I was pretty deeply involved in mapping through vertical markets when I was working for the CTO at Boeing in Phantom works from 1997-2004. And graduate work I did on vertical vs horizontal innovation at MIT 1995-1997 when Clay wrote his papers before Innovators Dilemma. You might also want to look at Utterback's work around product vs process innovation.
I do agree with Thiels view of these markets, and SpaceX's success. But also suggest that the time for vertical or horizontal integration is also correlated to markets latent needs. In Clay's context, if the market is seeking new features/product, vertical takes off. If price, horizonal takes off. The chart in this twitter post shows the general factors that drive the shift. https://x.com/jcarterwil/status/1828583162739409255
SpaceX is a novel success. The effort to build SpaceX though is akin to many things done before in aerospace. I was on the team that developed some of the manufacturing technology For the F/A-18 E/F in the early 90s that is key to both SpaceX and Tesla. The new leaders build on the technology from the predecessor. Boeing is a vertical system integrator. For all its recent troubles, there are 100+ products in its portfolio that are breakthroughs in vertical integration. F15, 747, DC3, F18, 777, Harpoon, JDAM, B52, etc. 737, launched in the late 60s, has flown 6.5 Billion people with 5,500 deaths.
Developing these systems is a level of complexity few people comprehend. Maybe global oil production is similar.
Boeing's recent trouble is in part a failure in 2004 timeframe to move back to product innovation. SpaceX will hit the same wall. 1st generation leadership (Musk, Bill Boeing, James McDonnell, Palmer Luckey) are vertical system integrators. Then the 2nd generation shows up. Who are process innovators. They reduce cost. They modularize. They subcontract. All the time looking awesome because they made the 1st generation lower cost and better. Then the 3rd generation shows up. They are product innovators, or losers. Creative Destruction takes off. The incumbent resets or gets destroyed.
In all cases the survivor is normally closely attached to customers. Thompsons aggregation theory is an important metaphor for vertical integrators to stay close to their customers. Boeing's process optimizers became distant to those customers.
TL;DR
Markets go vertical to innovate product, horizontal to reduce cost and scale. Back and forth over a 40-50 year cycle.
In markets needing disruption, (Ag, Defense, Energy, Health) there is a SpaceX like vertical optimization required. Thompson/Thiel lean on this from a internet/user context. I am living it in a move commodities around in Ag/Food context. Similar but different.
Your article is very helpful helping other people see a glimpse into the complexty. And a path to the fix. These are extrodinaryily difficult problems seeking a market answer, but a lot of twisting to get there. But the market answer is still the long term correct path.
Hands down. This essay (and the link to the Base deep dive) is one of the most interesting and eye opening pieces I read in a really long time. A huge thanks for sharing your ideas and for crafting this text. Shared it with lots of different persons.
This is also very relevant to clean energy and other energy infrastructure companies. Vertically integrated providers that are able to own supply chain and create high quality products are already showing more value than those who aren't.
Right. If you look at industry verticals you could make the case that cars/space, and energy are already being disrupted. What is interesting about those industries is the elasticity of demand is steeper than other industries. There are not a lot of substitutes. Where else can you go? Farming and agribusiness seems to be one to me. It is hard, and logistical coordination is necessary. Big huge incumbents (Cargill, ADM etc) and the meat business is very different than the grain business or fruit and vegetable business. Profit margins are tighter though. Air travel seems to be another one. Plenty of others.
I am deep now in Ag. This industry is straining between highly commoditized horzontal (i.e. Cargil) and new vertical markets. Shifting the market from low cost calories to affordable nutrition. The horizontal market serves the 1.7T we spend on food in the US. The vertical market will fix the 1.9T we spend on the healthcare cost of poor nutrition.
Very interesting - I think Terraform Industries' work is in this vein. They didn't have any major inventions, but they customized and assembled their system to create a business that doesn't exist yet (selling a machine that turns sunlight into natural gas).
These types of companies feel like a better fit for VC, especially since software companies have become so cheap to start & build.
I feel like private equity has had a long run in rolling up SMB's into more vertically integrated businesses. BUT, it's usually a financial-centric approach vs. one pushing frontier tech & trying to ~innovate~.
Hi Packy - in your “Technological Provenness” graphic, what’s the horizontal axis?
I’m trying to understand why Vertical Integrators are off on the right and Deep Tech’s in the middle. Apologies if it’s in the essay and I just missed it.
Got it, thanks - I was trying to figure out if it was relative amount of current capex, or complexity or risk, or something else. Appreciate the clarification!
Super interesting piece. got me thinking about Porter's Five Forces and other things. One hard part about vertically integrating from the ground up in a heavy industry is getting the startup capital; and trying to figure out where to start. The lead times are incredibly long-you don't build a car company and have revenue the first number of years etc. Existing VCs might be so conditioned to software, they and their LPs might find the jump to vertically integrated hardware impossible to fathom. The risk is also higher given the dollars needed to get from point A to B, but if you can have a "monopoly" the profit stream lasts until you get disrupted.
Great article.
I was pretty deeply involved in mapping through vertical markets when I was working for the CTO at Boeing in Phantom works from 1997-2004. And graduate work I did on vertical vs horizontal innovation at MIT 1995-1997 when Clay wrote his papers before Innovators Dilemma. You might also want to look at Utterback's work around product vs process innovation.
I do agree with Thiels view of these markets, and SpaceX's success. But also suggest that the time for vertical or horizontal integration is also correlated to markets latent needs. In Clay's context, if the market is seeking new features/product, vertical takes off. If price, horizonal takes off. The chart in this twitter post shows the general factors that drive the shift. https://x.com/jcarterwil/status/1828583162739409255
SpaceX is a novel success. The effort to build SpaceX though is akin to many things done before in aerospace. I was on the team that developed some of the manufacturing technology For the F/A-18 E/F in the early 90s that is key to both SpaceX and Tesla. The new leaders build on the technology from the predecessor. Boeing is a vertical system integrator. For all its recent troubles, there are 100+ products in its portfolio that are breakthroughs in vertical integration. F15, 747, DC3, F18, 777, Harpoon, JDAM, B52, etc. 737, launched in the late 60s, has flown 6.5 Billion people with 5,500 deaths.
Developing these systems is a level of complexity few people comprehend. Maybe global oil production is similar.
Boeing's recent trouble is in part a failure in 2004 timeframe to move back to product innovation. SpaceX will hit the same wall. 1st generation leadership (Musk, Bill Boeing, James McDonnell, Palmer Luckey) are vertical system integrators. Then the 2nd generation shows up. Who are process innovators. They reduce cost. They modularize. They subcontract. All the time looking awesome because they made the 1st generation lower cost and better. Then the 3rd generation shows up. They are product innovators, or losers. Creative Destruction takes off. The incumbent resets or gets destroyed.
In all cases the survivor is normally closely attached to customers. Thompsons aggregation theory is an important metaphor for vertical integrators to stay close to their customers. Boeing's process optimizers became distant to those customers.
TL;DR
Markets go vertical to innovate product, horizontal to reduce cost and scale. Back and forth over a 40-50 year cycle.
This is incredible context. I really appreciate it. I’ll check out Utterback’s work this week.
You captured the essential complexities well.
In markets needing disruption, (Ag, Defense, Energy, Health) there is a SpaceX like vertical optimization required. Thompson/Thiel lean on this from a internet/user context. I am living it in a move commodities around in Ag/Food context. Similar but different.
Your article is very helpful helping other people see a glimpse into the complexty. And a path to the fix. These are extrodinaryily difficult problems seeking a market answer, but a lot of twisting to get there. But the market answer is still the long term correct path.
Hands down. This essay (and the link to the Base deep dive) is one of the most interesting and eye opening pieces I read in a really long time. A huge thanks for sharing your ideas and for crafting this text. Shared it with lots of different persons.
Thanks again. What a piece of work!
Happy to hear that and thank you for sharing!
This is also very relevant to clean energy and other energy infrastructure companies. Vertically integrated providers that are able to own supply chain and create high quality products are already showing more value than those who aren't.
Right. If you look at industry verticals you could make the case that cars/space, and energy are already being disrupted. What is interesting about those industries is the elasticity of demand is steeper than other industries. There are not a lot of substitutes. Where else can you go? Farming and agribusiness seems to be one to me. It is hard, and logistical coordination is necessary. Big huge incumbents (Cargill, ADM etc) and the meat business is very different than the grain business or fruit and vegetable business. Profit margins are tighter though. Air travel seems to be another one. Plenty of others.
I am deep now in Ag. This industry is straining between highly commoditized horzontal (i.e. Cargil) and new vertical markets. Shifting the market from low cost calories to affordable nutrition. The horizontal market serves the 1.7T we spend on food in the US. The vertical market will fix the 1.9T we spend on the healthcare cost of poor nutrition.
Very interesting - I think Terraform Industries' work is in this vein. They didn't have any major inventions, but they customized and assembled their system to create a business that doesn't exist yet (selling a machine that turns sunlight into natural gas).
https://terraformindustries.wordpress.com/2024/06/24/how-terraform-navigated-the-idea-maze/
Yes! Big Terraform fan.
Really enjoyed this article - thank you for sharing!
These types of companies feel like a better fit for VC, especially since software companies have become so cheap to start & build.
I feel like private equity has had a long run in rolling up SMB's into more vertically integrated businesses. BUT, it's usually a financial-centric approach vs. one pushing frontier tech & trying to ~innovate~.
Agreed!
Hi Packy - in your “Technological Provenness” graphic, what’s the horizontal axis?
I’m trying to understand why Vertical Integrators are off on the right and Deep Tech’s in the middle. Apologies if it’s in the essay and I just missed it.
Poorly drawn graph - that's just the line above which it's proven and below which it's not
Got it, thanks - I was trying to figure out if it was relative amount of current capex, or complexity or risk, or something else. Appreciate the clarification!
Super interesting piece. got me thinking about Porter's Five Forces and other things. One hard part about vertically integrating from the ground up in a heavy industry is getting the startup capital; and trying to figure out where to start. The lead times are incredibly long-you don't build a car company and have revenue the first number of years etc. Existing VCs might be so conditioned to software, they and their LPs might find the jump to vertically integrated hardware impossible to fathom. The risk is also higher given the dollars needed to get from point A to B, but if you can have a "monopoly" the profit stream lasts until you get disrupted.
Yes! I actually had a Porter's Five Forces chart that I removed but might show back up in Part II...
Really enjoyed reading this! Going to check out part 2 and 3 soon
excellent.