4 ms·

According to Paul Graham, successful startups (weighted by eventual market cap) are ideas that: 1) Sound bad 2) But are actually good. So by selecting on ide

According to Paul Graham, successful startups (weighted by eventual market cap) are ideas that: 1) Sound bad 2) But are actually good. So by selecting on ideas that sound good on first glance, you're already selecting for startups that won't comprise the majority of the success mass.

marcinzm 2y ago
Obviously good ideas are ones that a large company department will also easily get funding for. That increases the competition a lot.
- api 2y ago
  Yeah but large companies often execute badly or develop solutions that are many many orders of magnitude more complex and expensive than needed. Large companies also generally push for all their products to work together, limiting their ability to build clean standalone things.
- bcrosby95 2y ago
  Not true when it comes to new tech. That's why anytime there's a new piece of tech there's a gold rush of new companies born. Internet, smart phones, social platforms, etc.
jszymborski 2y ago
In fairness, I think pg mostly concerns himself with start-ups whose goals are to become a unicorn or be acquired by one. I think there are plenty of boring, sustainable businesses built on good, logical sounding ideas that have found their niche in the industries they find themselves in.
nprateem 2y ago
Ah so post-hoc categorisation. That's really useful. I suppose it's blindingly obvious to VCs the bad ideas that are bad vs the bad ideas that are good, which is why they never lose money.
IshKebab 2y ago
Is that really true? The only example I can think of is online clothes shopping... But that certainly isn't a clear cut case (remember boo.com?). Online dating maybe? Certainly most big successful tech startups have pretty obvious businesses; they just executed them particularly well or at the right time or with the right connections.
- paulddraper 2y ago
  Not at all. Facebook...people posting their personal info online for everyone to see? AirBnb...couch-surfing in strangers houses? (One r*** and that idea is dead ) Netflix mailing DVDs to your house?
  - IshKebab 2y ago
    > people posting their personal info online for everyone to see? It's not for all to see, and it was even less for all to see when it started. > AirBnb...couch-surfing in strangers houses? That idea didn't really last though; now it's just private landlords. > Netflix mailing DVDs to your house? Obvious idea in the era of Blockbuster. I don't think Netflix were the only ones.
    - ghaff 2y ago
      And there were books on tape rental by mail and probably other things.
    - paulddraper 2y ago
      Lol I congratulate you on investing so well. It was far from obvious for most people that these were wildly successful ideas before they became popular.
      - IshKebab 2y ago
        I think you missed this bit: > they just executed them particularly well or at the right time or with the right connections.
nothercastle 2y ago
Most of those ideas are bad because to be a unicorn you need to achieve something that seems damn near impossible at the time. Think VR, AI, Take over taxi biz, compete with big finance. It’s not a bad idea per se but it’s bad because it takes ridiculous scale to do so and most likely to fail
- exe34 2y ago
  It's even easier if you can afford the lawyers to bend rules, break the spirit of the laws and in some cases the actual letter of the law - just long enough to achieve dominance and then you can help write the laws to stop others competing with you.

marcinzm 2y ago

Obviously good ideas are ones that a large company department will also easily get funding for. That increases the competition a lot.

api 2y ago
Yeah but large companies often execute badly or develop solutions that are many many orders of magnitude more complex and expensive than needed. Large companies also generally push for all their products to work together, limiting their ability to build clean standalone things.
bcrosby95 2y ago
Not true when it comes to new tech. That's why anytime there's a new piece of tech there's a gold rush of new companies born. Internet, smart phones, social platforms, etc.

jszymborski 2y ago

In fairness, I think pg mostly concerns himself with start-ups whose goals are to become a unicorn or be acquired by one. I think there are plenty of boring, sustainable businesses built on good, logical sounding ideas that have found their niche in the industries they find themselves in.

nprateem 2y ago

Ah so post-hoc categorisation. That's really useful. I suppose it's blindingly obvious to VCs the bad ideas that are bad vs the bad ideas that are good, which is why they never lose money.

IshKebab 2y ago

Is that really true? The only example I can think of is online clothes shopping... But that certainly isn't a clear cut case (remember boo.com?). Online dating maybe? Certainly most big successful tech startups have pretty obvious businesses; they just executed them particularly well or at the right time or with the right connections.

paulddraper 2y ago
Not at all. Facebook...people posting their personal info online for everyone to see? AirBnb...couch-surfing in strangers houses? (One r*** and that idea is dead ) Netflix mailing DVDs to your house?
- IshKebab 2y ago
  > people posting their personal info online for everyone to see? It's not for all to see, and it was even less for all to see when it started. > AirBnb...couch-surfing in strangers houses? That idea didn't really last though; now it's just private landlords. > Netflix mailing DVDs to your house? Obvious idea in the era of Blockbuster. I don't think Netflix were the only ones.
  - ghaff 2y ago
    And there were books on tape rental by mail and probably other things.
  - paulddraper 2y ago
    Lol I congratulate you on investing so well. It was far from obvious for most people that these were wildly successful ideas before they became popular.
    - IshKebab 2y ago
      I think you missed this bit: > they just executed them particularly well or at the right time or with the right connections.

nothercastle 2y ago

Most of those ideas are bad because to be a unicorn you need to achieve something that seems damn near impossible at the time. Think VR, AI, Take over taxi biz, compete with big finance. It’s not a bad idea per se but it’s bad because it takes ridiculous scale to do so and most likely to fail

exe34 2y ago
It's even easier if you can afford the lawyers to bend rules, break the spirit of the laws and in some cases the actual letter of the law - just long enough to achieve dominance and then you can help write the laws to stop others competing with you.