The P&L Case For Autonomous AI Agents

We draw a hard line between AI that chats and AI that works, then follow the money to show why autonomous digital employees are rewriting competitive advantage. We break down ROI, productivity multipliers, case studies, and the governance guardrails that separate winners from pilot purgatory. 
• Defining three tiers: chatbot, assistant, autonomous digital employee 
• Real example of an agent completing a refund workflow across systems 
• Baseline ROI and payback periods tied to P&L impact 
• Fully loaded labor cost versus digital employee operating cost 
• The inverse cost curve and why coordination kills scaling 
• Productivity multipliers in support, sales, analysis, and content 
• Hybrid teams where agents handle volume and humans handle exceptions 
• Case studies: Klarna outcomes, sales lifts, small business savings 
• Failure modes: shadow AI, pilot purgatory, mis-scoped “AI strategy” 
• Governance essentials: permissions, limits, approval gates, timeouts 
• Risks: selection power and cosmetic alignment that looks compliant 
• The recession stress test and compounding cost efficiency advantage 

Nonprofits and Businesses plan to automate at least 30% of all processes in 2026.  What is your plan? Who will be leading this effort?

Transcript

The Competitive Reality Check

SPEAKER_01 0:00

So if you are still typing prompts into Chat GPT and you're calling your business an AI-powered company, I mean we need to ask you a very uncomfortable question right up front. Are you actually competing or are you just sort of narrating your own obsolescence?

SPEAKER_00 0:14

Aaron Powell It's a harsh reality check, honestly. But we have to start there. Right. Because when you look at the enterprise data and uh the financial postmortems coming out right now, there is just this massive fracture in the market. There's a really clear divide between companies that are playing with AI as like a fun office toy and companies that are using it to fundamentally replace their workforce.

SPEAKER_01 0:39

Exactly. And that is exactly what our deep dive is about today. We are looking at this entire stack of research purely through the lens of your profit and loss statement.

SPEAKER_00 0:47

Yeah, no technical jargon today.

SPEAKER_01 0:49

Zero. We're focusing purely on the business strategy.

Chatbots Assistants Digital Employees Defined

SPEAKER_01 0:52

But to do that, you know, we have to cleanly define what we are even talking about because the terminology out there is just a total mess.

SPEAKER_00 0:59

Oh, it's terrible. Everyone is just slapping the word AI onto their product. Aaron Powell Right.

SPEAKER_01 1:03

So let's separate the modern business battlefield into uh three distinct tiers. First, you have your basic chatbot.

SPEAKER_00 1:11

Right, which is purely reactive.

SPEAKER_01 1:13

Yeah. It answers a single question when asked. It's basically just a dynamic FAQ page. You know, you ask what are your business hours, and it says, we are open until five. End of interaction.

SPEAKER_00 1:24

Exactly. Then step two is an AI assistant. And this is a tool that helps a human complete a task a little bit faster.

SPEAKER_01 1:32

Okay, like what?

SPEAKER_00 1:33

Like it might summarize a long meeting transcript or you know, help draft an email response, but fundamentally the human is still driving the car.

SPEAKER_01 1:41

Right. The human still has to initiate the task, review the work, and hit send.

SPEAKER_00 1:45

Yes. But the third tier, and this is where the entire market is shifting, is the autonomous agentic digital employee.

SPEAKER_01 1:53

And that's the wake-up call for this whole discussion.

SPEAKER_00 1:55

It really is. Because this is completely different from a chatbot or an assistant. A digital employee owns the entire job from start to finish. It runs 24 hours a day. It makes decisions based on complex rules. It takes action across multiple systems, and it hands off to humans only when absolutely necessary.

SPEAKER_01 2:13

And then it reports back on the outcome.

SPEAKER_00 2:15

Exactly. It is a worker, it is not a tool.

SPEAKER_01 2:18

Wait, hold on though. The word agentic gets thrown around a lot. What does that actually look like in practice?

SPEAKER_00 2:23

Yeah, it's a fair question.

SPEAKER_01 2:25

Because I mean, most bots I interact with just send me a link to an article when I have a problem. How is a digital employee actually doing the job?

SPEAKER_00 2:34

Well, that is the critical distinction because an agentic digital employee has permissions tied into your actual company software. It doesn't just talk, it acts.

SPEAKER_01 2:44

Okay, give me an example.

SPEAKER_00 2:45

Imagine a customer email saying their package didn't arrive and they want a refund.

SPEAKER_01 2:48

Sure.

SPEAKER_00 2:49

A chatbot says, here's our refund policy, but a digital employee reads the email, logs into your e-commerce backend to verify the shipping status, sees that the package is lost, logs into your billing platform, issues the refund, updates the CRM, and then emails the customer a personalized apology with the receipt.

SPEAKER_01 3:06

Wow. So it's executing the actual digital mouse clicks and keyboard strokes that a human worker would.

SPEAKER_00 3:14

Exactly.

SPEAKER_01 3:15

That is a staggering difference. I mean, it's the difference between buying your team a slightly faster typewriter and hiring an entire department of typists who never sleep, never take a coffee break, and you know, never make a typo.

SPEAKER_00 3:29

That's a great analogy. And this distinction is no longer some academic debate for technologists.

SPEAKER_01 3:34

Not at all.

SPEAKER_00 3:35

If you think your business is safe because your marketing manager bought a $20 chat subscription, I mean you are fundamentally misunderstanding the competitive threat coming for your market share.

SPEAKER_01 3:45

Aaron Powell And that brings us to the raw financial reality of

ROI Payback And Fully Loaded Labor Costs

SPEAKER_01 3:48

segment one. Because once you understand what a digital employee actually does, you have to look at the PL impact.

SPEAKER_00 3:53

Trevor Burrus, Jr. Right, the math that should wake every owner up.

SPEAKER_01 3:56

Exactly. Walk us through the baseline numbers.

SPEAKER_00 3:58

So across the board, the standard return on investment for deploying these autonomous digital workers is $3.50 to $3.70 returned for every single dollar invested.

SPEAKER_01 4:09

Wait, almost a four to one return just as the baseline standard?

SPEAKER_00 4:13

That is just the baseline. The top-tier operators, the organizations that are aggressively restructuring their workflows around this, they're pulling an incredible 10 to 1 return.

SPEAKER_01 4:25

10 to 1. That's unbelievable.

SPEAKER_00 4:27

It is. And for any founder or operator listening who manages cash flow, I mean the payback period is what really matters, right? Absolutely. The average payback period to completely recoup the investment across all business sizes is just 14 months.

SPEAKER_01 4:41

Just 14 months.

SPEAKER_00 4:42

Yeah. And for some specific functions like customer service routing or early stage lead qualification, the deployment pays for itself in the very first month.

SPEAKER_01 4:51

The first month. Okay, but let's ground this in everyday business expenses. Let's look at the human cost comparison because this is where the spreadsheet starts to look like a weapon.

SPEAKER_00 5:00

It really does.

SPEAKER_01 5:01

To handle a specific operational workload, let's say the workload of two to three full-time roles, a two to three person team costs an average of $250,000 to $555,000 per year.

SPEAKER_00 5:13

Right. Meanwhile, an autonomous digital employee operation handling that exact same workload runs roughly $18,000 to $48,000 per year.

SPEAKER_01 5:22

Okay, I have to push back a little here because half a million dollars for just a couple of roles, that sounds wildly inflated to me. Base salaries for three mid-level reps shouldn't hit anywhere near 500 grand.

SPEAKER_00 5:35

Well, base salary, no, but we are talking about the fully loaded cost of human labor.

SPEAKER_01 5:40

Oh, right.

SPEAKER_00 5:41

You're forgetting payroll taxes, healthcare benefits, 401k matching, computer equipment, monthly software seat licenses for every app they use.

SPEAKER_01 5:50

The management overhead.

SPEAKER_00 5:51

Exactly. Management overhead, the cost of recruiting them, and you know, the inevitable cost of replacing them when one quits in eight months.

SPEAKER_01 5:59

Yeah, we only ever look at the base salary on an offer letter. We completely ignore the massive hidden management drag that comes with hiring humans.

SPEAKER_00 6:07

Exactly. The math is actually terrifying if you are the one still paying those fully loaded costs while your competitor is paying $18,000 a year. Trevor Burrus, Jr.

SPEAKER_01 6:15

18 grand a year versus half a million. That friction is exactly what kills scaling.

SPEAKER_00 6:20

Yes. And it highlights a structural advantage that goes much deeper than just the raw dollar amount.

The Inverse Cost Curve Explained

SPEAKER_00 6:25

It's something the research calls the inverse cost curve.

SPEAKER_01 6:28

The inverse cost curve. Break that down for us.

SPEAKER_00 6:30

Aaron Powell Think about traditional hiring. When you hire humans, your cost goes up linearly, but your management complexity increases exponentially. Right. Two employees have one line of communication. Ten employees have 45 lines of communication. Wow. Suddenly you need middle managers, HR infrastructure, more office space, conflict resolution.

SPEAKER_01 6:52

It's the classic more people, more meetings problem. You hire people to do work, and eventually they spend half their day just coordinating with other people about the work.

SPEAKER_00 7:02

Precisely. Well, digital employees turn that entirely upside down. They invert the curve.

SPEAKER_01 7:06

How so?

SPEAKER_00 7:07

As your volume of work grows, say your customer ticket volume triples overnight because of a holiday promotion, the per unit cost of your digital workforce actually drops. Yeah, because once the core system is built, scaling it from 100 tasks to 10,000 tasks costs mere pennies in cloud computing. And the coordination overhead remains virtually zero.

SPEAKER_01 7:29

Because you don't need to hire a digital middle manager to oversee your digital customer service reps.

SPEAKER_00 7:35

Right. They scale instantly to meet the demand and they shrink instantly when the demand falls.

SPEAKER_01 7:40

Okay, so the math makes absolute sense for replacing humans to save money. But you know, you can only cut costs so far before you hit bone. True. Does this actually help a company grow its ceiling? Or is this just a massive, ruthless cost-cutting exercise? Because saving 100 grand is great, but it doesn't double my

Productivity Multipliers And Hybrid Teams

SPEAKER_01 7:58

revenue.

SPEAKER_00 7:58

That is the perfect question to ask because cost savings is only the defensive strategy. The offensive strategy is production ceilings.

SPEAKER_01 8:04

Okay, let's talk about the productivity multipliers.

SPEAKER_00 8:06

Yeah, when we look at the data here, the numbers honestly sound like typos until you look at the raw mechanics of how the work gets done. Let's walk through what these digital workers deliver function by function.

SPEAKER_01 8:18

Go for it.

SPEAKER_00 8:18

In customer support, the data shows a 15 times gain in productivity.

SPEAKER_01 8:23

Wait, 15 times the output? How is that even possible? Are they just typing that much faster than a human?

SPEAKER_00 8:30

No, no, it's about parallel processing and the total elimination of context switching.

SPEAKER_01 8:35

Context switching, right.

SPEAKER_00 8:36

A human agent typically handles 20 to 30 complex tickets per day. Why? Because they read a ticket, they open three different tabs to check the system, they get a Slack message from their boss, they lose their train of thought, they go back to the ticket, they write a response.

SPEAKER_01 8:52

Yep. Sounds like a normal Tuesday.

SPEAKER_00 8:54

Right. But an autonomous agent doesn't do things sequentially. It handles 200 to 500 interactions daily because it can process 300 tickets simultaneously in milliseconds.

SPEAKER_01 9:05

Without ever losing focus.

SPEAKER_00 9:06

Exactly. Without ever losing focus or switching contexts.

SPEAKER_01 9:10

That makes total sense. You aren't just speeding up the human. You're removing the human bottlenecks entirely.

SPEAKER_00 9:15

Exactly.

SPEAKER_01 9:16

What does this look like in the sales department?

SPEAKER_00 9:18

For lead qualification, we see a 12 times gain.

SPEAKER_01 9:21

12 times.

SPEAKER_00 9:22

Yeah. A human sales development rep might process 50 to 75 leads a day. They have to search the prospect, check their LinkedIn, find the company size, and write a custom email. Right. An autonomous system processes 500 to 1,000 leads a day. It hits data enrichment APIs instantly, scores the lead based on your ideal customer profile, and drafts and sends the outreach while your human sales team is asleep.

SPEAKER_01 9:49

That is wild.

SPEAKER_00 9:50

In data analysis, it's a 10 times gain in output. For content creation, it's a four times gain.

SPEAKER_01 9:56

So if you're a founder, I mean you aren't just giving your sales team bitter running shoes, you're handing them a teleporter.

SPEAKER_00 10:02

Basically, yes.

SPEAKER_01 10:03

You physically cannot hire enough humans to match a 15x output multiplier without bankrupting the company.

SPEAKER_00 10:10

You really can't. And what is really interesting is what happens when you create hybrid environments.

SPEAKER_01 10:14

What do you mean by hybrid?

SPEAKER_00 10:15

When you place these digital employees alongside your best human workers, letting the agents handle the massive volume of routine work, and letting the humans handle only the highly complex, nuanced exceptions.

SPEAKER_01 10:26

Okay, and what happens then?

SPEAKER_00 10:28

The overall per worker productivity gain across the entire team is roughly 73%.

SPEAKER_01 10:34

Aaron Powell A 73% lift across the board.

SPEAKER_00 10:37

Yes.

SPEAKER_01 10:37

So the core takeaway here is that this is not about doing the same job 10% faster. This is about producing 10 to 15 times the output for a fraction of the cost.

SPEAKER_00 10:49

Exactly.

SPEAKER_01 10:49

That is not a tool. That is a complete workforce transformation.

SPEAKER_00 10:53

It is the literal definition of workforce transformation. But you know, let's prove these multipliers aren't just theoretical spreadsheet math. Let's look at segment three, where the real money shows up in documented business outcomes.

Case Studies From Klarna To Small Retail

SPEAKER_01 11:05

Yes, let's get into the case studies. Start with customer service. Yeah. Because that seems to be where the bleeding edge is right now.

SPEAKER_00 11:10

It is. In customer service, these deployments routinely result in a 50 to 75% reduction in raw ticket volume that ever reaches a human being. Wow. And that directly translates to a 30% reduction in total departmental expenses.

SPEAKER_01 11:24

Okay, give us a real-world example.

SPEAKER_00 11:26

The gold standard case study for this is Klarna. They deployed an autonomous customer service operation that handled 2.3 million conversations.

SPEAKER_01 11:36

Wait, 2.3 million?

SPEAKER_00 11:37

Yes. That is equal to the work of 700 full-time human agents.

SPEAKER_01 11:41

700 agents.

SPEAKER_00 11:43

And by deploying this, Klarna drove an estimated $40 million in profit improvement.

SPEAKER_01 11:49

$40 million straight to the bottom line. That's incredible. But you know, customer service is a cost center. What about revenue generation?

SPEAKER_00 11:57

Okay, let's look at sales.

SPEAKER_01 11:58

Because replacing a support rep is one thing. But trusting an AI to close the gap on top line growth feels like a much bigger leap of faith for an owner.

SPEAKER_00 12:07

It is a leap, but the data proves it works. In sales and revenue operations, organizations deploying agentic workers are seeing a 13 to 15% overall revenue increase.

SPEAKER_01 12:18

Just from deploying agents.

SPEAKER_00 12:19

Yes. And a 10 to 20% lift in sales ROI. But here is the statistic that should make every sales director listening freeze in their tracks. What's your sales teams running with autonomous agents hit their annual quota 83% of the time. Teams without them hit quota only 66% of the time.

SPEAKER_01 12:38

Oh wow. That is the literal difference between a business hitting or missing its annual target.

SPEAKER_00 12:44

It really is.

SPEAKER_01 12:44

That's the difference between a massive bonus pool and laying off 20% of your staff.

SPEAKER_00 12:49

Absolutely. And we need to be very clear that this isn't just a game for massive giants like Klarna or global enterprise sales team.

SPEAKER_01 12:56

Right, because we are talking to small and mid-sized businesses too.

SPEAKER_00 12:59

Exactly. The back office and operational savings for smaller teams are just as impactful. Small businesses are quietly pocketing $7,500 to over $20,000 per year in pure operational savings.

SPEAKER_01 13:14

Let's talk about the boutique clothing store example from the research because this is where it gets incredibly relatable for the average operator.

SPEAKER_00 13:21

I love this example. There is a documented case study of a boutique clothing store with just seven employees.

SPEAKER_01 13:26

Just seven.

SPEAKER_00 13:27

Yeah. They implemented a digital employee to handle all of their inbound customer inquiries, sizing questions, and inventory checks.

SPEAKER_01 13:35

Okay. And what was the impact?

SPEAKER_00 13:36

They saved $1,000 a month in part-time labor costs, which is great. But the critical part is what happened to their customer experience. Right. Their customer satisfaction score actually climbed from 4.2 to 4.7 out of five.

SPEAKER_01 13:49

I mean, I hear massive numbers like Klarna and I get it, but that seven-person boutique example is what really gets me.

SPEAKER_00 13:56

Why is that?

SPEAKER_01 13:57

Well, they slashed their labor costs, they fired the human element from that specific touch point, but their customer satisfaction went up. How is that simply because a digital employee never has a bad day and never makes a customer wait on hold?

SPEAKER_00 14:13

That is exactly the mechanism. Think about what customers hate most. They hate waiting.

SPEAKER_01 14:17

Yeah, and repeating themselves.

SPEAKER_00 14:18

Exactly. The digital employee has infinite patience, it responds in three seconds, and it has perfect recall of the inventory database.

SPEAKER_01 14:26

It doesn't get overwhelmed and short-tempered during a Black Friday rush.

SPEAKER_00 14:30

It just executes calmly and perfectly every single time. And when you scale that exact same reliability up to the enterprise level, the numbers become astronomical.

SPEAKER_01 14:39

Like what?

SPEAKER_00 14:40

Well, JP Morgan Chase extracted $1.5 billion in efficiency and fraud prevention out of its workforce using these systems. Aaron Powell.

SPEAKER_01 14:50

Okay, I want to pause here and play devil's advocate for segment four.

Why Most AI Efforts Fail

SPEAKER_01 14:54

If the financial math is this obvious, I mean if the ROI is 10 to 1, the productivity multipliers are 15X, and a seven-person retail shop can pull this off, why isn't every single business successfully doing this?

SPEAKER_00 15:07

Aaron Powell It's a great question.

SPEAKER_01 15:08

Why aren't we seeing a massive wave of hyper-profitable companies everywhere we look?

SPEAKER_00 15:13

Aaron Powell Because the execution is vastly different from the theory, and most owners get the strategy completely wrong. They fall into the fatal trap of what we call shadow AI.

SPEAKER_01 15:21

Shadow AI. That sounds like something out of a spy novel. What is that?

SPEAKER_00 15:24

It is far more mundane and far more wasteful, honestly. Shadow AI is when a business owner buys a $20 monthly chat subscription, writes a few basic prompts, gives their human employees access to the tool, and tricks themselves into believing they have an AI strategy.

SPEAKER_01 15:41

Oh man.

SPEAKER_00 15:42

Yeah. They are still entirely relying on their human workers to remember to use the tool, to prompt it correctly, to fact-check it, and to manually copy and paste the results into their actual workflow.

SPEAKER_01 15:54

Aaron Powell Going back to my analogy, they bought this slightly faster typewriter and they think they hired the typist department.

SPEAKER_00 16:00

Yes. And this is exactly why 80% of corporate AI projects fail to reach meaningful production. 80%. Yep. They end up in the graveyard because they were never designed to be autonomous workers in the first place. They were designed as minor novelties to help humans brainstorm.

SPEAKER_01 16:15

Which naturally leads into what the source called pilot purgatory.

SPEAKER_00 16:18

Oh, yes.

SPEAKER_01 16:19

We have this terrifying concept in the research called the $500,000 warning. Is this where companies spend a year and a half million dollars building a custom system just to have it summarize a client email that I could have read in 10 seconds anyway?

SPEAKER_00 16:33

That is exactly what pilot purgatory is. Companies will waste months and hundreds of thousands of dollars in salaries building pilots that just sit on a shelf.

SPEAKER_01 16:42

Why do they fail so often?

SPEAKER_00 16:44

Because the business treats the AI like a magical black box that will solve all their problems. Rather than treating it like a new employee that needs a specific job description, clear metrics, and strict operational boundaries.

SPEAKER_01 16:58

Right. If you hire a human junior accountant, you don't just point them at the company bank account on day one, give them the password, and say, go figure out our taxes.

SPEAKER_00 17:05

Of course not. You give them guardrails.

SPEAKER_01 17:07

Exactly.

Guardrails Selection Power And Governance

SPEAKER_00 17:08

And yet that is how so many companies treat their AI deployments. You must treat an agent's actions like employee permissions. They need strict boundaries. They need specific spending limits, they need timeout constraints so they don't get stuck in an endless loop trying to solve a problem. And most importantly, they need human approval gates for high-stakes commitments.

SPEAKER_01 17:28

And this touches on a really fascinating concept from the research regarding how these models actually think. The concept is called selection power.

SPEAKER_00 17:37

Oh, selection power is absolutely critical to understand. It is the agent's ability to decide which options it even presents to you in the first place. Right. If you don't bound an agent's authority, it's selection power, it can confidently make terrible business decisions while looking perfectly compliant. We call this cosmetic alignment.

SPEAKER_01 17:57

Cosmetic alignment. Explain why that happens. Because on the surface, I mean, if the AI does what I ask, shouldn't it be aligned with my goals? Why would it fake it?

SPEAKER_00 18:06

Because these models are fundamentally built and trained to please the user. They are people-pleasing algorithms. They want to give you an answer that looks correct and sounds polite. So if they find a lazy or incomplete way to complete the task you gave them, but they can format it nicely and log their actions, they will take that path of least resistance.

SPEAKER_01 18:26

Just to make you happy in the short term.

SPEAKER_00 18:27

Exactly. The agent looks aligned cosmetically, it follows the formatting rules. But under the surface, it might be excluding the best strategic options or missing critical data because it optimized for finishing the task quickly rather than finishing it correctly.

SPEAKER_01 18:42

Aaron Powell So the biggest operational risk for a business owner isn't some sci-fi dystopian scenario where the AI becomes sentient and takes over the world.

SPEAKER_00 18:52

Not at all.

SPEAKER_01 18:52

It's that the AI, eager to please you, will confidently give away your core product for free or get stuck in a forty-seven thousand dollar recursive cloud computing loop over a long holiday weekend simply because you didn't give it a basic budget cap.

SPEAKER_00 19:07

That is the exact threat. The risks are incredibly mundane, but they are financially devastating. A digital employee will do exactly what you tell it to do, a thousand times a second, until it runs out of money. If you do not govern it like a human worker with a corporate credit card, you're asking for a PL disaster.

SPEAKER_01 19:25

It all comes back to governance and guardrails. You are managing a digital workforce, not just installing a new app on your phone.

SPEAKER_00 19:33

Precisely. And the organizations that understand how to govern selection power, the ones who know how to put those operational circuit breakers in place, are the ones who are safely extracting that 10 to 1 return on investment while their competitors are stuck in pilot purgatory.

The Recession Stress Test And Closing

SPEAKER_01 19:48

We have covered a massive amount of ground today, and it all points in one very stark direction. Let's summarize the hard truth for you, the operator listening to this right now.

SPEAKER_00 19:58

It's time to wake up.

SPEAKER_01 19:59

Autonomous AI agents have moved from an experimental side project to standard operating procedure for the modern business. The math is not ambiguous. Not at all. Whether you are a seven-person retail boutique saving $1,000 a month or JP Morgan Chase saving $1.5 billion, failing to deploy digital employees isn't just standing still. It is actively funding your competitors' advantage. Yes. You are paying half a million dollars for a workload that they're getting done for $30,000.

SPEAKER_00 20:29

And because of the inverse cost curve we discussed earlier, that gap in efficiency will only compound over time. The longer you wait, the further ahead their cost advantage gets.

SPEAKER_01 20:40

So I want to leave you with a chilling, forward-looking thought to chew on, something we haven't even fully explored yet, but that sits at the very edge of this research. What happens to your market share during an economic downturn?

SPEAKER_00 20:51

Oh, a recession is the ultimate stress test for a business model.

SPEAKER_01 20:55

Think about it. When your competitors have inverse cost curves, meaning they can scale up their digital operations instantly during peak seasons without hiring a single person, and then scale down instantly during a recession without paying painful severance packages or dealing with morale crushing layoffs.

SPEAKER_00 21:11

While you were trapped.

SPEAKER_01 21:12

Exactly. You are dealing with fixed human payrolls, employee burnout, middle management bloat, and total hiring freezes. When the market contracts, who survives?

SPEAKER_00 21:23

The competitive advantage always goes to the most adaptable cost structure.

SPEAKER_01 21:27

Right. The ultimate competitive mode of the future is not your branding. It is not even your product. It is the sheer compounding cost efficiency of your digital workforce.

SPEAKER_00 21:37

It's that simple.

SPEAKER_01 21:38

Precisely. So challenge yourself. Tomorrow morning, when you open your laptop and look at your profit and loss statement, I want you to look at those massive labor and operational line items.

SPEAKER_00 21:48

Take a really hard look.

SPEAKER_01 21:49

Remember the difference between buying a faster typewriter and hiring a team of tireless typists. And ask yourself the only question that matters on the modern business battlefield who is actually doing the work?