna try and empty as many seats as possible and pay as little upair employees as little as possible.

So Katie, when you think about corporate culture and AI, what are the things that come to mind for you?

Katie Robbert 1:11

The first thing comes to mind for me is insecurity.

And so I say that, because I feel like if there’s insecurity about, here’s the value that we provide, here’s, you know, the whole will AI take my job.

And we’ve talked about some of the psychology of that, I feel like if any of that exists within the culture as a whole, it’s going to be really difficult to introduce artificial intelligence, or any kind of automation, quite honestly, into the culture, because having seen it firsthand, people will cling so tightly to the work that they do, and make it as complicated as possible, to make sure that they don’t lose that job.

And so they won’t share how they do it, they won’t share what they do, they will just sort of leave it in this black box, and be like, okay, but I’m the only one who knows how to do this thing.

And I think if there, if that exists in the culture, especially from the top down, then it’s going to be really difficult.

So that’s what I think of when I think of, you know, the corporate culture and AI, there needs to be an openness, there needs to be a curiosity, there needs to be a willingness to experiment and try things and fail.

If you don’t have those elements, then I think, you know, introducing AI, or really anything else is going to be difficult.

What do you think, Chris?

Christopher Penn 2:35

It’s interesting, because I hear something very pointed in your comments there that ultimately it’s about sort of the emotions of your workplace, right? Is it a fear based workplace? Or is it an optimistic based workplace where people are excited to come to work? People are eager to learn more people are enthused.

They’re, you know, Monday morning, is it treated like a like some kind of crime, a Monday morning is something that people actually look forward to, to within reason, obviously, everybody likes to not have the alarm clock go off every now and again.

But we’ve worked in places, and even the last agency we worked at had this interesting transition from one to the other.

As management change and ownership change, the culture changed with it.

Going from a more optimistic to a less optimistic workplace.

So I think the, the the key thing to think about there, and this is especially true for leadership, is what is sort of the emotional tone of the workplace, on average, like everybody has bad days and sufficient.

I’m not saying you’ll always be sunny and happy.

But on balance, do your people actually enjoy the work they do? Do they get to do a good job of it? And how are you fostering that? And even more important, I guess from a leadership perspective is what is your vision for your workplace in terms of how it makes people feel not just the goals, right? Goals are easy, make a million dollars in revenue, do this do that? But what is the experience like for your team members on the way to that goal, because we’ve all been in those car trips, where you know, it’s fun, and you’re just singing along to your 90 Spotify playlist? And you’re there before you know and then there’s the those car trips where the like, Are we there yet? Are we there yet? So You’re so boring.

Are we there yet? And those are two really good contrast to the kinds of emotional environments so we can work and, and AI is just a tool.

Right? It’s, it’s like spreadsheets, the spreadsheets work, whether your workplace is happy or not.

But obviously they provide more benefit when people actually want to do the work.

Katie Robbert 4:54

It’s interesting, and you know, I think you’re absolutely right Chris.

It is about emotion.

And so, you know, let’s say I came to you and I said, we’re going to introduce waffle irons, for lack of a better thing, we’re going to choose waffle irons into our job.

Your initial reaction to that should be my sort of temperature reading of the kind of culture that we’ve created, you know, as the person who’s introducing it.

So, you know, in my head, I’m like, Oh, my God, if I tell Chris, we’re introducing waffle irons, he’s going to be so excited.

But if in reality, I say, Hey, Chris, we’re going to introduce waffle irons, and your initial reaction is great, one more thing for me to do.

That’s a mismatch in terms of the culture.

But that also tells me that you as a team member, you’re not enthusiastic about this, you’re not seeing this as an opportunity.

You’re seeing this as a burden.

And then, you know, if I dig deeper, I’m like, Well, you know, what’s the problem with wallflowers? Like, I don’t know how to use waffle loans, you know, why do we even need them? There’s that layer of fear and insecurity.

And, you know, substitute waffle iron for AI.

And you start to see who on your team can be an advocate and ambassador, and help other people understand like, this is a good thing.

This is a cool, like new experiment, this could save us a lot of time, so that we can do the deeper work of building relationships and doing those insights versus the people were like, Oh, great, one more thing that I have to try to fit into my already overbooked schedule for the day.

And I think that those are the kinds of cues that you can be looking for in your culture.

Christopher Penn 6:39

But are we actually introducing wallflowers? Because that’d be pretty cool.

I mean, like it worked for Nike.

Katie Robbert 6:46

You know, we will take that offline, but I’m not opposed.

I love waffles.

So why not?

Christopher Penn 6:53

I know every every year, our Thanksgiving newsletter is Will It Waffle? Can you continue? Exactly.

One other thing that I think is an important part of the discussion about AI and corporate culture is how it’s being used within that culture.

So a couple of examples, standout.

One in 2018, Amazon introduced AI into its hiring process, and unfortunately, it stopped hiring women, because they trained the dataset entirely on on male developers.

So there’s clearly a case where the culture of the machines can’t be wrong, as is incorrect.

But then there’s also the other aspect, which is using AI to analyze your corporate culture with a healthy dose of of caution.

But it can be done if your workplace is one in which something like that would be effective.

So for example, every McDonald’s in the world transmits its data, essentially back to to the national organization.

So you have every time the driving monitor is blinking red, saying This car has been in line for six minutes, 30 seconds, where the fries, that data gets recorded.

Now, up until, you know, the last 10 years that data never really got analyzed, because there’s just so much of it.

But now it’s possible with machine learning tools to actually analyze it and say, you know, franchise, 1172 is consistently behind, and it’s most behind on Tuesdays.

And we think this is the reason why.

So how do you feel? And how do you think about the use of AI to analyze your corporation itself, your culture and your workplace?

Katie Robbert 8:42

I think that it’s completely appropriate, you know, within reason.

And so the first sort of, say, within reason, because you as the company need to be transparent with your employees about the data that’s being collected, and what will be analyzed.

So really start with the privacy.

And then I think it’s absolutely appropriate, look, analyzing, you know, your company is not a new idea.

It’s just the tools that have, you know, evolved over time.

So, you know, keeping a, you know, a checkbook or a check register, or some sort of financial statement, that is analyzing your company, a profit and loss statement that’s analyzing your company, you know, employee retention, that’s analyzing your company, whether you do that manually, or you introduce AI, is really sort of, I see them like, it doesn’t really matter how you do it as long as you do it.

So I think that one of the things that AI can do, now, much easier than trying to do it manually is try to look for tone and sentiment.

So for example, Chris, if we were going to analyze the culture at Trust Insights, what we probably do is, you know, do some scraping of conversations in our Slack channels.

So, um, you know, internally just to sort of see like, you know, is Katie always cranky and frustrated? Well, the answer is yes.

But that’s her permanent state.

That however, she’s not complaining about the work itself.

So therefore, you know, she must be, you know, fairly satisfied as an employee.

But if you look at like a larger organization, where they have intranets and Microsoft Teams, and SharePoint, and all of these other places where communication happens, trying to find those moments of people are really unhappy and disgruntled and not just unhappy, disgruntled, because to your point, people can have a bad day.

But why? What is the cause of it? Is it people? Or is it that the common thread is people are constantly complaining about Katie as the manager, she’s just telling me to do things, but not telling me why she’s constantly moving deadlines without giving me any kinds of heads up and then getting mad at me for not meeting a deadline that she didn’t communicate in the first place.

Those are the kinds of things that you can be looking for.

But you can also be looking for all the positive, like, hey, you know, I really wish we had more research and development, I would love to try this, Hey, I’ve been looking at this thing.

And so it should give you that insight.

It gives you insight into not just the efficiency, but just you know, sort of that tone and sentiment into how your employees are actually feeling.

Christopher Penn 11:21

So who’s in charge that? Who makes those decisions about what to calibrate the machinery on? Because everyone’s going to have a different answer, right? The CFO is going to say we should be maximizing profit, right? That is our job is to maximize profit, the head of HR is going to say we should be maximizing employee retention, because replacing employees is costly.

The CMO is saying we should be maximizing brand awareness because we need people to know about us.

How do you How does anyone when you start deploying the these different tools, make a balanced decision so that you’re not overweight? In any one particular area?

Katie Robbert 12:01

Well, just like humans, you know, it’s complicated, there is no one answer to the question.

And so that’s exactly its balance.

And so, you know, as the leadership of your company, you need to make sure that your your mission, your vision, and your values are clearly outlined.

So if you don’t have that, then you know, start there.

And then you can clearly outline what your business goals are.

So you know, Chris, in your example, to make a million dollars, okay, great, everybody is moving towards that same goal.

And everything else is just noise.

And so everything that you’re measuring should align with that goal.

And that goal should align with your mission, and vision and your values.

And so that’s just one singular goal.

But when you start to break it apart of it’s not just okay, we need to bring in revenue, it’s we need to keep employees long term who have institutional knowledge, you can build those customer relationships, who can keep people on board and upsell, in order to meet to make those that million dollars, we need to have good quality products, which means we have to have happy QA and engineering teams with good product managers who understand what the thing is supposed to do.

And so you start to break it down team by team, it’s those user stories all over again, of how does this personally affect me, as an employee, what do I bring to the table to get to the goal.

So I as the CMO, I as the, you know, head engineer, I as the QA person, I, as you know, pick a roll, then you focus it back on that one singular goal of the company, which is to make a million dollars, because every one is going to play a different role.

And so therefore, there is no one single answer of we just need to look at the final output of million dollars.

Did we make it or not? If we didn’t, you know, the culture is crap.

That’s not true.

It’s very, it’s complicated.

But AI can help you get to those answers faster, because it’s complicated.

Christopher Penn 14:11

Do you feel though that that’s more comp more, I guess, likely to see a balanced use in privately held companies because I think about Wall Street, for example, the singular mission of like 99% of publicly held companies is maximize earnings numbers for this quarter.

Like we need to hit our numbers this quarter, by hook or by crook fire as many people as you need to, to make the number.

And that has obviously created a lot of myopia for publicly held corporations, they have a tendency to do really stupid things because they have such short term goals.

You know, for example, r&d is typically badly sacrificed to a lot of companies because there’s no quarterly return for it.

And if you start using artificial intelligence to accelerate that process, I think If you run the risk of having, making bad decisions faster, easiest way, whereas a privately held company that doesn’t, that doesn’t report to the street, or to anybody except its private shareholders might be easier to align them on that, like, for example, we’re a privately held company, you and I are the primary shareholders.

So we can say, Yeah, you know, what we’re okay with, you know, not maximizing our profit margins, you know, we’re okay with like a 20% profit margin rather than 82% profit margin, because we have these much longer term goals, and we’re willing to sacrifice short term rewards for much bigger, longer term rewards.

When you think about the use of artificial intelligence, it optimizes for whatever objectives you give it, but the culture in those companies is very, very short term.

And so I, I feel like people would use it naturally to maximize those those short term goals and ignore or damage everything else in the process.

Katie Robbert 16:04

So that’s the point, you know, the point of this conversation is emerging.

And so you’re absolutely you answered the question.

You know, so in a privately held versus a publicly held, the use cases are going to be very different.

And the you know, amount of control that you have over it, is going to also going to be very different.

And so, you know, when we think about AI in corporate culture, there’s good reasons to use it, there’s not so great reasons to use it.

And I think, unfortunately, and this is completely anecdotal.

You know, I don’t have like solid proof that this is what’s happening.

But this is more speculation is in those publicly held companies, it’s less about culture and employee happiness, and it’s more about efficiency and making more money.

And for some, for some people, that’s completely fine.

They just want to show up, do the job, move on.

But what we’re seeing in this as part of the whole great resignation, is that that’s not enough for people, they want the culture, they want to feel valued, they don’t just want to feel like they’re a cog in the machine.

And introducing artificial intelligence does have the risk of making you feel that way.

Because artificial intelligence is very, it does repetitive tasks.

And it can take those repetitive tasks away from you, which could be great.

But if you don’t have you know, sort of that alt alternate career path of what else can I do? That it can, it can be very jarring to be like, Oh, my God, I just a I’d my way out of a job.

So what do I do now? And so I think in those publicly held companies where every single dollar matters, there is going to be less of an emphasis on the culture, like, completely be cut.

And we’ve, I mean, we’ve seen that we’ve seen the toxic cultures in a publicly held company.

You know, we’ve seen those examples firsthand.

I mean, they’re making all kinds of, you know, documentaries about these cultures now.

Christopher Penn 18:06

So in those places, then, should employees be resisting the implementation of AI?

Katie Robbert 18:17

It? That’s, that’s a complicated question.

I don’t feel like they should be resisting the implementation of AI.

I feel like they should be questioning leadership on their end goal.

I feel like they should demand more transparency, they should demand a longer term vision of where’s this going to take us if the short term goal is we just need to make more money right now? That is very short sighted.

And so I think that it’s not the tools the like, the tools kit, like you can, I could show up to your house, Chris, and be like, here’s a waffle iron.

Okay, great.

It’s just the tool until you do something with it until you implement it.

Like the waffle iron is not the enemy.

AI is not the enemy in this situation.

It’s not the villain, it’s not the bad guy.

It’s the people who are making the decision, the decisions around how it’s going to be used, what it’s going to be used for what the end goal is of using the tool and so those are the conversations you need to be having, you know, it’s not the tools fault that it you know, took your job, it’s the person who made the decision to program the tool in such a way to take your job

Christopher Penn 19:37

so in that case, when you have a corporation where the culture is very much just go out and make us a bunch of money.

For people who want to change a culture sounds like their, their their best option is just to find a different place to work because the leader ship, and the objectives they are optimizing for, will naturally find its way into the code because machine learning learns from the data you give it and all the data you give it like Amazon found that 2018, if all the data you give it goes one way, the machine is going to come up with models that also go the same way, it reinforces what’s already there.

So in companies where you have a, a toxic or negative culture, all the data that the machines will learn from, will be that and it will say, Let’s do more of that, because that’s the objective, it will never have the opportunity to bring in new data from outside saying maybe there’s a a better way to optimize, and still achieve those numbers.

So it sounds to me, again, given what we know about machine learning, and the state of it today, that AI and machine learning would actually make even more permanent, the existing culture that your company has, because it’s reinforced the data that’s already been given.

So whatever the culture is of a company, and I will make more of that.

Katie Robbert 21:03

And I think that that’s exactly it.

And so that is definitely the cautionary tale.

However, the other side of it is, if that’s the culture that you’ve created, and that’s the culture that AI will learn, you as the company will have a really hard time finding and retaining that human talent.

And so all you’ll have left is AI, and that may not be enough for you to meet your goals.

And so that’s when you have to come to those, you know, hard self reflective decisions of did we do this wrong? Why can’t we keep people and you know, it’s, you can’t escape bad culture.

You can’t, you can’t, you know, just AI your way out of having a toxic culture, because that’s what the AI will pick up.

And so you as a company will then be putting yourself at risk, because you will have nobody to run, to execute the tools to meet the goals.

Christopher Penn 22:02

That’s actually a very interesting challenge that if you think about it, because if, let’s say, you know, the company is doing badly, the board votes, okay, the current CEO and all management, they’re a bunch of idiots, the other we’re not making up numbers, everybody get out, which happens a fair amount of time.

And they bring in, they re staff it, they bring in a whole bunch of new folks, but the machinery that is still in place, and the data that’s still in place is naturally going to guide even the new management, you know, when it comes to objective optimization to the same things that were wrong from the previous management.

So in this case, AI is sort of like institutional knowledge codified into systems that make permanent that bad culture until you have enough data to start overwriting it.

So it sounds like one of the cautionary tales would be if you are changing management, you also may want to just flush your data and say, Okay, let’s let’s put all these machine learning systems on pause, because we don’t want it to learn from our past mistakes.

And I don’t know that people think about that when they implement these things.

Katie Robbert 23:02

I don’t think they do.

I think you’re absolutely right, Chris.

And I think that, you know, back to that, you know, will AI take my job, it opens up new job opportunities for humans, because that was that scenario you just described.

100% needs human intervention, to tell the AI to say, hey, these historic scenarios that we’ve given you, they’re wrong, they’re bad, we don’t want to repeat them.

I need to give you new data, but someone needs to look at all of that information to go, you know, keep toss, keep toss, keep toss, the AI can’t do that themselves, because you’re not giving it the information to make a decision.

So all it knows is what you’ve given it.

And so that becomes new roles, new positions for humans.

That’s really interesting,

Christopher Penn 23:52

because you could see a cottage industry popping up for big known systems like SAP, for example, the arriba procurement system, various HRIS systems where you could see a cottage industry of people who manufacture synthetic data for the systems as the training data set like hey, if you want the if you want apples growth, you know, for example, here’s the synthetic dataset modelled on Apple not with Apple apples actually data.

But you can then put this as the training starter into your HR system for hiring.

It’s like the the data equivalent of a sourdough starter.

If you find someone who’s got a really good one, you can start your system with that, as opposed to just rely on the existing data or relying on on making the data as you go and say okay, let’s get our systems trained out of the box with no good data.

For example, imagine in marketing, you have a CRM with lead scoring.

Instead of coming up with your own lead scoring algorithm.

You can say hey, I would like to buy and import a synthetic version of Hubspot lead scoring because they seem to be a successful company.

Let’s import their lead scoring as a starting, then we can fine tune it, that would be a heck of an interesting industry for managing your corporate culture and your corporate systems by taking known working models, and then just using them as your starters.

Katie Robbert 25:15

I mean, it’s, again, sort of back to the auditing your company is not a new idea.

It’s just the tools that have changed, you know, think about blueprints and templates and stencils, and all of these things, that you just copy a recipe to recreate something else.

And so it’s just a new tool to do the same kind of work.

And so, you know, building and maintaining a corporate culture, not a new idea, historically, it’s been done manually, like whose culture do I, you know, want to emulate? I really like the culture at Apple, for example.

So what is it that they’re doing that historically, it’s all been very manual, because that’s what was available.

And so you would have, you know, a team of people who would say, Well, they do, you know, pizza Mondays, and bagel Fridays, and karaoke Wednesdays, and they give people this amount of time off.

And you know, what is all the other things that they do? And you would say, let’s start to introduce that into our culture, and see how it works.

using AI to do that is really no different.

All you’re doing is saying, Give me the blueprint.

Let’s introduce it.

And so I think, again, sort of the introducing AI into corporate culture doesn’t need to be a scary threatening, you know, I don’t know why we’re doing this, it’s a matter of how are you going to do it, and what you’re going to use it for?

Christopher Penn 26:44

Yeah, I’m intrigued by the idea of best practice models, though, for different systems within a corporation, because that would be, again, it’s a jumpstart, it’s a way to get a company up and running sooner rather than later without having to, you know, if you have an HR is that supports payroll optimization, instead of having to manually decide how do we scale bonuses and things.

If you have a known best practice model to work from based on a company that you aspire to be? It might come up with some different answers that you might not consider.

And I think that’s a really interesting, and I think, very well discussed idea behind how we can use AI to generate these, these examples, and then build them into our companies.

Katie Robbert 27:32

Well, I mean, Chris, think about when you’re working on a new code and product project, you know, I’m, I’m fairly certain you’ll go to places like GitHub to see has anyone solved this problem already? Can I use their code as a template? To solve my problem, instead of starting from scratch and trying to figure it out all by yourself? It’s the exact same thing.

And so it’s just a matter of, does the template solve the problem that you’re trying to solve? And so that’s where you start is, what is the problem I’m trying to solve? And then can I find examples of where this has been successful, and introduce them into my company?

Christopher Penn 28:10

It’s so funny, you mentioned that because I read all these things about hiring for various types of engineering and stuff in AI and data science.

And there’s all these developer interviews and coder interviews where they people ask coders to solve problems.

And at no point, does anybody in that entire process acknowledge now what you do, the first step in the real world is you go to Stack Overflow finds an existing version, just copy and paste that into your code.

And you’re done with the problem.

You know, that’s reality.

And these developer interviews are interviewing for in a lot of cases, situations that are not reality.

And so you might actually end up making the wrong hire, because you’ve hired somebody who will attempt to reinvent the wheel, rather than go for the simplest solution possible, which is find some existing code adapted to your needs and build it.

And I think when we’re talking about the implementation of AI within a company, I think that’s a very real risk as well, if you’re the hiring process for how you bring AI and has to be as rigorous as the hiring process for bringing humans in.

If it’s not, I think you’ve got a uphill battle.

Katie Robbert 29:20

And that goes back to you know, the whole conversation about corporate culture, it’s what is the problem you’re trying to solve? And how does AI solve it? And so, you know, you may be saying, well, at a very small scale AI is going to help us with our reporting, you know, our mountain of reporting that we have to do.

Okay, great.

That’s a very specific use case.

And so you know, that one is easier to implement, because it’s very, you know, you have very clear boundaries around it, you know exactly what it’s supposed to do.

The output is very clear.

When you start to broaden the scope of why you think you want to introduce AI, that’s where it gets more complicated.

And that’s where, you know Surprise, surprise, you need to spend more time upfront with those business requirements of why are we doing it? Who needs to be involved? What is the process? What are the outcomes? How do we know if it was successful? You know, it’s, you know, again, if I just drive to your house, Chris and drop off and waffle on and you’re like, Great, this does not solve a single problem that I was having.

It’s the same with AI, you, you can just, you know, spend a lot of money to buy, quote, unquote, AI, which is really you’re buying a piece of software.

But if you don’t know how you’re going to use it, then you’ve just wasted a lot of money.

Christopher Penn 30:35

Yep.

All right.

So wrapping up.

At the end of the day, your your machines will reflect to you humans and the culture they set.

And if you don’t figure the humans out, you’re definitely going to just make things worse, worse with the machines.

If you’ve got some stories to share of your own, about the implementation of AI and how it was received in your corporate culture, pop on over to our free slack group go to trust insights.ai/analytics for marketers, where you and over 2300 other marketers are asking and answering each other’s questions every single day.

And wherever it is you watch or listen to the show.

If there’s a challenge you’d rather have it on, go to trust insights.ai/t AI podcast where you can find the show on most other networks.

Thanks for tuning in.

We’ll talk to you soon.

Take care


Need help with your marketing AI and analytics?

You might also enjoy:

Get unique data, analysis, and perspectives on analytics, insights, machine learning, marketing, and AI in the weekly Trust Insights newsletter, INBOX INSIGHTS. Subscribe now for free; new issues every Wednesday!

Click here to subscribe now »

Want to learn more about data, analytics, and insights? Subscribe to In-Ear Insights, the Trust Insights podcast, with new episodes every Wednesday.

Leave a Reply

Your email address will not be published. Required fields are marked *

Pin It on Pinterest

Share This