4 ms·

Low margin businesses are all the rage because all the large, high-margin opportunities are either gone or have such huge barriers to entry that startups have t

by exelius 11y ago

Low margin businesses are all the rage because all the large, high-margin opportunities are either gone or have such huge barriers to entry that startups have to raise a half billion dollars before making an honest attempt. At least that's what I see. Not that we shouldn't chase the expensive ones, but I don't know that VC is the right framework for that.

nostrademons 11y ago
Maybe the solution to that is to go after small, high-margin opportunities, ideally growing ones. http://paulgraham.com/startupideas.html http://paulgraham.com/startupideas.html
- exelius 11y ago
  This is currently the model that most big consulting companies use: find a thousand niches and fill them all with people. I have a feeling that if you enter these markets as a startup, you'll find they're not really as high margin as they look, and that the margin will get eaten up by integrators anyway (integration work doesn't scale exponentially and causes your costs/outputs to lag your pipeline). Anyway it's not a bad idea; just not one I feel is especially well suited to the VC investment model.
  - ghaff 11y ago
    It's also the model for a lot of business software that caters to some specialist requirement, industry vertical, etc. Just look at the huge partner ecosystem of Salesforce for example. Something like 160,000 people attended Dreamforce. Some of that is large software firms and in-house end-user development of course, but a lot of it is small ISVs with niche add-ons.
  - nostrademons 11y ago
    I think the idea is to pick niches that are small now but will become big in the future. Build the only programming language available for the Altair and you have a high-margin monopoly on a market of a few thousand machines. Build the only operating system for every IBM PC and clone, and you have a high-margin monopoly on a market of hundreds of millions of machines. That's behind the common SV wisdom of "catch a wave", "ride a trend that's larger than yourself", "find markets, don't try to create them", etc. The pitfall is that it requires some degree of future-prediction, and predicting the future is left as an exercise to the entrepreneur. The guy who built the first programming language for the Altair is the richest man in the world; the guy who built the first programming language for the Alto [1] has largely been forgotten by history. [1] https://en.wikipedia.org/wiki/Xerox_Alto https://en.wikipedia.org/wiki/Xerox_Alto

by exelius 11y ago

nostrademons 11y ago

Maybe the solution to that is to go after small, high-margin opportunities, ideally growing ones. http://paulgraham.com/startupideas.html http://paulgraham.com/startupideas.html

exelius 11y ago
This is currently the model that most big consulting companies use: find a thousand niches and fill them all with people. I have a feeling that if you enter these markets as a startup, you'll find they're not really as high margin as they look, and that the margin will get eaten up by integrators anyway (integration work doesn't scale exponentially and causes your costs/outputs to lag your pipeline). Anyway it's not a bad idea; just not one I feel is especially well suited to the VC investment model.
- ghaff 11y ago
  It's also the model for a lot of business software that caters to some specialist requirement, industry vertical, etc. Just look at the huge partner ecosystem of Salesforce for example. Something like 160,000 people attended Dreamforce. Some of that is large software firms and in-house end-user development of course, but a lot of it is small ISVs with niche add-ons.
- nostrademons 11y ago
  I think the idea is to pick niches that are small now but will become big in the future. Build the only programming language available for the Altair and you have a high-margin monopoly on a market of a few thousand machines. Build the only operating system for every IBM PC and clone, and you have a high-margin monopoly on a market of hundreds of millions of machines. That's behind the common SV wisdom of "catch a wave", "ride a trend that's larger than yourself", "find markets, don't try to create them", etc. The pitfall is that it requires some degree of future-prediction, and predicting the future is left as an exercise to the entrepreneur. The guy who built the first programming language for the Altair is the richest man in the world; the guy who built the first programming language for the Alto [1] has largely been forgotten by history. [1] https://en.wikipedia.org/wiki/Xerox_Alto https://en.wikipedia.org/wiki/Xerox_Alto