HN Explorer

Ask pg: Why MUST good startup ideas look like bad ideas initially?

4 points by chacham15 ↗ HN
You have said many times that "good startup ideas have to look like bad ideas initially."

The reason that you suggest in your essays that the idea MUST look like a bad idea is because if it didn't then other people would do it. What leads me to doubt whether this is true is the idea that it ignores innovation. Therefore, what if the reason that other people dont do it is simply because they don't know how?

For example, if a startup were able to create a piece of software that could take a block of text and produce a summary (not keywords and such, but a real summary that a human would write if asked). I would think that that seems like (and IMO is) a good idea. Here, the problem is how do I do this rather than should I do this.

A historic example rather than a theoretical one could be Googles' basic concept of page-rank. It seems like a good idea (this is not just hindsight, IIRC there are even stories about Larry Paige explaining the concept at the time and people thinking that it was genius). Just, how many people at that time thought of it? Even if people did think of it, it still is a technically difficult algorithm when you take into account the sheer amount of data that it must process. So again, the problem is not should I, but rather, how do I.

Do you see any error in my logic?

8 comments

[ 0.79 ms ] story [ 23.2 ms ] thread
quanticle
The reason he says that a good startup idea looks like a bad idea is that good startup ideas must be hard, and if they weren't, many people would have already done it by now.

For example, if a startup were able to create a piece of software that could take a block of text and produce a summary (not keywords and such, but a real summary that a human would write if asked). I would think that that seems like (and IMO is) a good idea. Here, the problem is how do I do this rather than should I do this.

The problem truly is, "How do I do this profitably?" Sure, lots of problems are hard. But there aren't very many hard problems that are also profitable. In my opinion, that's the greatest difficulty with a startup. It's not convincing investors that the problem is difficult. It's convincing investors that people are willing to pay for the solution.

A historic example rather than a theoretical one could be Googles' basic concept of page-rank. It seems like a good idea (this is not just hindsight, IIRC there are even stories about Larry Paige explaining the concept at the time and people thinking that it was genius). Just, how many people at that time thought of it? Even if people did think of it, it still is a technically difficult algorithm when you take into account the sheer amount of data that it must process. So again, the problem is not should I, but rather, how do I.

Again, you're misunderstanding the problem. Yes, PageRank was an "obviously good" algorithm. Heck pg himself thought of something very similar while he was at Yahoo. But you have to also remember that, at the time, web search was considered a "dead" field. Everyone thought that the future was manually created indexes and portals (like Yahoo) rather than in search. Google's real accomplishment was, IMHO, showing that web search could be profitable. Yes, they couldn't have done it without PageRank, but the new algorithm was only a necessary factor, not a sufficient one.

chacham15
IIRC, many startups have been funded by investors based upon purely traction, not profit. I am not saying that that is the optimal path, but still a viable one.

For example, Facebook went a LONG time before they actually started making money. They created something that people wanted and consequently gained a lot of traction. Later down the line, they were able to turn the traction into profit.

"In September 2009, Facebook said that it had turned cash-flow positive for the first time." Facebook was founded in 2004.

orangethirty
My take on this (I'm actually deciding what the fuck I'm going to do for my startup right now)is that Paul refers to one thing. Big ideas come with a lot of criticism. People will call you crazy and what not. Thus, they seem like bad ideas to most because most people are afraid of ridicule and failure. Imagine if I woke up one day, and said "I'm going to take down Google." Everybody on HN would laugh histerically. To most of you, it would seem like a bad idea. But Google is not as big and mighty as it appears. Anyone in here can build something that will compete with it (in terms of software). The problem lies in being stupid enough to believe that you can do it.

I thought about taking on Google, but my passion seems to lie outside of search. I'm not ruling it out though.

masterzora
The problem with your logic is that it assumes that you uniquely are able to solve a problem. This is probably not true, especially if I've never heard of you before. You really should assume there are multiple people smarter than you (a trivial assumption) and that they've thought of the same idea as you. Given this assumption, I think PG's quote is really a snappier way of saying "Ask yourself why these smarter people haven't done it yet." Do you truly believe you stumbled on a solution smarter people legitimately haven't? What advantage did you have that they didn't? Or, if they have thought of it, would they have problems with execution or traction or funding or even just explaining the idea?
chacham15
Please dont get personal, this is a theoretical argument. It is natural to assume that there is a variation in intelligence in the computer science world and that there will be people who are outliers. By definition, there are few people who are smarter than the outliers. The idea that someone smarter would have thought of it doesnt apply to them because the number of people smarter than them is very small.

The second problem is: how do we recognize these outliers? Take for example, Larry Paige. The problem with 'especially if I've never heard of you before' is demonstrated here. People are never claimed to be really smart until they've done something noteworthy. This, however, is the classic illustration of hindsight. We deem them smart because they have done something smart, not because they are. Therefore, would it be fair to ask Larry Paige those questions at the time that he was founding Google?

Furthermore, pg has an entire essay on why smart people have bad ideas: http://paulgraham.com/bronze.html. Therefore, even if there are people smarter than you, that is not to say that they will have that brilliant idea instead of you.

An example to that point is Drew Houston and Dropbox. Many people, arguably smarter, had thought of that idea before but Drew was able to execute it in a way that no-one else had thought of.

masterzora
I am not trying to get personal. I use the second person as a rhetorical device because properly consistent use of the third person sounds really stuffy.

That said, you are touching on something I left implicit in my original post. Yes, there is a really small chance that you are the smartest person with the most relevant experience relative to a given problem to approach it. And, if this is the case, that is your valid answer to why you haven't seen someone else do it before. But for most people this is simply not the case and even for the people you are thinking of now as the smartest should assume it isn't the case, at least at first.

If analyzing the situation and doing research legitimately leads you to "I am better/luckier than everyone else who's approached this", good for you because you're off to a great start.

As far as Larry Page goes, a sibling post said that pg, to name one person, had something very similar to PageRank futher driving in the point that you usually can't count on being the first to figure out a solution. But, regardless, it absolutely would have been fair to ask Larry those same questions at the time he was founding Google. No matter how smart you are, due diligence is a key concept.

chacham15
How would Larry Paige or Drew Houston have done research to conclude that they are "better/luckier than everyone else who's approached this"? After all, Google was one of the last search engines to enter the industry.

I am not even convinced that that is true. As I said in the last post, given a certain number of people that can solve a problem, what are the chances that they are working on the same problem as Larry or Drew? Furthermore, the amount of time that it takes to prove that an idea is actually a good one is a few years at the least. Next, the range of ideas is so large, that this group of people is going to be heavily segmented in different areas. Lastly, as I pointed out with Dropbox, innovation doesn't always lie entirely in the difficulty of the problem, but also the execution. All of these factors go to increase the minimal size of smart people who can be the first to innovate in their fields.

masterzora
> How would Larry Paige or Drew Houston have done research to conclude that they are "better/luckier than everyone else who's approached this"?

In a negative sense, mostly. If it's absolutely a good idea that you know you have the solution for and it doesn't require a ridiculous amount of capital to execute and said execution is fairly straightforward and a sizeable market exists and etc, etc then the only possible conclusion is that you are legitimately close enough to the first one to solve the problem that you can go ahead. But odds are against this being the case so you should assume that it's not and then figure it out from there rather than assume that it is the case and you have an easy business ahead of you. Anything else would just be a bad investment of your time and resources.