Brian Potter, "The Origins of Efficiency" (Stripe Press, 2025)

Summary

---

Podcast Summary: Brian Potter on "The Origins of Efficiency"

New Books Network | Hosted by Gregory McNip | October 28, 2025
Guest: Brian Potter, Senior Infrastructure Fellow at the Institute for Progress and author of "The Origins of Efficiency" (Stripe Press, 2025)

---

Episode Overview

This episode features a wide-ranging conversation between host Gregory McNip and author Brian Potter on his new book, The Origins of Efficiency. The discussion explores why some industries become more efficient over time while others lag behind, examining the technical, historical, and systemic drivers of productivity improvement. Potter draws on his experience in construction and his research to uncover the mechanics and broader societal implications of efficiency—offering accessible explanations, vivid case studies, and insightful reflections for anyone fascinated by the progress of technology, manufacturing, and civilization itself.

---

Key Discussion Points & Insights

1. Why Write The Origins of Efficiency? (02:06–05:53)

• Personal Motivation:
  "For most of my career… [the construction industry] seemed quite inefficient to me as well." (02:49, Brian Potter)
  Potter’s time at Katerra, a construction startup, shaped his questions about why productivity advances succeed in some fields and stall in others.
• Audience:
  Written to satisfy Potter’s own curiosity, the book targets anyone interested in improvement—business, government, or the curious layperson.

2. Defining Efficiency (06:22–07:01)

• Resource Minimization:
  Potter frames efficiency as reducing the resources (labor, materials, energy, cost) required for a process:
  "How stuff gets cheaper over time, which is… a pretty good measure of… the amount of inputs and resources it requires…" (06:52, Brian Potter)
• Broader Progress:
  Efficiency unlocks wider access to goods and underpins the story of societal advancement.

3. Case Studies: Penicillin, Printing, and Progress (07:13–10:12)

• Penicillin as a Metaphor for Civilization:
  "The story of penicillin is the story of civilization… It started out very, very expensive and difficult to make… Eventually, we figured out how to mass produce it…" (07:13–09:01, Brian Potter)
• Scalability Through History:
  Example: “By 1600, roughly as many books were printed in Western Europe every five years as had been produced in the entire thousand year period between 450 and 1450.” (09:55, Gregory McNip quoting book)

4. The Five (Plus) Routes to Efficiency (10:12–17:09)

Potter outlines five core ways to improve efficiency in production processes:

• 1. Fundamental Method Innovation:
  E.g., The progression in nail manufacturing from hand-forged to mechanical to modern machines.
• 2. Cheaper Inputs:
  Locate production where resources (like labor or electricity) are inexpensive.
  Example: Iceland’s aluminum industry leverages cheap hydroelectric power.
• 3. Scaling:
  Making more units reduces average costs due to fixed costs spread over outputs.
• 4. Reducing Variation:
  Minimize errors and unpredictability, leading to fewer defective products and waste.
• 5. Smoothing Process Flow ("Buffers"):
  Align process steps—reduce inventory between steps, increase reliability, cut costs.
• 6. Cutting Steps ("Process Simplification"):
  Remove unnecessary motions or tasks (a major insight from lean manufacturing and the Toyota Production System).
• Continuous Flow Ideal:
  “Very efficient production processes… tend to kind of converge on this similar form of a continuous transformation of goods from inputs to outputs.” (18:10, Brian Potter)

5. Value-Added vs. Non-Value-Added Work (19:50–22:06)

• Pareto Principle in Operations:
  Only about 10% of typical process steps are truly value-adding; the rest are ancillary or waste.
  “Only the like putting the bracket in place and attaching the two screws are actually contributing to the final product… everything else… is not strictly necessary…” (21:17, Brian Potter)

6. The S-Curve of Technological Change (22:45–27:52)

• S-Curve Concept:
  Technological progress follows an S-curve: slow improvement, then rapid gains, then plateau as the limits of a technology are reached.
  • Invention > Rapid Improvement > Plateau (limits) > New paradigm needed
  • Light bulbs as an example: “You needed to… introduce a fundamentally new technology [like LEDs]… each… with a higher luminous efficiency…” (27:16, Brian Potter)

7. Design for Manufacturability & Value Engineering (28:03–33:44)

• Production Process Reflects Product Design:
  Efficient production depends on designing products that are easier and cheaper to manufacture.
• Value Engineering:
  “It’s basically just looking at whatever it is you’re producing… and whether you can sort of get similar value for less expenditure.” (33:23, Brian Potter)
  Even tiny savings (e.g., swapping a cover) become major at scale.

8. Mechanization’s Benefits (29:56–31:28)

• Human labor is costly and limited; machines can work faster, increasing output and lowering cost per unit.

9. Vertical Integration: Pros and Cons (34:43–35:53)

• Benefit: Lower costs, more reliable supply.
• Drawback: Can become cumbersome, less agile in changing markets.
• Automotive case study: Ford’s shifting approach to vertical integration over time.

10. Manufacturing Innovations – Toyota & Tesla (36:39–38:48)

• Both moved the industry towards greater efficiency.
  • Toyota: Lean manufacturing, minimal waste, flexible and efficient systems.
    • “Flexibly at lower volumes.” (37:30)
  • Tesla: Large aluminum castings to consolidate car frames, lowering cost and complexity.
    • “A real pioneer of these sorts of large castings.” (38:30)
• Their once-unique methods are now widely adopted.

11. Economies of Scale & Learning Curves (38:54–41:48)

• Scale is vital: “With enough scale and enough accumulated improvements, production costs can be driven down by orders of magnitude.” (interview summary)
• Learning curve: Each doubling of production leads to a constant % decrease in cost (solar PV as example).

12. Division of Labor (41:48–43:10)

• Not always a panacea; benefits depend on context.
• Toyota (again) sometimes finds less division brings more efficiency by cross-training workers.

13. Shipping Container’s Transformative Impact (43:12–43:56)

• Containerization: Dramatically reduced shipping costs, enabling global concentration of production and massive economies of scale.

14. Variation, Statistical Process Control (43:56–45:09)

• Variation: Deviations from desired output add cost and waste.
• Control: Statistical methods (originated at Bell Labs) help identify and remove sources of variation.

15. Bundled and Cascading Process Improvements (45:14–46:43)

• Positive cycles: Lower costs → higher demand → more output → even lower costs.
• Negative cycles: Rising costs → falling demand → even higher costs.

16. Assembly Line & Mass Production (47:02–48:42)

• Incremental to Breakthrough: The assembly line wasn’t a sudden leap, but an extension of earlier approaches to high-volume, continuous flow.

17. AI’s Potential for Efficiency in Resistant Sectors (48:53–50:46)

• Limited in already-optimized domains (e.g. mass manufacturing): “A humanoid robot would be so much worse at doing this task… than this specific dedicated equipment…” (49:43, Brian Potter)
• Biggest potential: Sectors with personalized, variable tasks (healthcare, education, car repair) where current approaches resist standardization and automation.
  • “If you have an AI that can respond flexibly based on environmental contexts quickly and cheaply, that will—maybe—have... improvements in areas that have historically been resistant.” (50:31, Brian Potter)

---

Notable Quotes & Memorable Moments

• On the Penicillin story as a metaphor for civilization (07:13):
  “The story of penicillin is the story of civilization… that basic process where we’re trying to sort of take these things that are very, very expensive to produce and figure out how we can make them more cheaply and easier so they’re more widely available—that is basically the story of how civilization has progressed over the course of human history.”
  — Brian Potter (08:04)

• On value-added vs. non-value-added steps (21:17):
  "Only the like putting the bracket in place and attaching the two screws are actually contributing to the final product. Everything else... is not strictly necessary."
  — Brian Potter

• On the S-curve of technology (23:49):
  “Eventually you get to a point where things start to click... performance starts to improve more and more rapidly... And then you start to run out of improvements... you weren't really seeing dramatic improvements in performance anymore, just because you had reached the ceiling of what this technology was capable of.” — Brian Potter (24:17–27:07)

• On AI and resistant sectors (49:50):
  “Where I think you’re going to see the benefits from these types of improvements in AI and robotics… is in these areas that have not seen really huge improvements… things like car repair, medical care, and education… where it’s expensive in part because it’s tailored to a specific set of circumstances…”
  — Brian Potter

---

Key Timestamps

• [02:06] — Why Brian wrote the book
• [06:22] — Defining efficiency
• [07:13] — Penicillin as a case study for civilization’s advance
• [10:12] — The five main drivers of efficiency
• [19:50] — Value-added vs. non-value-added process analysis
• [22:45] — The S-curve and technological progress
• [29:56] — Mechanization’s impact
• [34:43] — Vertical integration: pros and cons
• [36:39] — Toyota and Tesla as efficiency innovators
• [38:54] — The power of scale and learning curves
• [43:12] — Shipping container’s influence on global manufacturing
• [45:14] — Cascading and bundled process improvements
• [47:02] — Origins of the assembly line
• [48:53] — AI’s potential in hard-to-standardize sectors

---

Conclusion

Brian Potter’s The Origins of Efficiency reframes “efficiency” as a pillar of human progress, showing that transformative advances in how we produce goods have made civilization richer and better-off. Through real-world examples, clear-sighted historical analysis, and practical frameworks, Potter helps listeners (and readers) understand not only how industries evolve, but also why persistent inefficiencies emerge, and how breakthroughs can ripple through society. His insights emphasize that while some lessons of efficiency are nearly universal, every industry and era brings its own puzzle—and future advances (perhaps thanks to AI) may come not where processes are already optimized, but where complexity and customization have long resisted improvement.

---

Transcript

[Marshall Po]

Hello everybody. This is Marshall Po. I'm the founder and editor of the New Books Network. And if you're listening to this, you know that the NBN is the largest academic podcast network in the world. We reach a worldwide audience of 2 million people. You may have a podcast or you may be thinking about starting a podcast. As you probably know, there are challenges basically of two kinds. One is technical. There are things you have to know in order to get your podcast produced and distributed. And the second is, and this is the biggest problem, you need to get an audience. Building an audience in podcasting is the hardest thing to do today. With this in mind, we at the NBM have started a service called NBN Productions. What we do is help you create a podcast, produce your podcast, distribute your podcast, and we host your podcast. Most importantly, what we do is we distribute your podcast to the NBN audience. We've done this many times with many academic podcasts and we would like to help you. If you would be interested in talking to us about how we can help you with your podcast, please contact us. Just go to the front page of the New Books Network and you will see a link to NBN Productions. Click that, fill out the form and we can talk. Welcome to the New Books Network.

[Gregory McNip]

Welcome to the New Books Network. I'm your host, Gregory McNip and I'm excited to be joined by Brian Potter, the author of the Origins Efficiency. The book was published by Stripe Press in the United States in October of 2025. Brian Potter is the author of the Construction Physics newsletter and a Senior Infrastructure Fellow at the Institute for Progress. He writes about the technology and economics of building construction with a focus on improving productivity and reducing costs. He previously managed an engineering team at Katera, a softbank backed construction startup, and has 15 years of experience as a structural engineer. He holds a Bachelor's in Civil Engineering from Georgia Tech and a Master's in Systems Engineering from the University of Central Florida. I selected the Origins of Efficiency because it connects the mechanics of how we produce things with the broader story of how civilization improves itself. Brian takes what might seem like a technical question, why some industries get more efficient and others don't, and turns it into a lens for understanding progress itself. Hello Brian, thank you for joining me today to discuss your book.

[Brian Potter]

Yeah, thank you for having me.

[Gregory McNip]

Brian, why did you write the Origins of Efficiency and who is the target reader?

[Brian Potter]

Yeah, so I, like you said, my.

Background is in the construction industry. I spent 15 years as a structural.

Engineer and I had previously worked at this construction startup called KATERA and an important fact about the construction industry is that measures of, by measures of productivity, it's never really gotten more efficient over time.

You know, in terms of like the.

Amount of labor required to produce a.

Given amount of building or produce a house, you know, the cost of building has basically steadily risen at the level of overall inflation or faster than overall inflation.

And from inside the industry, it often quite seemed quite inefficient to me as.

Well, for most of my career.

But then in 2018, I had a chance to join this construction startup, Katera, that said, oh, we're going to make this, you know, we're going to improve how this industry works.

We're going to make it so much more efficient. We're going to move everything into factories and it's going to be, you know, so much faster and easier and cheaper. We're going to be the, the Henry Ford of housing.

And it, it basically didn't work. And there was, you know, there's various reasons why that happened, but it partially, I came to believe that their sort of thesis that they were operating under that is if you don't just move this stuff into a factory and that's.

All you need to do to kind.

Of make this process more efficient wasn't quite enough to sort of get them there. And I wanted to kind of better understand what specifically it would take to.

Sort of make some process more efficient.

And as you said, why some, why.

Some of these industries seem to sort of improve their productivity over time and.

Others don't really do that. And what are the specific things that.

Make it possible in some industries and not in others.

And so I wanted to understand when a process is sort of getting more.

Efficient, what specifically kind of makes that possible.

And so that was sort of the.

Origins of the book.

[Interviewer]

And in terms of target reader, I mean, certainly I think anyone can benefit from that works in any type of business or government. I mean, it's really understanding what you do and how to do it better. But I'll let you.

[Brian Potter]

Yeah, yeah, pretty much a, you know, a flaw or advantage of the way that I work is that I'm usually writing things with myself as the primary audience in that I have. There's some given thing that I don't.

Feel like I understand very well.

And I personally want to sort of understand how this thing works better. And because of, you know, the way that I think the easiest way for me to do that is just like produce some sort of written explanation that's.

Like as clear and simple as possible.

It's like here, how's these, here's how.

These parts fit together, here's how this whole thing works.

Uh, and then it turns out that other people find those sorts of things valuable as well. But with the book it was kind of the same thing. Whereas like, you know, I just joined this construction startup that tried to sort.

Of make this industry more efficient and.

It did not work at all. Lots of other companies have tried to.

Try to do sort of do the same thing.

I don't understand what sort of why they're all failing to do this. I don't really understand why. What specifically can make some process more efficient over time, you know, make it.

Cheaper and, and easier to produce.

I have some sort of vague ideas, I know, like things that have worked in specific places, but I don't have like a general theory that kind of.

Ties it all together.

And so the whole, the exercise of the book was so I could understand.

Myself how this thing worked together.

But yeah, as you say, it's people.

That work in industry, people that work in sort of government, people that sort.

Of are interested in sort of just how things kind of work in general.

I think they'll all kind of. Yeah. Be able to find value in the book.

[Interviewer]

Yeah, I mean, towards the end of.

[Gregory McNip]

The book, I really thought the last.

[Interviewer]

Chapter and the conclusion was very insightful. And you talk about transferring some of.

[Gregory McNip]

The productivity improvements or process improvements from.

[Interviewer]

High volume manufacturing to areas that have historically been resistant to efficiency improvements like education, healthcare, et cetera. So I definitely want to ask you about that, but starting sort of at the beginning of the book, how do you define efficiency in terms of the context that you're writing here?

[Brian Potter]

Sure.

So in the book I basically talk about things that can make a process.

Get cheaper over time.

So reducing the amount of resources that go into it.

So less labor, less material, less, you know, energy, less cost overall.

There's other ways you could kind of contextualize it.

Some people think of efficiency more in terms of how fast you can do something.

And a lot of those, you know.

Similar, these ideas will apply to sort of these other similar concessions.

But when I'm, when I'm focusing on the book, book is mostly how stuff.

Gets cheaper over time, which is, you know, a pretty good measure of, you.

Know, the amount of inputs and resources.

It requires to sort of make something.

[Interviewer]

Yeah. You begin the book with several case studies. The first one, I think is the story of penicillin. And you say, quote, the story of penicillin is the story of civilization. What do you mean? By this.

[Brian Potter]

So, yeah, the sort of arc is civilization. If you go back, you know, hundreds or thousands of years, people were like, you know, vast majority of people were.

Extremely, extremely, extremely poor. They just had very, very, very little.

And a big, you know, part of that is just, you know, a lack of, like, technology. They didn't have modern technology. But closely coupled to that idea is that it took a lot of time and effort to sort of make any given thing.

You know, whether you're talking about steel, whether you're talking about, you know, books, whether you're talking about clothing, any of it.

It just took huge, huge, huge amounts.

Of, like, physical effort and toil to kind of make all this stuff.

And because it took so much effort, it was so expensive. And so it was just not very available to sort of the vast majority of people.

Most people couldn't afford to have most things.

And the story of the last, you.

Know, 100, several hundred years of human. There are thousands of years of human civilization is just civilization just getting richer and richer and richer over.

And a big part of that is because we figured out how to sort of make this stuff for any given.

Thing that you want to make, whether it's cloveling or steel or books or.

Whatever, try to make those things less.

Expensively, take fewer inputs over time.

And so the story of penicillin is this thing that started out very, very.

Very expensive and difficult to make. They could only produce a very, very tiny amount of it early on.

And a big project during World War.

II was trying to figure out ways to sort of be able to mass produce this. So it cost. So could they produce in very, very large numbers for, you know, the entire arm allied armed forces, basically. And this project was successful and they. And they figured out how to do that.

But that basic process where we're trying to sort of take these things that.

Are very, very extensive to produce and figure out how we can make them for more cheaply and easier so they're more widely available to more people. Is basically the story of how civilization has progressed over the course of human history.

[Interviewer]

Yeah, I found that fascinating because we normally associate the product with the inventor, Edison, obviously, or the printing press or penicillin. But as you point out in the book, it can take decades or even longer to refine the process and get it to the point where we can actually mass produce it. So kudos to those people who are involved in that. I wish they were more recognized and hopefully your book will do that. I want to cite one. You provide a lot of metrics but one that stood out for me. You mentioned books. By 1600, roughly as many books were printed in Western Europe every five years as had been produced in the entire thousand year period between 450 and 1450. So it seems like the printing press and being able to produce it in math quantities or produce books really have like a scaling function there.

[Gregory McNip]

It's really impressive.

[Brian Potter]

Yeah, for sure.

[Interviewer]

Just you what's. I just want to move into the production process. You spend a significant amount of time in the book just defining it. You come up with five factors. Could you briefly describe them and talk about continuous flow?

[Brian Potter]

Sure.

So yeah, A production process is basically a step by step sequence of actions you take to sort of transform some given set of inputs, whether that's material or labor or energy or whatever, into some output. And it's the easiest way to think.

About that is some factory that takes raw material and transforms it and outputs some widget or whatever.

But you can think of that same basic model as applying to anything, whether.

You'Re growing crops or performing some service like cleaning a house, you can still kind of think about it as sort of a step by step sequence where you're starting with some set of inputs and gradually transforming it into some sort of finished output.

And yeah, I sort of, I talk about, there are, I categorize sort of five different ways that you can kind of intervene in a process like this.

To kind of make it cheaper and more efficient and reduce the amount of resources that is required to do it.

The first is that you can sort of invent some sort of fundamentally new method of doing part of the process.

Sort of, this would be sort of.

Introducing new kind of technology. And so in the book I talk about, you know, I give lots of.

Examples of this, but one of them is I talk about nail manufacturing, you know, nails to nail pieces of wood together.

Where in the early, you know, late.

18Th century, early 19th century, these were made basically by hand forging. Somebody would, you know, hammer a piece of iron into the correct shape.

This eventually got replaced by a machine that could like cut nails out of.

Sheets of sheets of iron.

And then those machines got replaced with.

Even better machines that made nail in a different way out of, out of pieces of iron and steel wire.

And so by changing sort of the.

Fundamental method of what you're doing, you can sort of change the amount of resources that a process requires.

And then the next sort of point of intervention is you can reduce the amount of inputs that a process requires. This is very similar to sort of.

That first point of intervention. But it's got some subtle differences.

Basically, you can kind of think of a production process as like a recipe.

And you take some series of ingredients and you do some specific things to them. You get, you know, whatever the recipe is for.

And if you can reduce the sort of cost of those ingredients, you can, whatever it is that you're making, you.

Can kind of make it more, more cheaply over time.

And so, you know, an example of.

This is finding cheaper sources of labor. This has been a very, you know.

Important or, you know, widely used strategy.

For manufacturers over time to sort of reduce their costs is that they're constantly moving around to new sources of low cost labor.

But you also see it for other things too. So an example that I talk about in the book is that Iceland produces.

A surprisingly large amount of aluminum, like almost as much as aluminum as the U.S. even though Iceland is a tiny, tiny country and the US is a huge, enormous country.

And the reason is that Iceland has.

Very, very inexpensive electricity. They have cheap hydroelectric power.

And that makes it.

Because aluminum requires such huge amounts of electricity to produce it, it's very valuable to locate your aluminum manufacturing in a place where your electricity is very inexpensive.

And so the next one that I.

Talk about is scaling effects. Basically, the more that you make of something, if you make something in larger and larger volumes, you can produce it more and more cheaply. And there's kind of a variety of different mechanisms for this. But one easy one that everyone kind of, I think intuitively understands is that.

When you're making something, some of your.

Costs associated with that thing are fixed and don't increase the more that you make of something. And so if you can take your fixed costs and spread them over a larger and larger output, that sort of unit cost for whatever that you're producing falls over time.

And, you know, there's several different actually.

Scaling mechanisms that you can do, and I kind of go over all of.

Them in the book, and then the last two are kind of closely related, is that you can reduce the sort of variation in your process. A given production process. It will never like, produce exactly what.

It'S supposed to do every single time. There will always be some sort of variation in what. If it does, sometimes it will fail and produce the wrong thing or produce something that's out of tolerance or whatever.

And the more that you can sort.

Of reduce those failures and make your process more and more reliable, the more inexpensive that whatever you're producing will be. And then sort of the last one.

Is there's, you Know, for any given. We talked before about, hey, you know.

Production process is sort of this multi step process where, you know, you have one thing and then the next thing and then the next thing.

And if those steps in the process.

Aren'T like completely aligned in like how fast they work and when one part feeds into the next, the next part may not immediately be ready for it because it's running at a different rate, or it's because there's some failures in the process anyway.

But because of these sort of differences.

In different parts of the process, you can have what are called buffers of partly produced material build up between the different steps.

And if you have these buffers, those.

Buffers basically add cost to your process. Because if you have all this stuff, you need space to store, you need people to sort of keep track of it and all these things.

So if you can kind of reduce.

The cost of your buffers, and basically a big part of that is sort of making your process more reliable and predictable. So one part of it smoothly and predictably feeds into the next part, you.

Can reduce the cost of those buffers.

And reduce the cost of whatever it is that you're producing.

So those are the five steps. Then I also have like a six.

Thing that you can do, which is that you can also just cut a step out of the production process completely. And when you're able to do that, you can, you know, every cost associated with that step goes with it as well. This is a big insight of the sort of Toyota production system and lean manufacturing. But it's been noticed by manufacturers for a very long time that a lot of the steps in a given process are not actually contributing anything of value to what you're producing. They're things like moving things around from point A to point B, that maybe if you were more clever in your arrangement, you didn't actually have to do those sort of movements and you know, storing inventory that maybe you don't actually need to store and things like that.

So if you can cut out these.

Sort of wasteful, unneeded steps, you can reduce the cost of your process overall.

And you mentioned continuous processes and kind of when you're able to sort of do all these improvements and really push them forward, what your process kind of.

Starts to remember or resemble, excuse me.

Is a sort of continuous transformation of, you know, inputs into outputs that it.

Just goes smoothly and swiftly through your process without any sort of waiting or breakdowns or buffering or anything like that. Your process kind of resembles one basically giant machine just moving stuff through it continuously.

You know, people have, you know, talk.

About things working like a well oiled machine. And this is sort of that kind of similar intuition that when everything is like working smoothly and correctly and as it should, everything just goes from step to step to step to step immediately without sort of any waiting or delays or problems at all.

And so kind of efficient, very efficient.

Production processes kind of tend, whatever they are in whatever industry you're talking about tend to kind of converge on this similar form of like a continuous transformation of goods from inputs to outputs.

[WhatsApp Ad]

When did making plans get this complicated? It's time to streamline with WhatsApp, the secure messaging app that brings the whole group together. Use polls to settle dinner plans, send event invites and pin messages so no one forgets mom 60th and never miss a meme or milestone. All protected with end to end encryption. It's time for WhatsApp message privately with everyone.

[eBay Ad]

Learn [e-mail] this episode is brought to you by eBay. Buying parts for your car. You'll know that will it work Feeling. But on ebay, buying parts is different. EBay's free returns means if it doesn't fit or if it isn't what you expected, you just print a label, drop it off and get your refund fast. No haggling, no stress, and at least 30 days to return any eligible item. Millions of parts, free returns, eBay, things people love Eligible items only exclusions apply.

[Indeed Ad]

This episode is brought to you by Indeed. Stop waiting around for the perfect candidate. Instead use Indeed Sponsored Jobs to find the right people with the right skills fast. It's a simple way to make sure your listing is the first candidate. C According to Indeed data, Sponsored Jobs have four times more applicants than non sponsored jobs. So go build your dream team today with Indeed. Get a $75 sponsored job credit@ Indeed.com podcast. Terms and conditions apply.

[Interviewer]

Yeah, that comes across very clearly and I want to circle back on Toyota. Obviously they seem to have really been a pioneer in manufacturing. I did want to ask about the value and the non value steps. I think at one point you say roughly 10% of the overall production process. Remember you cite a statistic is value added and the remaining is non value added. Is that correct? Is that how you think about it?

[Brian Potter]

That's certainly what like the consultants that.

Like study these sorts of things have.

Found is that yeah, a lot of, you know, if you talk when you're talking about like value versus value add.

Versus non value add, the value add.

Is stuff that like actually is making.

The physical thing or doing the physical transformations that are fundamentally necessary to produce whatever it is that you're producing. But then surrounding these fundamentally necessary things.

Are all these other ancillary steps that.

Kind of serve to support the value.

Added parts of it, but aren't strictly.

Necessary to produce whatever it is you're producing.

So I talk about, I gave sort.

Of an example in the book of like, you know, a sketch of somebody.

Like installing a part on a car.

In a factory and sort of, you know, they take the, they take the Bartlett bracket or whatever out of the box, they walk over to, you know, where it needs to be installed, they put it in place, they attach two screws and then they walk back.

And of that, of all those steps, only the like putting the bracket in.

Place and attaching the two screws are.

Actually contributing to the final product. Everything else like the walking and the.

Picking stuff up and the carrying and.

Things like that is like not strictly.

Necessary to sort of make the product. It's sort of an ancillary step that you've put in there to sort of help you out.

But if you can sort of be.

A little bit more clever in your.

Arrangement, you can find ways to sort.

Of eliminate a lot of these like non value adding steps which as you know, tend to be.

There tend to be a lot of.

Them in the production process or at least historically there were. Manufacturers have over the past several decades become a lot more sensitive to sort of the presence of these nonvaluated steps and have worked harder to kind of eliminate them.

[Interviewer]

Yeah, I really enjoyed the sort of write up or your analysis of Gilbreth's study of motion. I thought that was fascinating how they went through each step in the process and tried to sort of provide a classification or taxonomy system to identify what was value add and what wasn't and how to I guess simplify and streamline the process. I want to move on to. You talk about S curves and I think everyone's familiar with them. But how do you think about S shaped curves in terms of technology? Obviously innovation and disruption, but where the most likely improvements in the production curve occur in the S curve.

[Brian Potter]

Yeah. So S curs is kind of how.

I think about technological progress, which is.

Basically for any given technology, it's kind of performance can be thought of like.

Moving forward on like an S shaped curve.

We're in the very early stages of.

A technology when it's sort of first being invented. At first being worked out, its performance is basically zero or very, very low.

It barely works at all. If it works, fixing one problem just.

Tends to reveal another problem.

So you're putting all this effort into sort of developing it, and it's still.

Really only kind of barely working.

Your sort of effort isn't showing up.

In dramatic improvements in performance because you're trying to figure out how to sort of make this thing work correctly at all.

And so this is sort of the.

Bottom part of the S where as time goes along, performance is like, pretty flat. Maybe it's improving a tiny, tiny bit.

But then eventually you get to a point where sort of things kind of.

Start to click and it's kind of.

Basically working somewhat correctly. And this is sort of as sort.

Of the bottom sort of shoulder of that S. And performance starts to improve more and more rapidly as you put more time and money and investment into it, it works better and better and better. And this is, like I said, the vertical part of that S where a performance of SEC technologies is getting quite a bit better as time goes by and as investment, more investment goes into it.

But then eventually, and this eventually may.

Last a long time, years or decades or even centuries in some cases, you.

Kind of start to run out of improvements.

There's sort of this fundamental limit in what a sort of given arrangement of technology is actually capable of achieving.

And this is when you kind of.

Reach the top part of the S where performance kind of flattens out. And even if you put more and more investment in, you're going to see less and less return in terms of performance.

And usually what happens after that is that if you want to kind of continually see an increase in performance and.

Find ways to do whatever you're doing better or faster or cheaper or whatever.

You need to switch your technology to.

Sort of some fundamentally new concept that requires some new principle that doesn't have the same limits as the technology that you're working, working with currently.

And so an example I give this is in the book is sort of.

The incandescent light bulb, which, you know.

Started out, it took years and years.

And years for this to basically work at all.

Before Edison, there was like, you know.

Dozens of people that had worked to try to invent an incandescent light bulb. They did all sorts of experiments and developments, and it still kind of was very crummy and didn't really work. So you have decades going by, lots of investment, and the performance of a bulb is not really. Still not very good at all.

And then Edison, you know, comes along.

In the 1870s and finally is able to kind of make it click and.

Put things together in a.

In a way that, you know, resolved a lot of these problems. And he finally is able to kind of get performance where it needs to be and then.

Or, you know, good enough that it.

Could be a actual successful product.

And then since, you know, this is sort of the bottom, you know, as the.

The curve kind of starts to turn a little bit. And then as time goes on from Edison, light bulbs get better and better and better. They find out how to make them.

Even cheaper, use less expensive materials, get.

More light per unit of electricity and all this stuff. And that improvement goes on for several.

Decades, but then eventually it kind of tapers off, and then they've kind of. In terms of the performance of the.

Light bulb, a common measure of performance is called luminous efficiency, which is how.

Much light you get for a given.

Amount of energy input.

[Interviewer]

It.

[Brian Potter]

And there's sort of fundamental limits in how efficient a light bulb that uses.

Incandescent can be just to. Just based on sort of the laws of physics. And as time went on, they got closer and closer to this sort of fundamental limit. And so you were kind of at the top part of this S curve, where you weren't really seeing dramatic improvements in performance anymore, just because you had reached the ceiling of what this technology was capable of.

And so for further improvements in sort.

Of luminous efficiency to kind of keep making these bulbs better and better and.

Better, you needed to kind of introduce.

A fundamentally new technology. And so you have different, you know.

Light bulb technologies that didn't have these.

Limits and so could have high luminous efficiency, like fluorescent light bulbs and sodium light bulbs. And, you know, LEDs is what we're using now. And each of these have, like, a higher luminous efficiency than sort of the technology that came before it.

So technology progress tends to look at.

This sort of like series of overlapping S curves where, you know, working on some technology for a while eventually kind of figure out performance improves. You kind of start to approach a ceiling of what this technology can do.

And then you sort of transition to.

Another technology on its own S curve that has, like, a higher performance limit. And you sort of keep climbing these different S curves over time.

[Interviewer]

So you write a given production process reflects a particular product design. Can you expand on that?

[Brian Potter]

Yeah. And this is sort of this sort of basic observation that for any given.

Thing that you're making, you know, the.

Parts and components and things that are going to make that up, every given.

Thing needs to be made by some specific process, some specific series of steps.

And so the specific series of steps that you have of in that process.

You know, the machines that you're using to sort of make whatever it is.

That you're making are going to be.

A reflection of the sort of specific design of whatever it is that you're making, like the specific materials that you've chosen and the shapes and arrangements that they kind of go into.

And a kind of fundamentally important thing.

In like, making a process more efficient.

Is that if you can design whatever.

It is that you're making such that.

It'S easy to produce or, you know, uses less expensive inputs, you can, you.

Know, have a different production process that might be much, much easier to. To do. So this has been a. Historically a pretty important way that that stuff has gotten cheaper over time. There's this whole manufacturing discipline called design for manufacturing and assembly, which basically is.

Like, you have this, you know, maybe.

You have a design for a product.

That works, but it requires, like, expensive.

Operations to sort of produce.

And if you can sort of change.

The design of that a little bit.

You can make it a lot easier.

For the machines that produce that thing to sort of do their job or use different machines and different processes that are cheaper than the ones that you. That you started out with, that you.

Can kind of have the same or very similar product and produce it much more cheaply.

[Interviewer]

Can you discuss the benefits of mechanization?

[Brian Potter]

I mean, yeah, that's pretty straightforward, is that human labor is pretty expensive. And so if you. And it's pretty expensive. And also there are fundamental limits as.

To how fast or accurate that a human can do a given task.

And so if you can, you know.

Recognize some process and take the manual labor component out of it, you can.

A reduce your cost because you can.

Replace it with a machine that maybe.

Is less expensive than the human that.

Requires to do it.

And also then you're no longer coupled.

To the rate limits of how fast that a human can work. And those things are closely coupled together because oftentimes what you have is a machine that is on its own is quite expensive, but because it can work very, very, very quickly, the unit costs of whatever it can produce are much lower.

So an example example of this, we.

Talked earlier about nail manufacturing. And when your nails are manufactured by.

A person, a blacksmith, they can only.

Make those nails so fast, right? There's only so quickly A human can swing a hammer and cut off a piece of iron from a rod.

So there's, you know, the rate that they can produce those nails is quite limited.

But a machine can move much, much faster than a person, right? And So a process that produces nails via machine can make those nails much, much more quickly than a person can.

[Interviewer]

You cite a study that says a dollar spent on value engineering reduced 23 to $300 in construction costs. What is value engineering?

[Brian Potter]

So, yeah, value engineering is pretty. Is conceptually quite similar to kind of.

What I talked about earlier, which is design for manufacturing.

But it's basically like looking at this.

Process or, you know, and the design of what we're making and the steps that we're using to do it and.

Trying to figure out if you can get a identical or a very, very.

Similar outcome for much more cheaply.

And if you can do that, you.

Can cut out lots of expensive parts of your process.

So I give some examples of the book.

But some people are designing some electrical.

Widget, and there's a cover on it.

That has to be affixed by several screws or whatever. And the COVID is maybe quite expensive. And it takes time and effort to install this thing. And they kind of looked at this like, well, why do we even have this cover here? Why is it for. And it turns out that it was there just to basically protect, you know.

Dirt from getting in sort of inside this mechanism. But in the final assembly, this small.

Cover would be protected by an even larger cover.

And it. So this smaller, you know, cover was not actually really needed.

It could be something much, much less.

Expensive and less robust because of the.

It wasn't actually needed to sort of.

Contribute very much to the value of.

This sort of electrical widget. And so they replaced this.

This electrical cover, which maybe costs 11.

Cents, with sort of a thin sheet of plastic that maybe cost 2 cents or something like that, which is a.

Very small amount in absolute dollars. But if you're making a million of.

These things, those sorts of savings can add up quite significantly.

So, yeah, it's basically just looking at.

Whatever it is you're producing, thinking about.

What function it's serving and what it's.

Doing, and thinking about what are the part. Whether the parts that are sort of in this, you know, widget or whatever it is that you're making or the product that you're creating and whether they're actually contributing to the value of it and whether you can sort of get similar value for. For less expenditure.

[State Farm Ad]

This episode is brought to you by State Farm. Listening to this podcast. Smart move. Being financially savvy. Smart move. Another smart move. Having State Farm help you create a competitive price when you choose to bundle home and auto bundling. Just another way to save with a personal price plan. Like a good neighbor, State Farm is there. Prices are based on rating plans that vary by state. Coverage options are selected by the customer. Availability, amount of discounts and savings and eligibility vary by state.

[Lowe's Ad]

Upgrade your laundry routine with a durable and reliable Maytag laundry pair at Lowe's. Like the new Maytag washer and dryer with performance enhanced stain fighting power designed to cut through serious dirt and grime. And what's great is this laundry pair is in stock and ready for delivery when you need it the most. Don't miss out. Shop Maytag in store or online today at Lowe's.

[Interviewer]

What are the benefits and drawbacks of vertical integration?

[Brian Potter]

Yeah, I mean the benefits are that you can often reduce the sort of cost of making something either because you're.

Not having to pay for somebody else's profit margin or you can just be.

More reliably get an input of something.

And so you don't have this downtime.

And then the costs are just that.

You can become sort of big and unwieldy and become less agile in terms of what you're sort of producing. So if you need to change what you're producing or change, you know, how much of it you're producing, having sort of a big vertically integrated process that you control every aspect of it can be, can become a drawback.

And so you, what you kind of tend to see is that as the.

Sort of market, various markets for things shift over time, the value of being vertically integrated shifts as well. And companies kind of tend to change how much they're vertically integrated depending on sort of the specifics of the market that they're operating in.

And so in the book I talk.

A little bit about the car manufacturing industry and how it's kind of changed over time. And you know, at the very, very early stages of, of car manufacturing, they're like not vertically integrated at all. And they're, you know, they're buying all their parts, you know, including like engines and chassis and stuff like that from existing manufacturers. And then, you know, over time some of them get, you know, very, very vertically integrated. A company like Ford in the sort of early 20th century was producing a very, very large fraction of the parts in a given car. And then over time it's kind of continued to shift and from Ford it got less vertically integrated and it's just kind of continued to sort of transform over time, just depending on how the market has evolved.

[Interviewer]

Brian, what's unique about Toyota and Tesla's manufacturing processes? Why are they different in the car industry?

[Brian Potter]

So I don't know if they're necessarily unique per se. Toyota and Tesla are both, both notable for having done a lot to sort of think about what makes a manufacturing.

Process more efficient and trying to sort of drive ways of making those processes more efficient.

Toyota is very famous for its Toyota.

Production system, which then got popularized as lean manufacturing, which is basically trying to.

Sort of minimize the amount of. Of waste in a process, basically.

And they have all these different categories of waste, one of which is excess.

Inventory and sort of designing a process that can sort of efficiently produce more.

Flexibly at lower volumes.

And their methods that they have developed.

Have become kind of quite popular over.

Time and adopted by other manufacturers as well. So they would no longer be like.

Unique in sort of these methods that they sort of initially developed, which were in kind of the 1950s and 60s and 70s.

And Tesla is a little bit similar where they've kind of thought about what.

It takes to sort of make some process more efficient.

How can we change this process of.

What we're making to reduce the cost of whatever they're doing?

And sort of the example that I.

Give in the book is their strategy for using very large aluminum castings for large parts of the car's frame, which.

You know, cut out a large number.

Of separate individual parts from the frame, which made it much cheaper and kind of easier to produce.

And so they were like a real.

Pioneer of these sorts of large castings. But this is another example where they've kind of. Other car manufacturers have kind of started adopting these methods as well. So both Tesla and Toyota were pioneers and kind of exploring ways of how the car manufacturing process could be done more efficiently.

But neither of them are like.

Like neither of them have sort of a monopoly on those methods currently.

[Gregory McNip]

Perfect.

[Interviewer]

I want to move to economies of scale because that's sort of a major theme in your book. You have a chapter devoted to it. And in the end, in your synopsis, you basically conclude, with enough scale and enough accumulated improvements, production costs can be driven down by orders of magnitude. Can you talk about how important economies of scale are and making sure we. I guess a company has the right economies of scale and isn't fooled by false economies of scale as you talk about.

[Brian Potter]

Yeah, economies of scale are quite important. There's a lot of different mechanisms that.

Make stuff, you know, you could make possible to sort of make things more cheaply as you make more of something.

And those mechanisms will often, you know.

Continue to making stuff in very, very, very large volumes.

An important thing that I talk about in the book, related economies of Scale.

Is the learning curve, which is this idea that the more you make of something, the sort of more you find ways to make it more cheaply and more inexpensively.

And sort of the learning curve is this observation that you tend to have.

A constant reduction in costs in percentage terms, every single doubling of production. So when you go from 10 units to 20 units, its cost may be followed by 20% going to 20 units to from 40 units, they will fall by 10%. Again, 40 to 80, they'll fall by 10% or whatever.

So if you can ride down these.

Learning curves for a very, very long time and have a lot of doublings, you can have these really, really dramatic reductions in cost.

Even if the maybe improvements on like.

You know, a per unit level aren't accumulating particularly don't seem like they're accumulating particularly quickly.

So like solar photovoltaics is an example.

Of the solar panels which when they.

Were first came along in the 1950s.

They were incredibly expensive on like a, you know, per amount of electricity they produced.

They were only justified in places where.

There was basically no other option, like satellites, where it was, you know, there was no other option to, you know, no other good option to provide a lot of power to something, you know, up in outer space.

But as we sort of made more.

Of them and found more and more ways to kind of make them more cheaply, the cost of these solar panels fell dramatically over time by huge, huge, huge, huge magnitudes, to the point where they're now one of the cheapest ways of producing electricity until you can use them to provide large utility scale power.

So all these improvements are predicated on.

Being able to produce things in very large volumes. So, yeah, economies of scale have historically kind of been a very important mechanism for stuff to get cheaper over time.

[Interviewer]

Is division of labor, does it always lead to a better production process?

[Brian Potter]

Again, I would not necessarily characterize it like that. Division of labor is kind of an.

Important way that you can sort of make stuff cheaper over time.

And I go into the book, there's.

Various specific effects that are happening there that make that possible.

It's similar to how you can't just.

Move operations into a factory and magically expect it to get more efficient. If there's other constraining factors at work.

You can't just divide your labor up.

More finely and expect it to magically make your process more efficient.

And there's some examples where less division.

Of labor has actually kind of made stuff more efficient. Terada kind of is an example here where they will sort of have.

One of the ways that they have.

Sort of structured their operations to have workers able to attend multiple machines in a single cell of operations and doing multiple tasks.

And so division of labor is important, but again, it's important to kind of.

Look at the specific constraints that you're sort of operating under and what, you know, when those sorts of things will have improvements and when they won't.

And I talk about this, you know.

Go into the specifics in the book.

[Interviewer]

Yep. Why did the standard shipping container lead to such a dramatic reduction in manufacturing costs?

[Brian Potter]

I would not necessarily say the shipping container itself led to a drama. Well, I guess how I characterize this is that it basically shipping container lowered.

The cost of all, like global transportation, which made it more cost effective to sort of ship things very long distances.

And what that did was allow sort.

Of more concentration in manufacturing operations. And so you could have a factory producing some small part of some process in very large volumes and sort of ship those things all over the world.

[Interviewer]

What is variation in the production process and how can we reduce it?

[Brian Potter]

It.

So variation is basically just when the.

Process doesn't do exactly what you want.

It to do, you know, or, you know, the, the it's, you know, it's.

Producing something that's like slightly too big or slightly too small or slightly faster, slightly slower than it should. And when you have variation in a process, it ultimately kind of corrodes it.

You're ultimately sort of producing, you know, not producing as.

As fast or as that.

You could.

Different parts of the process aren't aligned, or you're producing stuff that you have to either throw away or rework. And so it basically adds cost to your production process.

And you basically reduce it by finding, you know, any given variation is a.

Result of some specific thing happening in your process.

And if you can find those causes.

And eliminate them, you can kind of reduce the variation as well. And so there's this industrial improvement system.

They invented at Bell Labs in the.

First half of the 20th century called statistical Process Control, which is all about investigating these. These causes of variation and eliminating them.

[Interviewer]

What are bundle and change of improvements and how can they impact the production process?

[Brian Potter]

Yeah, so this is kind of the idea that oftentimes an improvement to a process will either either come bundled with several improvements at once or, or it will basically trigger additional improvements. And so an example I give of this in the book is with economies.

Of scale, where if you make more.

Of something and you find ways to.

Sort of produce it more cheaply, if.

You'Re now selling it more cheaply, the demand for that thing is going to rise and then you can give you.

An opportunity to sort of produce even more. And then again, now that you're producing even more, your costs can fall even further.

So you can have this sort of.

Virtuous cycle where every time you figure out how to lower your cost and you can sell even more of it and get greater economies of scale and lower your costs even more.

But then this sort of virtual cycle.

Can also work in reverse into a vicious cycle where if you sort of.

Lose output and your cost rise and.

Then your market dwindles even more because people are interested in buying at the higher cost, and then you sort of lose, you know, more economies of scale and your costs rise even more.

And so, yeah, there are conditions and.

Cases where, like these bundled, you know, improvements can sort of cascade and one can lead to another another.

And I talk about a little bit in the book, and this is a.

Sort of a way that the Toyota.

Production system is structured where they've sort.

Of arranged it so one improvement leads to sort further improvements, which is why they are so insistent that the sort of their whole system needs to all be implemented together, because you can't just implement one small part of it and expect it to work.

[Interviewer]

The impact of the assembly line and mass production cannot be overstated in terms of the nature of consumption and standard of living. Was this a dramatic change or were these processes and approaches, did they appear earlier in terms of the concepts decades before the assembly line came about? How was that? Like it was a gradual or an instant change. It felt like it was building on top of previous concepts.

[Brian Potter]

Yeah, it was building on previous concept. These ideas, you know, the assembly line you can kind of think about and you know, it's so closely associated with.

Ford and the Model T and car manufacturing.

But the assembly line is really this idea of a continuum continuous process applied.

To a very large complex manufactured good. And for assembly line is just going again smoothly and swiftly, one step after another, with as little waiting and downtime and delays as possible.

But you kind of see these similar.

Ideas, these similar sort of continuous manufacturing processes exist much earlier, kind of in the late 19th century. You start seeing them for like the manufacture of simple manufactured goods like matches or cigarettes or stuff like that, where you have these machines that could just again do every step of the cigarette production process, including putting them into packs.

So really what's notable about the assembly line is that now it's kind of.

Extending this concept of like a continuous manufacturing process into a very, very complicated manufactured good with like thousands and thousands of parts in it.

[Interviewer]

Last question. Can you give us your take on the extent of AI's impact on both high volume manufacturing and those areas historically resistant to efficiency improvements like education, healthcare, housing?

[Brian Potter]

Yeah, I talk about this in the book where we've, you know, for things that we can make a very, very.

In very, very large volumes, you know, very, some sort of very, very repetitive process.

Those things are things that we can.

Make like incredibly inexpensively. And the benefits from sort of some dramatic improvement in say AI or robotics are maybe going to be less pronounced there.

There's something, a sort of common thing that you see on sort of in.

Twitter areas where people talk about manufacturing.

Is they'll show some, you know, really.

Really fast, efficient like industrial process that's.

You know, something being produced with like dedicated automation equipment.

And it goes like really, really quick.

And then basically saying, look, there's all this enthusiasm about things like humanoid robots, but a humanoid robot would be so much worse at doing this task than.

This specific dedicated equipment that can already do it really, really, really quickly and efficiently.

So where I think you're going to.

See the benefits from these types of improvements in AI and robotics and all these sorts of similar technologies in art.

Is in kind of these areas that have not seen really huge improvements by.

Some process that is repetitive and works the same way over and over again. So things like car repair and things like medical care and things like education.

Where it's expensive in part because it's.

Tailored to a specific set of circumstances that you can't necessarily.

Work out exactly.

What you need to do ahead of time until you're kind of, of faced with it.

But if you have sort of an AI that can respond flexibly based on environmental contexts quickly and cheaply, that will.

Maybe have, you'll kind of start to see some of these improvements introduce sort.

Of automation in areas that have been.

A little bit historically resistant to it. Great.

[Interviewer]

That concludes our interview. Again, the book is the Origins of Efficiency by Brian Potter.

[Gregory McNip]

Brian, thank you so much for your time in writing. Such an interesting and insightful.

[Brian Potter]

Yeah, thanks for having me.