First of all, well-written essay. Thank you so much for sharing your thoughts.
It resonates with me, but also I think the direction is ... let's say inefficient. The core thesis - that deep knowledge beats emotional trading and investment snacking - sounds right. Like saying you should eat your vegetables and exercise regularly. Hard to argue with that.
But, there's a small logical problem here. The essay basically says "Step 1: Pick great companies. Step 2: Study them deeply. Step 3: Profit." Except... Step 1 is doing all the heavy lifting. If you already know how to identify truly great companies, you could skip steps 2 and 3 and still do pretty well.
The real challenge in investing isn't knowing everything about Apple's supply chain or Costco's membership renewal rates. It's developing the pattern recognition to spot the next great business before everyone else does. And for that, you need exposure to lots of companies - both the winners and the losers. And everything in between.
Think of it like machine learning (because every analogy these days has to involve AI). You don't train an algorithm by showing it the same 10 perfect examples over and over. You feed it thousands of varied cases so it learns to distinguish signal from noise. Your investing brain works the same way.
That's why I prefer what we might call horizontal scaling rather than vertical scaling. Instead of going 100% deep on 10 companies, try going 20% deep on 1000 companies or more. You might say "Your math is not working there." Ah, actually it does.
When we are talking about understanding companies, it's not like filling up a glass of water, where each minute of research adds the same amount of knowledge. It's more like trying to get to the center of the Earth. The first few kilometers are pretty straightforward - you learn the basics from earnings reports, investor presentations, industry analysis. But then it gets exponentially harder. Each extra bit of insight requires more effort than the last, and eventually you're trying to drill through magma-heated rock just to squeeze out one more detail about a company's third-tier supplier relationships. And by the way, you'll never get to the center. Just like no one ever understands 100% of any company.
That's why the math of "20% effort on 1000 companies" actually makes more sense than it sounds. And depending on your efficiency, that number can grow or shrink.
So when we talk about horizontal scaling, we're really talking about living in that sweet spot of the effort-to-insight curve. Get the key patterns, understand the business model, grasp the competitive dynamics, and then move on. You'll develop much better pattern recognition. Plus, you'll have a wider pool of candidates in order to spot something promising.
The essay's ideas about research checklists and mental schemas are solid. But the Apple example, even though it makes sense in the Apple scope, it misses the point a bit. What do I mean? A truly useful mental framework shouldn't be company-specific - it should help you understand patterns across many different businesses. What makes some companies get stronger as they grow while others plateau? Why do some benefit from scale while others get crushed by complexity?
Those are the kinds of patterns worth studying. Because knowing everything about Apple might help you trade Apple stock. But understanding how great businesses work helps you spot the next Apple before everyone else does.
I do think this essay nails something important: you need both knowledge and patience to be a great investor. The twin engines framework is solid. And the idea that some companies actually get stronger as they grow - while passing those benefits to customers - that's a powerful insight.
But maybe the real magic happens when you combine approaches: Use those research checklists and mental frameworks to efficiently study hundreds of companies. Let the patterns emerge naturally. Then when you spot something truly special, sure, dive deep. Because at that point, you're pressure-testing a thesis built from analyzing hundreds of others. In essence, that's how you go through step 1, the "finding the great companies".
And that's really what investing is about, isn't it? Building the judgment to see what others don't. Yet.
Hi @Silba thanks for taking the time to read the article and for your thoughtful comment. I agree with many things you mention there. Finding the "sweet spot of the effort-to-insight curve" if something I think a lot about and try get better at. There are decreasing marginal returns in allocating time to research a company, 100%. I'd would argue though that different companies have different "depths" (alluding to your drilling metaphor), and that sweet spot may be reached by few investors in complicated setups or businesses.
Of course, we are generalising here. Every company is unique, and every investor has a unique background and “model” in his brain. So “your mileage may vary“, I agree. We are essentially trying to find an abstraction that can work well for most cases.
What's striking about this essay is that whilst explicitly prescribing a methodology for spotting good management practices (and therefore potentially excellent companies) you implicitly illuminate to business owners and founders the lodestar that might guide them toward said practices. Engaging and enjoyable on both levels. I look forward to the next piece.
Build that list of companies that could develop this kind of advantages and follow them over the years. It takes work but I think it’s a valid exercise. Start with the As to the Zs. Early indicators? Its hard but I focus on any type
of customer feedback, reviews or experience sharing I can find.
Polymath Investor is one of my favourite Substack writers. Brilliantly articulated and thought-provoking content.
Thanks James~! 🙏🙏
First of all, well-written essay. Thank you so much for sharing your thoughts.
It resonates with me, but also I think the direction is ... let's say inefficient. The core thesis - that deep knowledge beats emotional trading and investment snacking - sounds right. Like saying you should eat your vegetables and exercise regularly. Hard to argue with that.
But, there's a small logical problem here. The essay basically says "Step 1: Pick great companies. Step 2: Study them deeply. Step 3: Profit." Except... Step 1 is doing all the heavy lifting. If you already know how to identify truly great companies, you could skip steps 2 and 3 and still do pretty well.
The real challenge in investing isn't knowing everything about Apple's supply chain or Costco's membership renewal rates. It's developing the pattern recognition to spot the next great business before everyone else does. And for that, you need exposure to lots of companies - both the winners and the losers. And everything in between.
Think of it like machine learning (because every analogy these days has to involve AI). You don't train an algorithm by showing it the same 10 perfect examples over and over. You feed it thousands of varied cases so it learns to distinguish signal from noise. Your investing brain works the same way.
That's why I prefer what we might call horizontal scaling rather than vertical scaling. Instead of going 100% deep on 10 companies, try going 20% deep on 1000 companies or more. You might say "Your math is not working there." Ah, actually it does.
When we are talking about understanding companies, it's not like filling up a glass of water, where each minute of research adds the same amount of knowledge. It's more like trying to get to the center of the Earth. The first few kilometers are pretty straightforward - you learn the basics from earnings reports, investor presentations, industry analysis. But then it gets exponentially harder. Each extra bit of insight requires more effort than the last, and eventually you're trying to drill through magma-heated rock just to squeeze out one more detail about a company's third-tier supplier relationships. And by the way, you'll never get to the center. Just like no one ever understands 100% of any company.
That's why the math of "20% effort on 1000 companies" actually makes more sense than it sounds. And depending on your efficiency, that number can grow or shrink.
So when we talk about horizontal scaling, we're really talking about living in that sweet spot of the effort-to-insight curve. Get the key patterns, understand the business model, grasp the competitive dynamics, and then move on. You'll develop much better pattern recognition. Plus, you'll have a wider pool of candidates in order to spot something promising.
The essay's ideas about research checklists and mental schemas are solid. But the Apple example, even though it makes sense in the Apple scope, it misses the point a bit. What do I mean? A truly useful mental framework shouldn't be company-specific - it should help you understand patterns across many different businesses. What makes some companies get stronger as they grow while others plateau? Why do some benefit from scale while others get crushed by complexity?
Those are the kinds of patterns worth studying. Because knowing everything about Apple might help you trade Apple stock. But understanding how great businesses work helps you spot the next Apple before everyone else does.
I do think this essay nails something important: you need both knowledge and patience to be a great investor. The twin engines framework is solid. And the idea that some companies actually get stronger as they grow - while passing those benefits to customers - that's a powerful insight.
But maybe the real magic happens when you combine approaches: Use those research checklists and mental frameworks to efficiently study hundreds of companies. Let the patterns emerge naturally. Then when you spot something truly special, sure, dive deep. Because at that point, you're pressure-testing a thesis built from analyzing hundreds of others. In essence, that's how you go through step 1, the "finding the great companies".
And that's really what investing is about, isn't it? Building the judgment to see what others don't. Yet.
Hi @Silba thanks for taking the time to read the article and for your thoughtful comment. I agree with many things you mention there. Finding the "sweet spot of the effort-to-insight curve" if something I think a lot about and try get better at. There are decreasing marginal returns in allocating time to research a company, 100%. I'd would argue though that different companies have different "depths" (alluding to your drilling metaphor), and that sweet spot may be reached by few investors in complicated setups or businesses.
Of course, we are generalising here. Every company is unique, and every investor has a unique background and “model” in his brain. So “your mileage may vary“, I agree. We are essentially trying to find an abstraction that can work well for most cases.
What's striking about this essay is that whilst explicitly prescribing a methodology for spotting good management practices (and therefore potentially excellent companies) you implicitly illuminate to business owners and founders the lodestar that might guide them toward said practices. Engaging and enjoyable on both levels. I look forward to the next piece.
Great article. Any thoughts on spotting these moat and mindsets earlier? Thanks.
Build that list of companies that could develop this kind of advantages and follow them over the years. It takes work but I think it’s a valid exercise. Start with the As to the Zs. Early indicators? Its hard but I focus on any type
of customer feedback, reviews or experience sharing I can find.
Wonderful article!
Thanks John!