7 ms·

What Kills Startups? A study.

98 points by entangld 15y ago

InnocentB 15y ago
The study indicates premature scaling is the number one cause of startup failure. Y-Combinator's results seem to indicate the number one cause of failure among its startups is kind of the opposite: the startup just kind of peters out, and the founders go work on something else. I assume this is a result of selection bias on both sides: Y Combinator only funds extremely small startups (generally 2-3 people), and this study likely (though I'm having trouble verifying) only includes startups that got beyond this phase. Does this sound reasonable?
- maxmarmer 15y ago
  Hey Max here...one of the authors of the study, Our study contains startups that are in stages across the board, (Discovery, Validation, Efficiency, Scale). While many of the YC startups don't reach the Scale stage, and maybe don't scale up their team or or raise too much money, they can still prematurely scale the product by over-engineering the product and not doing enough customer development. There are more nuanced case of premature scaling that are also discussed in the report.
  - brlewis 15y ago
    I was going to post here asking what sort of scaling the report was talking about since asking here would be faster than registering to get the report. Now I have your answer, but the answer is intriguing enough that I'll probably go ahead and register.
Hrothgar15 15y ago
"98% of startups fail because 98% of startups are unnecessary. They're trying to solve artificial problems." -Joey Pfeifer
- maxmarmer 15y ago
  The question is whether the founders are capable. Do you think 98% of founders are incapable of creating something useful?
  - steverist 15y ago
    My experience is that you're not going to have the problem as clearly defined as you thought in the beginning. If you have a capable team and a flexible strategy you can and must redefine the problem you're trying to solve as you learn more about the domain. I would add that as a form a failure in startup culture: To get so enamored with a perceived problem and your proposed solution as to ignore new evidence that should alter that first impression. Sometimes the new information leads you to incremental or 'attainable' adjustments and you adapt and move forward, sometimes you need to pull the ripcord and go back to the drawing board.
- Hrothgar15 15y ago
  Today I read an article with the headline "How to Pick the Right Idea for Your Startup." That right there epitomizes exactly what is wrong with startup culture. Don't have an idea? Don't have a startup. Ideally, problems could be solved and ideas implemented without the need to form a company, get funding, or turn a profit. Those are simply (often optional) measures that must be taken to help realize the end goal.
  - pg 15y ago
    Don't have an idea? Don't have a startup. I disagree with that. Some of the better startups we've funded changed their idea completely during YC. It's the founders that matter, not the idea. So I'd say instead: if you're not the right sort of person to start a startup, don't start a startup. But of course that is a complicated matter to decide because (a) people often don't know if they are the right sort of person and (b) people change.
    - jsdalton 15y ago
      So, why not recruit small teams of awesome founders -- comprised of the right kind of people -- and "assign" them ideas?
pg 15y ago
"Herrmann said that the team’s research found that of the 90 percent of startups that fail, 70 percent scaled prematurely" They must have had a strangely biased sample. Since the Bubble (when it was common) I've seen few startups do that. What kills startups is making something users don't want.
- nicholasjbs 15y ago
  Maybe by "scaled prematurely" they mean "started hiring and spending lots of money before they made something people want." Edit: This wasn't intended to be snarky, I was being serious.
  - pg 15y ago
    I assume that's what he meant, but it is comparatively rare. Nowadays most failed startups die small.
    - maxmarmer 15y ago
      Max here—one of the co-authors with Bjoern Herrmann, The way we've defined premature scaling, "making something users don't want" is included. For example, on the customer interaction metrics, which is what is used for "actual stage" a startup would be in the discovery or validation stages (stage 1 or 2). On the behavioral stage if they're making something people don't want, they're probably focused on streamlining their product or making it more scalable, which is a stage 3, or efficiency stage action. This would cause the startup to show up as a Behavioral Stage 3 / Actual Stage 2 and be labeled as premature scaling, just not the drastic kind, that for example WebVan exhibited — "Behavioral Stage 4 / Actual Stage 1". They had a team that was completely scaled up without even having shipped their product.
    - bjoernlasseh 15y ago
      The way we've defined premature scaling, "making something users don't want" is included. For example, on the customer interaction metrics, which is what is used for "actual stage" a startup would be in the discovery or validation stages (stage 1 or 2). On the behavioral stage if they're making something people don't want, they're probably focused on streamlining their product or making it more scalable, which is a stage 3, or efficiency stage action. This would cause the startup to show up as a Behavioral Stage 3 / Actual Stage 2 and be labeled as premature scaling, just not the drastic kind, that for example WebVan exhibited — "Behavioral Stage 4 / Actual Stage 1". They had a team that was completely scaled up without even having shipped their product.
kadabra9 15y ago
Disclaimer: Purely anecdotal evidence ahead. I've worked for two startups that failed and did consulting work for another one that idled lifelessly while the founder refused to admit defeat. In general, aside from having a clunky, ugly product that didn't really solve a genuine problem, I also noticed another trend: The founders were too stubborn to recognize or listen to their users (if they had any). The word "pivot" was not in any of these founders' vocabulary. We'd look at the data, and see users not touching these new features we added and instead the founder(s) just shrugged and said, "No, the users just aren't using the product correctly". No, you idiot. Your product sucks. I, along with a few other employees tried in vain to convince the founders that we needed to stop adding useless features and start listening to the data, and the founders just decided to try to convince the users to use the app the "correct" way instead. This, in turn, led to employees (and even co-founders) basically just giving up on trying to convince the founder/CEO of a need to pivot, so in essence the employees had already given up on the product. And when your employees (and cofounders, ffs) have given up on the product, you've already lost. interesting read.
udits 15y ago
Only the startups that solve real problems are able to survive. As a startup you should answer the following questions: 1. Would you use your own product? 2. Would you pay for it? If the answer to both the questions is a BIG yes, then you have a solid product and a good chance to survive.
- huhtenberg 15y ago
  And 3. Are you the only one? :)
marquis 15y ago
Warning: I filled out 80% of this, stopped for a call then hit the back button accidentally when I got back. All my input seems to be erased and I'm not spending another 20 minutes on filling it out again to find out what happens at the end. (I just submitted this bug to their uservoice).
gabaix 15y ago
Paul Graham's version on what kills startups: http://www.paulgraham.com/startupmistakes.html http://www.paulgraham.com/startupmistakes.html
vikas5678 15y ago
I'd use this product if the company name was not a required field.
- alanh 15y ago
  Petty quibble. Save it for https://startupgenome.uservoice.com/ https://startupgenome.uservoice.com/ — and note that if your company doesn’t have a (working) name yet, then you’re too early stage to see much benefit here. (If it’s a privacy issue, fake the name.)
dennisgorelik 15y ago
"A core feature is something that is required for your service to deliver value. For facebook a core feature is the ability to friend people. An email notification from facebook that someone sent you a message is a nice to have but not a core feature." Is it a joke? Without email notifications Facebook would never grow to any meaningful size (if at all).
- AltIvan 15y ago
  100% true. "Look! A lot of people just commented on things i commented on that new thing called Facebook and i faithfully believe that those are answers to what i said! Lets check those out!."
torsten1979 15y ago
It's an interesting study, no doubt. However I keep wondering if you can really judge a start-up's chance to succeed without taking the actual business idea/concept as well as the entrepreneur's personality into account. Since it's very hard to operationalize those, I do understand that you refrained from including it. I've just got the feeling that the quality of the business idea and the entrepreneur's traits might be a pretty powerful confounding variable for what you examine. Therefore it seems necessary to me that these variables are being controlled somehow. I'd like to illustrate that issue. You name premature scaling as the most important factor for start-ups to fail. While this conclusion might not be wrong on the one hand, it may be an illusion that the premature scaling is the actual reason for failure. As an example, premature scaling should be very likely to happen to a "megalomaniac" entrepreneur who just overestimates his business idea's chance to succeed or its growth rate by far. If megalomania is in place at that stage, one have to assume that it also was when the entrepreneur founded his business - by highly overestimating a crap idea who wouldn't have deserved to be turned into a business. In this case, the business' failure wouldn't have been caused by the premature scaling (although it seems to be since that's what you were looking at) but by the entrepreneur's tendency to overestimate the quality of his idea. That said the premature scaling is just another outcome of the true, underlying reason for failure - but not a reason itself. It's just an example, but it might show that it's risky to name your variables as crucial factors for failure as long as you don't control what the business is based on in the first instance: the business idea and the entrepreneur's personality. What are your thoughts on that?
- maxmarmer 15y ago
  The initial idea isn't that important. Entrepreneurs change their idea all the time. In the report we show that consistent companies or ones that scale properly, are more focused on discovering whether their idea makes sense where as companies that scale prematurely are more focused on validating that they are right. And as far as personality, while it's an interesting variable to look at in the future, in the end of the day we just look at whether the company produces results regardless of their predisposition. It's also common wisdom that there are many different kinds of entrepreneurs that have been successful.
fatalerrorx3 15y ago
I think another major point that wasn't discussed in the post is that the startup world is a rocky one. You might be full of confidence in your idea one day, and then the next day you feel your idea is garbage. It's a roller-coaster ride, and not everyone can deal with that. Especially until you have users that are signing up for your service, it's really just your own personal opinion, you need validation in an idea to continue to give you the confidence to pursue it no matter what your opinion of it at the moment is

98 points by entangld 15y ago

The study indicates premature scaling is the number one cause of startup failure. Y-Combinator's results seem to indicate the number one cause of failure among its startups is kind of the opposite: the startup just kind of peters out, and the founders go work on something else. I assume this is a result of selection bias on both sides: Y Combinator only funds extremely small startups (generally 2-3 people), and this study likely (though I'm having trouble verifying) only includes startups that got beyond this phase. Does this sound reasonable?

maxmarmer 15y ago
Hey Max here...one of the authors of the study, Our study contains startups that are in stages across the board, (Discovery, Validation, Efficiency, Scale). While many of the YC startups don't reach the Scale stage, and maybe don't scale up their team or or raise too much money, they can still prematurely scale the product by over-engineering the product and not doing enough customer development. There are more nuanced case of premature scaling that are also discussed in the report.
- brlewis 15y ago
  I was going to post here asking what sort of scaling the report was talking about since asking here would be faster than registering to get the report. Now I have your answer, but the answer is intriguing enough that I'll probably go ahead and register.

Hrothgar15 15y ago

"98% of startups fail because 98% of startups are unnecessary. They're trying to solve artificial problems." -Joey Pfeifer

maxmarmer 15y ago
The question is whether the founders are capable. Do you think 98% of founders are incapable of creating something useful?
- steverist 15y ago
  My experience is that you're not going to have the problem as clearly defined as you thought in the beginning. If you have a capable team and a flexible strategy you can and must redefine the problem you're trying to solve as you learn more about the domain. I would add that as a form a failure in startup culture: To get so enamored with a perceived problem and your proposed solution as to ignore new evidence that should alter that first impression. Sometimes the new information leads you to incremental or 'attainable' adjustments and you adapt and move forward, sometimes you need to pull the ripcord and go back to the drawing board.
Hrothgar15 15y ago
Today I read an article with the headline "How to Pick the Right Idea for Your Startup." That right there epitomizes exactly what is wrong with startup culture. Don't have an idea? Don't have a startup. Ideally, problems could be solved and ideas implemented without the need to form a company, get funding, or turn a profit. Those are simply (often optional) measures that must be taken to help realize the end goal.
- pg 15y ago
  Don't have an idea? Don't have a startup. I disagree with that. Some of the better startups we've funded changed their idea completely during YC. It's the founders that matter, not the idea. So I'd say instead: if you're not the right sort of person to start a startup, don't start a startup. But of course that is a complicated matter to decide because (a) people often don't know if they are the right sort of person and (b) people change.
  - jsdalton 15y ago
    So, why not recruit small teams of awesome founders -- comprised of the right kind of people -- and "assign" them ideas?

pg 15y ago

"Herrmann said that the team’s research found that of the 90 percent of startups that fail, 70 percent scaled prematurely" They must have had a strangely biased sample. Since the Bubble (when it was common) I've seen few startups do that. What kills startups is making something users don't want.

nicholasjbs 15y ago
Maybe by "scaled prematurely" they mean "started hiring and spending lots of money before they made something people want." Edit: This wasn't intended to be snarky, I was being serious.
- pg 15y ago
  I assume that's what he meant, but it is comparatively rare. Nowadays most failed startups die small.
  - maxmarmer 15y ago
    Max here—one of the co-authors with Bjoern Herrmann, The way we've defined premature scaling, "making something users don't want" is included. For example, on the customer interaction metrics, which is what is used for "actual stage" a startup would be in the discovery or validation stages (stage 1 or 2). On the behavioral stage if they're making something people don't want, they're probably focused on streamlining their product or making it more scalable, which is a stage 3, or efficiency stage action. This would cause the startup to show up as a Behavioral Stage 3 / Actual Stage 2 and be labeled as premature scaling, just not the drastic kind, that for example WebVan exhibited — "Behavioral Stage 4 / Actual Stage 1". They had a team that was completely scaled up without even having shipped their product.
  - bjoernlasseh 15y ago
    The way we've defined premature scaling, "making something users don't want" is included. For example, on the customer interaction metrics, which is what is used for "actual stage" a startup would be in the discovery or validation stages (stage 1 or 2). On the behavioral stage if they're making something people don't want, they're probably focused on streamlining their product or making it more scalable, which is a stage 3, or efficiency stage action. This would cause the startup to show up as a Behavioral Stage 3 / Actual Stage 2 and be labeled as premature scaling, just not the drastic kind, that for example WebVan exhibited — "Behavioral Stage 4 / Actual Stage 1". They had a team that was completely scaled up without even having shipped their product.

kadabra9 15y ago

Disclaimer: Purely anecdotal evidence ahead. I've worked for two startups that failed and did consulting work for another one that idled lifelessly while the founder refused to admit defeat. In general, aside from having a clunky, ugly product that didn't really solve a genuine problem, I also noticed another trend: The founders were too stubborn to recognize or listen to their users (if they had any). The word "pivot" was not in any of these founders' vocabulary. We'd look at the data, and see users not touching these new features we added and instead the founder(s) just shrugged and said, "No, the users just aren't using the product correctly". No, you idiot. Your product sucks. I, along with a few other employees tried in vain to convince the founders that we needed to stop adding useless features and start listening to the data, and the founders just decided to try to convince the users to use the app the "correct" way instead. This, in turn, led to employees (and even co-founders) basically just giving up on trying to convince the founder/CEO of a need to pivot, so in essence the employees had already given up on the product. And when your employees (and cofounders, ffs) have given up on the product, you've already lost. interesting read.

udits 15y ago

Only the startups that solve real problems are able to survive. As a startup you should answer the following questions: 1. Would you use your own product? 2. Would you pay for it? If the answer to both the questions is a BIG yes, then you have a solid product and a good chance to survive.

huhtenberg 15y ago
And 3. Are you the only one? :)

marquis 15y ago

Warning: I filled out 80% of this, stopped for a call then hit the back button accidentally when I got back. All my input seems to be erased and I'm not spending another 20 minutes on filling it out again to find out what happens at the end. (I just submitted this bug to their uservoice).

gabaix 15y ago

Paul Graham's version on what kills startups: http://www.paulgraham.com/startupmistakes.html http://www.paulgraham.com/startupmistakes.html

vikas5678 15y ago

I'd use this product if the company name was not a required field.

alanh 15y ago
Petty quibble. Save it for https://startupgenome.uservoice.com/ https://startupgenome.uservoice.com/ — and note that if your company doesn’t have a (working) name yet, then you’re too early stage to see much benefit here. (If it’s a privacy issue, fake the name.)

dennisgorelik 15y ago

"A core feature is something that is required for your service to deliver value. For facebook a core feature is the ability to friend people. An email notification from facebook that someone sent you a message is a nice to have but not a core feature." Is it a joke? Without email notifications Facebook would never grow to any meaningful size (if at all).

AltIvan 15y ago
100% true. "Look! A lot of people just commented on things i commented on that new thing called Facebook and i faithfully believe that those are answers to what i said! Lets check those out!."

torsten1979 15y ago

It's an interesting study, no doubt. However I keep wondering if you can really judge a start-up's chance to succeed without taking the actual business idea/concept as well as the entrepreneur's personality into account. Since it's very hard to operationalize those, I do understand that you refrained from including it. I've just got the feeling that the quality of the business idea and the entrepreneur's traits might be a pretty powerful confounding variable for what you examine. Therefore it seems necessary to me that these variables are being controlled somehow. I'd like to illustrate that issue. You name premature scaling as the most important factor for start-ups to fail. While this conclusion might not be wrong on the one hand, it may be an illusion that the premature scaling is the actual reason for failure. As an example, premature scaling should be very likely to happen to a "megalomaniac" entrepreneur who just overestimates his business idea's chance to succeed or its growth rate by far. If megalomania is in place at that stage, one have to assume that it also was when the entrepreneur founded his business - by highly overestimating a crap idea who wouldn't have deserved to be turned into a business. In this case, the business' failure wouldn't have been caused by the premature scaling (although it seems to be since that's what you were looking at) but by the entrepreneur's tendency to overestimate the quality of his idea. That said the premature scaling is just another outcome of the true, underlying reason for failure - but not a reason itself. It's just an example, but it might show that it's risky to name your variables as crucial factors for failure as long as you don't control what the business is based on in the first instance: the business idea and the entrepreneur's personality. What are your thoughts on that?

maxmarmer 15y ago
The initial idea isn't that important. Entrepreneurs change their idea all the time. In the report we show that consistent companies or ones that scale properly, are more focused on discovering whether their idea makes sense where as companies that scale prematurely are more focused on validating that they are right. And as far as personality, while it's an interesting variable to look at in the future, in the end of the day we just look at whether the company produces results regardless of their predisposition. It's also common wisdom that there are many different kinds of entrepreneurs that have been successful.

fatalerrorx3 15y ago

I think another major point that wasn't discussed in the post is that the startup world is a rocky one. You might be full of confidence in your idea one day, and then the next day you feel your idea is garbage. It's a roller-coaster ride, and not everyone can deal with that. Especially until you have users that are signing up for your service, it's really just your own personal opinion, you need validation in an idea to continue to give you the confidence to pursue it no matter what your opinion of it at the moment is