6 ms·

I think there's something missing from this analysis - YC companies might duplicate one another's products but that doesn't say anything at all about their targ

I think there's something missing from this analysis - YC companies might duplicate one another's products but that doesn't say anything at all about their target markets, route to market, product focus, etc. As an example, my first startup was a requirements management app that, on paper at least, was a copy of IBM DOORS. Except DOORS is a massive enterprise app targeted at companies who are building a new airplane that has 250,000 technical requirements that need fully traceability, and my app was aimed at small businesses doing software projects with 1000 requirements. I was not competing with IBM; I was trying to apply the value of that app to much smaller projects (the end goal might have been to become a competitor one day but we failed long before then.) You can't just say "These two products do the same thing so the companies are dupes." It's far more complicated than that.

meiraleal 2y ago
That's not the case for most dev tools YC startup, they are targeting the same HN users/ other YC startups (trying to sell shovels)
- fidotron 2y ago
  The great suspicion with YC is what proportion of YC companies have all their customers being a mix of other YC companies and those with shared investors? There is a real danger with current era Bay Area tech that it is just a game of musical chairs played with money, with remarkably little external value being generated.
  - cudgy 2y ago
    > There is a real danger with current era Bay Area tech that it is just a game of musical chairs played with money, with remarkably little external value being generated. So, you don’t think that is already the case?
    - fidotron 2y ago
      It was nothing like as bad as this 10 years ago, no. Just look at the state of successful exits - it is awful. This is not the same as an ecosystem producing Sun, HP, Apple or even Google, which all had enormous positive externalities.
      - meiraleal 2y ago
        Yes it was like this for Apple, Microsoft, Google, Facebook. The difference is that at that time the genius youngster (which is just the son of some already successful people) creating startups were cool. Now nobody wants to make or watch a movie about Sam Altman and OpenAI.
        HWR_14 2y ago
        Did anyone think Zuckerberg was cool? Even the movie about him seems to have made him seem more antisocial and nerdy than he really is. (Based on 3rd party reports about him)
      - jjulius 2y ago
        Ten years ago was 2014 - Sun, HP, Apple and Google were all very much "old hat" and entrenched. Hell, HP was in the process of significantly cutting back it's business and eliminating jobs[1] (34,000 lost jobs in 2013 and ~16,000 in 2014 is an interesting "positive externality" lol). There was no ecosystem producing them, they were the people who were either running SV, or realizing that their time in the sun was coming to an end. [1]https://en.wikipedia.org/wiki/Hewlett-Packard https://en.wikipedia.org/wiki/Hewlett-Packard
  - spacebanana7 2y ago
    This can be a good thing. One of the hardest things about building new software products is getting initial customers to discover your bugs, missed requirements etc and also add credibility. As an enterprises SaaS buyer, I'd much rather use a code documentation tool that'd been circle jerked around a YC batch a few times compared to one with no prior customers.
hiAndrewQuinn 2y ago
Incidentally, if anyone else wants to reinvent IBM DOORS for ~1000-requirements fields, email me. There's a vast market out there for combined hardware/software shops that are in the no man's land of "not designing an entire literal train from top to bottom" and "not just hiding a Raspberry Pi in a trench coat".
- MichaelZuo 2y ago
  Why can’t IBM just sell a watered down “DOORS-lite”?
  - onion2k 2y ago
    They could, but it's IBM so any 'lite' version wouldn't stay lite for very long. They mostly excel at making huge, complicated things to sell to other huge, complicated businesses. They aren't very interested in, or very good at, making small things they need to sell lots of (and, to their credit, they recognize that and stick to what they're good at.)
    - MichaelZuo 2y ago
      Well they clearly have the potential to sell the same but without any of the add ons, or most of the support or guarantees, and with functionality restricfed. Wouldn’t they do so once they notice there is a serious challenger?
      - lmm 2y ago
        That's the famous innovator's dilemma. It's very hard for a company to sell a product that undercuts a product that makes a lot of revenue for them, even if it's the "right" play. So that wouldn't be a risk I'd worry about in practice.
        MichaelZuo 2y ago
        Well in this case it wouldn’t be undercutting, since the lite version would be totally unusuable for a big complex company.
mrandish 2y ago
Yes, I agree and would add other reasons. It's always seemed obvious to me that YC would invest in multiple startups working on the same problem space for these reasons: First, compared to a traditional VC, YC is micro-investing in far more startups and doing so faster and more frequently. Second, YC highly prioritizes founder/team quality over ideas but teams come to YC by pitching an idea. I just read an article yesterday citing evidence that in the last 10-15 years the diversity of startup ideas has declined significantly. This makes sense. Enormous amounts of information about what other very early stage startups are doing as well as which areas are currently being funded is readily available in near real-time. As a serial entrepreneur who did my first startup in the 80s, second in the 90s and third in the early 00s, I can attest we had nothing like this visibility in earlier eras. Ideas which are getting funded and which sound like good ideas in hot areas are obviously going to influence founder idea selection and cause clustering. Third, YC knows that many startup teams will pivot away from their initial idea once they engage with real customers in real markets. So much so, YC considers it a sign of a smart team working well together. However, like all VCs, at any given time YC has broad sector themes it considers especially ripe for various reasons ranging from new technologies enabling disruption to expected high growth in an emerging sector. Fourth, is the reason you identified. Having a good enough high-concept idea is necessary but far from sufficient for startup success. In addition to the execution details like go to market that you mentioned there's also timing. Even being 12 months earlier or later can make a difference toward making it. Finally, there's the luck factor. While it's true that quality teams are better able to maximize good luck when it happens and also have a tendency to increase their overall odds too, luck still matters. In my successes we had to get a lot of things right but there were also three or four serendipitous things that made a big difference at important moments. The only way for YC to solve for both the micro-timing and luck factors is betting on multiple similar startups around the same time.
- ribadeo 2y ago
  I wonder how successful market pivots are, on average. Seems like flailing to me. How can one take money to achieve x, build plans and hire people to achieve x, and then do y instead and not fall down in the necessary interval, unless the y product happens to be a side effect of x?

meiraleal 2y ago

That's not the case for most dev tools YC startup, they are targeting the same HN users/ other YC startups (trying to sell shovels)

fidotron 2y ago
The great suspicion with YC is what proportion of YC companies have all their customers being a mix of other YC companies and those with shared investors? There is a real danger with current era Bay Area tech that it is just a game of musical chairs played with money, with remarkably little external value being generated.
- cudgy 2y ago
  > There is a real danger with current era Bay Area tech that it is just a game of musical chairs played with money, with remarkably little external value being generated. So, you don’t think that is already the case?
  - fidotron 2y ago
    It was nothing like as bad as this 10 years ago, no. Just look at the state of successful exits - it is awful. This is not the same as an ecosystem producing Sun, HP, Apple or even Google, which all had enormous positive externalities.
    - meiraleal 2y ago
      Yes it was like this for Apple, Microsoft, Google, Facebook. The difference is that at that time the genius youngster (which is just the son of some already successful people) creating startups were cool. Now nobody wants to make or watch a movie about Sam Altman and OpenAI.
      - HWR_14 2y ago
        Did anyone think Zuckerberg was cool? Even the movie about him seems to have made him seem more antisocial and nerdy than he really is. (Based on 3rd party reports about him)
    - jjulius 2y ago
      Ten years ago was 2014 - Sun, HP, Apple and Google were all very much "old hat" and entrenched. Hell, HP was in the process of significantly cutting back it's business and eliminating jobs[1] (34,000 lost jobs in 2013 and ~16,000 in 2014 is an interesting "positive externality" lol). There was no ecosystem producing them, they were the people who were either running SV, or realizing that their time in the sun was coming to an end. [1]https://en.wikipedia.org/wiki/Hewlett-Packard https://en.wikipedia.org/wiki/Hewlett-Packard
- spacebanana7 2y ago
  This can be a good thing. One of the hardest things about building new software products is getting initial customers to discover your bugs, missed requirements etc and also add credibility. As an enterprises SaaS buyer, I'd much rather use a code documentation tool that'd been circle jerked around a YC batch a few times compared to one with no prior customers.

hiAndrewQuinn 2y ago

Incidentally, if anyone else wants to reinvent IBM DOORS for ~1000-requirements fields, email me. There's a vast market out there for combined hardware/software shops that are in the no man's land of "not designing an entire literal train from top to bottom" and "not just hiding a Raspberry Pi in a trench coat".

MichaelZuo 2y ago
Why can’t IBM just sell a watered down “DOORS-lite”?
- onion2k 2y ago
  They could, but it's IBM so any 'lite' version wouldn't stay lite for very long. They mostly excel at making huge, complicated things to sell to other huge, complicated businesses. They aren't very interested in, or very good at, making small things they need to sell lots of (and, to their credit, they recognize that and stick to what they're good at.)
  - MichaelZuo 2y ago
    Well they clearly have the potential to sell the same but without any of the add ons, or most of the support or guarantees, and with functionality restricfed. Wouldn’t they do so once they notice there is a serious challenger?
    - lmm 2y ago
      That's the famous innovator's dilemma. It's very hard for a company to sell a product that undercuts a product that makes a lot of revenue for them, even if it's the "right" play. So that wouldn't be a risk I'd worry about in practice.
      - MichaelZuo 2y ago
        Well in this case it wouldn’t be undercutting, since the lite version would be totally unusuable for a big complex company.

mrandish 2y ago

Yes, I agree and would add other reasons. It's always seemed obvious to me that YC would invest in multiple startups working on the same problem space for these reasons: First, compared to a traditional VC, YC is micro-investing in far more startups and doing so faster and more frequently. Second, YC highly prioritizes founder/team quality over ideas but teams come to YC by pitching an idea. I just read an article yesterday citing evidence that in the last 10-15 years the diversity of startup ideas has declined significantly. This makes sense. Enormous amounts of information about what other very early stage startups are doing as well as which areas are currently being funded is readily available in near real-time. As a serial entrepreneur who did my first startup in the 80s, second in the 90s and third in the early 00s, I can attest we had nothing like this visibility in earlier eras. Ideas which are getting funded and which sound like good ideas in hot areas are obviously going to influence founder idea selection and cause clustering. Third, YC knows that many startup teams will pivot away from their initial idea once they engage with real customers in real markets. So much so, YC considers it a sign of a smart team working well together. However, like all VCs, at any given time YC has broad sector themes it considers especially ripe for various reasons ranging from new technologies enabling disruption to expected high growth in an emerging sector. Fourth, is the reason you identified. Having a good enough high-concept idea is necessary but far from sufficient for startup success. In addition to the execution details like go to market that you mentioned there's also timing. Even being 12 months earlier or later can make a difference toward making it. Finally, there's the luck factor. While it's true that quality teams are better able to maximize good luck when it happens and also have a tendency to increase their overall odds too, luck still matters. In my successes we had to get a lot of things right but there were also three or four serendipitous things that made a big difference at important moments. The only way for YC to solve for both the micro-timing and luck factors is betting on multiple similar startups around the same time.

ribadeo 2y ago
I wonder how successful market pivots are, on average. Seems like flailing to me. How can one take money to achieve x, build plans and hire people to achieve x, and then do y instead and not fall down in the necessary interval, unless the y product happens to be a side effect of x?