I've been creating stuff for the last 5 years and if I had to tell you just one lesson from all of it, is this. I have seen in the last year how all my ideas from 3 or 4 years ago are becoming the big successes of the current web scene.
I'm not talking about an idea I had that resembled a current service. No, I'm talking about the same exact actual products and services. I'm in a perpetual WTF? state at how identical my ideas are.
Difference? Execution.
This also has made adopt a posture about the complexity of the idea itself. IMHO strip you idea of unnecessary complexity (advanced functionalities, etc, etc) in order to be able to execute it as good as you can as fast as you can. Simplicity allows better execution.
This seems obvious but is damn hard to accomplish.
> IMHO strip you idea of unnecessary complexity (advanced functionalities, etc, etc) in order to be able to execute it as good as you can as fast as you can. Simplicity allows better execution.
I think this is one of the important facets that determine how successful your execution will be. Put in too little functionality and it could fail. Put in too much and it could fail as well. All other things equal (design, marketing, etc) it's almost as if you need just enough to introduce the core functionality while still being useful, then iterating on that to bring in additional functionality once your customers grok the initial set.
I wonder if it would help to think about what job the user wants done (instead of how good the app is).
Christensen suggests the user is hiring your app, as an employee, to get something done for them (it might be a work task, or it might be fun, to pass time, start conversation, break ice etc). So: what's the minimum the app has to do, to achieve that, for some user?
I need to send this to every person who tells me about their amazing iPhone app idea. Angry Birds was a different take/execution on an idea from an old Flash game. Doodle Jump was nearly the SAME app as Super Papi Jump but with new graphics and execution. How many unit conversion apps are there? The guys who made Convert were the last app to do it but they went to #1 thanks to brilliant execution. Fart apps? I'm friends with the guy who made the #2 fart app, and I can bust his chops all I want, but he's now a millionaire.
In the App Store it's actually rare for a mega hit app or game to truly be original: they're all mostly riffs on existing apps or games.
Reminds me of Ed Catmull's talk about Pixar management at http://www.youtube.com/watch?v=k2h2lvhzMDc#t=21m50s -- "If you have a good idea and you give it to a mediocre group, they'll screw it up. If you give a mediocre idea to a good group, they'll fix it, or they'll throw it away and come up with something else."
I hear a lot of people that say "I have a huge idea, I can't tell you about it." In most cases, they really haven't thought it through, and it has many flaws.
I love when people tell me that they have the "next big idea," but that they can't disclose it. I lose interest the second someone tells me that, because if it were truly a great idea, they would instead be implementing it. If you have a 'great idea' that you don't know how to implement, it's probably not that great of an idea, because you inherently lack the skills to discover its weaknesses.
So, let's see: I have a 'great idea'. One pill taken once that will cure any cancer. World shaking idea!
This example clarifies this whole thread: The need to cure cancer is obvious. So is the desirability of such a one pill. Now let's set aside the obvious. The challenge is how to make that pill. The 'idea' with meaning, then, is how to make this pill.
If someone does know how to make the pill, then they definitely should not tell anyone about it without good legal advice, rock solid NDAs, etc.
But part of what makes this thread a mess is that at present the HN context of information technology entrepreneurship just assumes that, with current computing, the part analogous to finding the pill is trivial and the only challenge is identifying that people would like to have such a pill. This situation won't last very long!
I signed plenty of NDA's only to hear a really crappy idea. And even if the idea was nice, it would be nearly impossible to execute. Like having some original-ish idea like square is nice, but takes millions to execute. If you don't have the means to get to those millions, you can have the idea but it doesn't really matter. Someone who does have that access will win.
You are making a common but fundamental and potentially very costly mistake: You are looking for a needle in a haystack. Then you are drawing conclusions about the needle from the hay you do see and not the needle you don't.
It may be that for the needle, you will have to sign an NDA and, then, be very glad you did.
We're looking for needles, guys.
We can grab some hay and just hope we find a needle there.
Or we can try some solid means to find and/or create needles. Examples? Sure, look at the last 70 years of the US DoD. E.g., they bet on just the 'idea', just on paper, for, for just a short list, the proximity fuse, the atom bomb, the turbojet engine (GE once told me that the first prototype for any jet engine costs > $100 million), the engines for the SR-71 (both a turbojet and a ram jet), the SR-71, the Navy version of the GPS, the F-117, and much more. They start with just paper, and the failure rate from there on is surprisingly small. The US DoD can evaluate far out projects just on paper with high reliability. I almost forgot, the Internet.
It was DoD, the CIA, and NASA that mostly got Silicon Valley going. Now just what is it about how to evaluate projects that got Silicon Valley going that Silicon Valley now doesn't understand?
May I have the envelope, please? Here it is: DoD does evaluations with people with both technical and military competence, and Silicon Valley uses marketing, 'biz dev', Wall Street financial analysis, management consultant people. Hmm ....
No: Your evidence does not lead to your conclusion.
You are working with your "a lot of people" and drawing a conclusion about business success in, since this is HN, information technology entrepreneurship. Here's where you get off track: Something under 1% of the people with such 'ideas' achieve success. Or, what fraction of the people can be in the upper 1%?
So, the evidence that would be needed would be the role of ideas for that upper 1% or less.
As I wrote elsewhere on this thread, what is being assumed here about an 'idea' is just some short description of a 'cutomer need' to be met by, say, a new Web site, say, paper napkins with personalized printing ordered on a Web site and delivered via FedEx. Then, sure, execution is everything.
What is being missed, however, is old, traditional, and rock solid: Take a problem millions of people want solved, or that a few people want very, very much to have solved, find a solution, and then execute. Here the implication is that, because the problem has not been solved yet, finding the solution, the 'idea', is challenging. Then commonly people would protect the idea with a patent.
If the idea was really good, then the execution should be routine. Examples? Business history is awash in them.
I say again: Bad ideas are easy and, then, execution is everything and likely difficult. Good ideas are likely difficult and then execution can be routine.
'The Narrow Road', the Felix Dennis book that's quoted from, is a condensation of his earlier 'How To Get Rich' (not 'How To Make Money', that's identical to 'The Narrow Road'.) I found the longer version equally valuable and vastly more entertaining.
While Felix is worth reading - a 'get rich' book from some who actually built a business and got rich! - he comes from the publishing world and his advice is very different from the practice of most technology startups.
For example, he advises never giving away any equity in your company, ever, and includes the story of how, early on in his publishing career, he fired all of his senior staff because they asked him for a small cut of the equity pie.
Is there really any debate whether you have to execute an idea to make it successful? Are there really people who think you can have an idea, keep it in your head, and become rich just by thinking it? Of course you have to execute the idea, everyone knows that. Ah, but the sage advice is to execute it 'correctly.' That will be a big help to all those poor souls who thought it was better to execute their ideas incorrectly. If they'd only known...
The catch is that there is no way to predict whether your execution is 'correct' a priori. It's like saying the way to win the lottery is to pick the correct ticket. Technically true, but not particularly helpful in practice.
"Is there really any debate whether you have to execute an idea to make it successful?"
Mostly no. There are some cases where next to no 'execution' is needed for the big bucks. One example was the Sears employee who proposed a button on the socket wrench that would eject the socket. His 'execution' was a lot of legal wrestling with Sears after they implemented the idea but didn't pay him. Eventually he got paid.
"Are there really people who think you can have an idea, keep it in your head, and become rich just by thinking it?"
No.
"Ah, but the sage advice is to execute it 'correctly.'"
I believe that more "sage" advice is to have a good idea where execution is routine!
Here HN is trying really hard to get into a swamp and ignore
the standard, old "sage" advice:
(1) Identify some target customer 'pain'. Want a bad pain so that a solution is a "must have" instead of just a "nice to have" (I know, Facebook and Twitter are so far mostly only nice to have).
(2) Find a solution to the pain. Hopefully this solution is new, much better than anything else, and can be protected.
(3) Use the Sequoia statement: "A huge market with customers yearning for a product developed by great engineers requires very little firepower." I don't fully believe this, but it's not all wrong. It's a version of "Build a better mouse trap and the world will beat a path to your door.", which is only partly true, also. Still, I didn't see a lot of Super Bowl ads for Google, Facebook, Twitter, etc.
Here a crucial part, that I would call the valuable 'idea', is in (2), for the solution.
Nice, I just ordered "The Narrow Road" on Amazon. Felix Dennis's "How to Get Rich" is one of the best business books I've ever read, I highly recommend it.
Sometimes the startup world devises its own theories on what makes businesses successful, and it's good to balance that out with the views of an old school no-nonsense, and massively successful, old media founder like Felix Dennis. He lays it out straight for you and he's all entrepreneur. Tons of wisdom in his books.
The last Startup Weekend that took place in Tel Aviv[1] has showed me exactly how important proper execution is. Great ideas are never a guarantee for success.
Having an environment that cultivates top-notch execution is critical.
I couldn't attend Startup Weekend, but I have experienced the same phenomenon at other startup events in Tel Aviv. The less technically focused the event is, the more people will sit around and discuss ideas, without coding a single line. If you want to focus on execution, attend a Python or Ruby event, instead of a general startup meeting.
Cool. I've wanted to attend PyWeb-IL for a long time, but I haven't been able to make it. Hopefully I'll be at the next one - look for a student with long hair.
To my untrained eye, it seems like "raise lots of money, buy many, many patents, build nothing so the target can't bargain with their patent portfolio, and then start negotiating". Once you have that machine working you can iron out the details. It's like domainers - get there first, buy everything that looks vaguely useful and extract rents.
It's no harder than "build a faddy social network thing that's slightly better than MySpace", or "build a slightly better search engine than AltaVista" to other untrained eyes, I guess.
> We often talk about the Number One startup killer at TechStars—making a product for which there is no interesting market. TechStars accepts just 10 of more than 600 startups that apply and presumably they are among the best. But still, we find that at least one-third of those startups are attempting to build a product that they want, or that no one wants, instead of what the market wants.
So, no, the idea alone doesn't make you rich, but having a good idea is worth something, given how many bad ones there are.
Here's the solution: Borrow from Edison and the electric light. Look at lighting before the good electric lights -- it sucked. There were open flames that were dangerous. The light quality was awful. Etc. Yes, it was expensive and dangerous, but you have to remember, the light quality sucked.
So, there was no doubt about the customer 'pain' or the 'market' or the 'market potential'.
So, why was there work to do? Because no one knew how the heck to "Do that".
So, the challenge was in how to do that. Eventually Edison pulled it all together, and the earnings have been a lot of what has kept GE going to the present.
It would have been still better if there had been some patents.
It can be still better if the solution is in software, essentially impossible to 'reverse engineer' and very difficult to duplicate or equal, and quite secure in code on servers.
So, TechStars needs to start with customer 'pain' that is huge and clear and not solved. Then TechStars needs to find a solution to this pain. The solution, the 'secret sauce', should give much better results than anything else and be difficult to reverse engineer, duplicate, or equal. For a Web site to order custom printed paper napkins delivered by FedEx (put the print shop in Memphis and let orders in by 10 PM Memphis time be delivered by 10:30 AM local time), that's essentially impossible to protect. So, f'get about it.
The prerequisite for executing something, is the idea that you have an idea. Otherwise you will never go through the hassle of making something in the first place :)
Before you write something it feels like you already know what you’re going to write. Still, writing is though and it takes a long time before you come to a result that says what you figured you want to say all along.
That is, because what actually happens, is that you work hard on converting this network of vague assumptions in your head, which already feels meaningful, into something which can actually be meaningful to others.
That may be, but it's the rare equity negotiation I enter into where the party with "the idea" understands that they're not going to get majority ownership.
While I feel that I agree, I'm having a hard time thinking of any counter examples this morning. Perhaps there are there specific industries where the focus isn't executing or creating but instead just being the first? Pharmaceuticals?
Pharmaceuticals seems to be a good example: it's one of the few industries where getting the consumer to know of your product is not the issue (Steve Gary Blank wrote that, if a company found a cure for cancer, marketing would be a non issue). At the same time, being the first with the product guarantees patents and there for lots of money.
My professor used to drill this saying into us in my Intro to Entrepreneurship class 2 years ago. I still use that quote most when giving advice to other students who have an idea but are hesitant to share it.
My freshmen year in college, I met two students who had just received VC funding, and they told me one tip that had helped them most: they shared their idea with anybody who would listen, because they never knew how somebody might be able to help them along.
If you truly have a good idea, somebody else is working on it too. You might as well talk about it and get all the help you can.
"If you truly have a good idea, somebody else is working on it too."
Standard HN cliche number 298,498,229 and nonsense.
I've had several ideas that I can essentially 100% guarantee that no one else was working on. My Ph.D. dissertation was officially judged "new, correct, and significant" -- note the "new". Similarly for each paper of original research I've published. Similarly for the crucial, core idea for the company I'm starting.
Just what is it about original research that HN just cannot understand?
Research is a different beast than creating a marketable product. If you're creating a product that nobody is working on or hasn't tried to make before, you have to stop and wonder, why has nobody done this before?
Before Tylenol existed, many people had attempted to tackle the problem of pain management. Tylenol just happened to make a more efficient chemical formula with their research, but the idea of using drugs to control pain was not new. When you talk about your idea, you don't need to reveal how you do something differently, just like Tylenol can talk about their idea without revealing their formula and Google can talk about their idea without revealing their algorithm.
I was in St Louis last week for a business pitch competition, and I watched a med school PhD candidate read his dissertation paper and present his research for the elevator pitch competition. He had absolutely no business knowledge or any idea on how to market his research into a profitable product. But because he was able to talk about his idea and research without revealing anything that would allow somebody else in the room to just copy it, he was able to build contacts with several other business-oriented people in the room who liked the idea and were interested in helping him take it to market.
"Research is a different beast than creating a marketable product."
Well there are people in research looking for fundamental knowledge, e.g., string theory in physics or the finest details of biochemistry and genetic material in biology. And there are people doing make-work, junk-think, prof-scam, busy work. And in neither case is there much connection with a product or market.
But your claim is not true in all cases! I'm creating a "marketable product". That the product is desirable and "marketable" is mostly obvious. The challenge is how to create the thing.
I found a way. That way was some original research in applied math. Since I have Ph.D. for such math, I'm able to make a competent attempt at such research and to judge any results for correctness as math and significance for the real problem.
There's essentially no doubt that my idea is "new" and that no one else is working on it. Reasons: First, my work has some advanced prerequisites nearly unknown anywhere among full profs in computer science. They didn't take enough of the right grad math courses, and, actually, neither did more than a tiny fraction of math profs. Some of the courses are rarely taught; the combination of courses is especially rare. Second, the crucial part of the work is clearly original; there's no chance it's in the books or journals, and there's essentially no chance that anyone else with the prerequisites would be trying to get the same results.
To me, that research I did was the crucial 'idea' for my business. At this point, the rest is routine, a simple Web site, some simple usage of relational database, some fairly simple initial data gathering, and some routine publicity.
You are insisting on saying that the research is not business and not the idea and that the idea is the market for a solution for the customer pain. That definition of an 'idea' is new and strange and a big break with centuries of history in technology for business, medicine, and the military and with the whole history of the USPTO.
Yes, your new definition of an 'idea' has become very common in Web 2.0 information technology entrepreneurship.
The people who invented the formula for Tylenol would agree that that invention of the chemical formula, and possibly how to do the manufacturing, was the crucial 'idea'. Just a better pill to alleviate pain was not the idea.
For the market I'm addressing, sure, lots of other people have tried. It's just that for the customer pain I'm addressing, the results from the other guys just suck and the results from my idea, as is quite clear from some of the math, should be much better. Eventually it will dawn on some people in computer science that I must have made good progress in having computer software work effectively with how humans view 'meaning', where they will be correct, and start to say my work is close to earth shaking. But I don't want their characterizations, praise, prizes, approval, admiration, gratitude, tenure, invitations to give lectures, etc.; I want a yacht. Then I want to compose music, starting roughly with the techniques of Richard Strauss, and then pursue mathematical physics, often on my yacht.
Yes, other people have tried to solve the same customer pain, but I do not conclude from this fact that they have been working on the same "idea" I am.
One pill taken once to cure any cancer is not a significant 'idea' or 'market'. The good idea would be how to make such a pill, and then the 'market' would be quite new.
I count aspirin, Tylenol, Advil, Benzocaine, etc. as great ideas for alleviating pain. I don't count the market for alleviating pain as a meaningful 'idea', but you seem to be.
Your new definition for an 'idea' actually can have some value in Web 2.0: It is not very clear just what people want from Web 2.0, the 'social graph', etc. so that a guess at what they want and how to provide it, even if all the necessary technology is trivial, might be a valuable 'idea'. There seems to be some progress here at
Still, this definition of an 'idea' is strange, new, very narrow, and in strong conflict with centuries of important history and nearly everything else in 'ideas' today from 'creative' in ads to the USPTO, DoD, NSF, and NIH funded projects, pharmaceutical research, and much more.
I have one more idea: Send Jane Fonda and Al Guru to Japan!
I think we are arguing the same point but with different terms.
To me, the 'idea' is how you would describe the product to a customer in one sentence. The execution is how this idea is made. In your case, this would be your research and unique math algorithm. The business plan is how you plan to market and monetize this product. A successful business requires all three factors.
When I said that an idea is being worked on elsewhere, I meant that somebody has at some point tried to solve that pain point for your customers. They are your competition. I stand by this statement. You seem to be arguing that your execution is unique, and I absolutely agree. Your execution has to be unique, or else you would be guaranteed to fail. Based on your description, it seems you are in an ideal position to execute your idea successfully, and I wish you the best of luck.
Based on the above definitions, I am suggesting that you can and should talk about your idea without talking about your execution. Share what problem you're solving, but definitely don't tell anybody the results of your math research. Using your example of a cancer pill, you should talk about your pill to anybody who will listen without sharing how the pill is going cure cancer.
I disagree that business plan factor are trivial after execution is handled. You could make the perfect product with your algorithm, but if you are unable to figure out a way to monetize or market your product, then it will fail. Of course, the relative difficulty of each factor depends on the industry and product.
"To me, the 'idea' is how you would describe the product to a customer in one sentence."
Yes, that seems to be a better definition of the 'new', Web 2.0 definition of an 'idea' than I was able to formulate. I just don't think it is a very appropriate, significant, or meaningful use of 'idea'.
I regard the important ideas as the crucial, difficult, powerful 'secret sauce' and the 'execution' as the routine stuff that comes later, especially after customers arrive. With a sufficiently good idea, say, the first Xerox machine, TWA in its first months as it had the only Lockheed Connies in the world, and more, as relatively 'routine' and this because the work before was so darned good. The US is absolutely awash in ability to do routine execution, from a corner store up to a giant corporation, well.
For my project, I've worked out more than the math: The math is now in code ready for production -- tested, timed, documented. It's shockingly fast. I've worked out a server farm 'architecture' that is dirt simple and looks like it will 'scale' beyond what the world could use. Much of the simplicity is because there is a lot of parallelism and redundancy and a lot of data that is read-only for at least a day and maybe a week. Actually, all the big data is read-only in this sense. And, it does appear that in production this big read-only data can be just in virtual memory off either rotating magnetic disks or the newer, faster solid state disks (that love read-only data). Windows Server 2008 (I'm staying with Microsoft) can have 16 TB of virtual memory, and my software does the accesses with very nice, and, thus, fast, locality of reference.
The role of the math is to take the available data, 'crunch' it, and give the desired results. The quality of the results is key and where the work of the other guys sucks.
There is a 'larger business plan' beyond just the crucial, core math: As a byproduct of the work itself, the Web site gets some data from the users. The users give that data as part of getting the results they want and won't mind giving the data because there is no privacy threat.
But from some subtle results in advanced math (the Radon-Nikodym theorem), that data will be the main source of 'base data' for the whole system for everything. So, in particular, compared with the other guys, I have some new data, of unusually high quality, and especially valuable. People won't see why the data is valuable, but it is! In particular, the data will let me do some good ad targeting, and that's where the 'monitization' and 'revenue model' come from. I don't really want to use third party cookies to connect with the ad targeting of the ad networks, and I should be able to do better ad targeting just with my own data and some more of my math.
The software promises to be astoundingly efficient: From the timings, data rates, and data sizes, it appears that for the on-line production servers, four racks should be enough for about $4.5 billion a year in revenue, and 2000 square feet of on-line rack space should be sufficient to serve the world. Right: If 4 racks can give $4.5 billion a year in revenue, then 2000 square feet of rack space should give one of the most valuable companies in the world. Of course would want more space for redundancy, backup and recovery, testing, and development. For more reliability and lower latency, might want more than one data center. Could make a little use of some caching on the network.
I'm not eager to say what the problem is much before I'm ready to go live.
Actually, telling people what the problem is seems to have no upside at all: Plenty of people in venture capital have had descriptions of the work polished to more luster than any gold and really don't give a hoot. The situation is very much as in, "before good ComScore numbers, the project is just an idea, and ideas are easy and plentiful. Execution is everything and doesn't start until after the ComScrore numbers." The flip side of this thinking is an opportunity for me.
Net, what they care about are ComScore numbers. But, as threatens to be increasingly the case, for a successful Web site, there will be a window narrow in time between when the ComScore numbers are (1) good enough for the VCs to invest and (2) so good that I won't need or take their check. The advantages of being 100% owner of a 100% private, very profitable Subchapter S corporation, or even a C corporation, are just huge.
Actually, I'm terrified of having a Board: (1) They could kill my company just from taking up 25% of my time developing information to report to them. (2) There is no way they could possibly understand anything significant about the future of the business and no way I could explain it to them or keep them happy about it.
For the 'algorithms', I would say that they are trivial because the real content is in the math; the only claim of power or quality from the code is the math it implements; actually, given the math, the implementation was not difficult. So what matters is the math, not the algorithms.
This is a general situation in newer computing: An algorithm standing alone has little to recommend it, and just 'experience' will be a long, slow path to 'validity'. Instead, for 'validity', quality, whatever, an algorithm needs a prior 'specification' of what it does. Often that specification will have to be some relatively advanced math. Currently this math is a huge bottleneck for promise in computing.
You just provided a whole lot of technical information about the efficiency of your algorithm, and even some financial projections for your company. That's the sort of information that you should keep to yourself unless specifically asked for because it is the valuable part of your company. The problem you are trying to solve is the part you should actively talking about, because that's what's going to get people's interest.
You clearly have a great understanding of the technical side of things, but the business/marketing/financial side is just as difficult. It's impossible to be good at everything, you will want help, either in the form of a cofounder and/or a board that can advise you.
Another saying that my prof used to repeat was "you can be rich or you can be king, but not both." Entrepreneurs like to have control, that's why we start companies in the first place. We think we can solve a problem better than anyone else can. As a result, we try to control every aspect of the company. Unfortunately, this is the fastest way to hurt the company. If you want to control 100% of the company, you will only be able to grow so much. If you are hoping to own a yacht as you mentioned earlier, you're going to have to be willing to give up control and accept advice and help from others. Think of it this way: 25% of a big number is better than 100% of a small number.
Do you think Zuckerberg could have built Facebook to where it is today without help from more business-savvy people? Do you think Google would be as big as it is today if Page and Brin had maintained control instead of bringing in Schmidt to be CEO? Do you think Woz would have built Apple without Jobs? These are all examples of extremely intelligent people who needed help to grow their company. Like I said, it is impossible to be good at everything.
Running a startup is like raising a kid. It's hard enough to do that under normal circumstance. Why make it harder by making yourself a single parent?
Just some advice based on my own experiences, observations, and advice I've received from others.
For "interest", venture capital wants to see, first, just the UI/UX and then, second, ComScore numbers. For the UI/UX, that's the last software I'm writing, not the first. Nothing else I said here interests venture capital at all. Nothing. Actually, for the advantage and technological barrier to entry from the math, they just get torqued because they know it's not within their 'deep domain knowledge' and they want to feel that they know more than the entrepreneur. And I tell them the problem I'm solving as the users see it, and that doesn't impress them, either.
It's like, and I suspect, and this is much of the reason for the recent rise of the angels, that to stop another bubble the few, large LPs got together and wrote on the back of a 3 x 5" card the dirt simple criteria for each of seed and Series A, B, C, D. The criteria rule out considering anything I've said here or even the users' problem I'm solving. With these rules, could teach a dog to be a VC in a weekend. So, the LPs won't let the VCs actually 'review' a business plan and apply 'judgment'. So, that's why some VCs blog: That's about the only advantage they have left, that is, to get the 'deal flow', and otherwise are reduced to being just another dog following the 3 x 5" card.
I've seen a lot in business, from small companies to large and successful ones to large and sick ones. I helped start a large, successful one and learned a lot there. One of the main things I learned is, there isn't a lot to learn. I've been a B-school prof where I also ran a consulting company with four important clients in business. So, for someone doing a startup, I am relatively well informed on business and how to start and run one. None of this background impresses VCs either!
Actually, I learned that the most important thing to learn for business is People 101.
Besides, on Main Streets all across the US there are sole proprietors being successful. Actually, they own nearly all the yachts under 100 feet long. My business is really different only in the advantages of Moore's law, the technical core, and the potential. They do it; I should be able to do it, too.
The Koch brothers never went public; they just ran their business.
That I have some technical knowledge doesn't really mean that I have too little 'business knowledge'!
Sure, if I go public, then I'll have staffs helping me entertain Wall Street, the institutional investors, the WSJ and the NYT, CNBC, keep the SEC happy, keep me out of trouble with Sarbanes-Oxley, entertain Congress, etc.
I'm eager to learn: E.g., I work hard on the Internet to learn, from AVC.com, TechCrunch, VentureBeat, C|Net, SAI, HN, and the best I can get. AVC.com has some of the best information, mostly from people who comment.
Early on I won't try to do bookkeeping, accounting, or legal. ASAP I will just get the best paid technical support I can from Microsoft and deliberately resist learning all the details of Windows Server administration. Same for Cisco's products: I have a basic understanding of TCP/IP, DNS, and even MPLS and have written socket software, but I will work to 'outsource' as much as possible on networking to the right people at Microsoft and Cisco.
But a pattern is developing: I have to get this project to some ComScore numbers with just my own eight fingers and two thumbs. And I won't want the ComScore users much before I have some ads to get me some revenue. Then my planning arithmetic indicates that if I do get users, early on the 'free cash flow' from two weeks of revenue can pay to double the capacity of the server farm.
Why? Take reasonable values for charge per click or charge per 1000 ads displayed, take the small size of my simple Web pages, add in current prices for servers and bandwidth, and conclude that in two weeks a busy server can make money enough to replace itself. The key is to get the server busy!
So, after a few doublings, there's a good chance that I'll have so much revenue that a Series A check for a few million dollars would not be very interesting.
So, the venture capital people want me to wait so long that they are very close to having me wait so long I won't want venture capital. Or if I've been live for a few months and go for venture capital, then it's because my business is less good than I planned!
For my having a good, complete, well rounded team, on the Internet no one can tell that a Web site is run by a team, one person, or a dog. If people really like the Web site, then I'm awash in money to 'outsource', hire, etc. Really, the next key, nearly both necessary and sufficient, is just for people to like the Web site.
Maybe the LPs think that VC is like PE where just wait for an owner of a good business to go for fast women, slow horses, and cheap booze, get into a nasty divorce, etc. and then buy him out cheap, put in their own CEO to do just routine things, rebuild the company, and sell it.
There is the example of Plenty of Fish: A romantic matchmaking site in Canada, ads just from Google with no additional targeting, $10 million a year in revenue, two old Dell servers, and one guy. So, right, his business is worth ballpark $1 billion.
For examples, I liked Gates at Microsoft, Ellison at Oracle, Viterbi at QUALCOMM, and a few more.
For Schmidt at Google, it seemed to me that his main role was to make the investment bankers and institutional investors happy. Now he's booted himself upstairs. Seemed to me that Page and Brin were doing fine and now are again.
Even if my company could be the most valuable in the world, I don't much care. When someone offers me $1 billion for it, I'll sell, buy the yacht, and write music. The buyer can take the company and run it up to $300 billion if he wants. If no one wants to buy and I have a Board eager for the rest of the value, then I'll stay and help, out of obligation, and for some pride in seeing the success, but it will delay my work on music and physics.
There is an issue of power and public responsibility: A company or a person worth $300 billion has a lot of power. I would want at least to see that the power was not abused. It would be ugly music to write to leave the company and have someone else take the power of $300 billion and abuse it.
For some of the best advice I've seen in business, I'm not super thrilled. I'm crossing rough waters in a small canoe, and it's not clear to me that having someone else in the canoe will make the crossing easier. I just have to get that canoe across the water. That work is my responsibility, and there's no way I can 'outsource' or really even share that.
I don't understand why people continually compare the two as if they're mutually exclusive. Why can't we just settle on the fact that they're both equally critical to a successful company?
Is it because some outlying success stories seem to be deficient in one and it causes people question whether that was necessary in the first place?
The neverending rebuttal of "a great idea is enough".
It seems to me that having a great idea would help... but I wonder if even that is necessary... I've been reading Christensen, on "disruption", and there's the concepts of:
- targeting non-consumption (or non-consuming contexts - e.g. users of phones, but not while mobile)
- asymmetric motivation, where low income or low margins or small markets etc. look great to (poor) you, but a rounding-error to (rich) incumbents. So they don't fight you.
- independent value networks, where you don't tap into any part of the way the incumbents do business. You don't use their suppliers; you don't use their sales channels, you don't target their customers. To use a value network, you have to fit in with it, you have to integrate with; you have to work with it - these means that you must constrain your business in some way. This has two effects: 1. if it's their network, your business will be more similar to theirs, and therefore easier for them to adapt to it, to attack you on your home ground, and co-opt the new market you created; 2. being constrained by some pre-existing way of doing business makes you less adaptable to whatever this new market needs, what the customers need, what the business idea needs. You are hampered from serving it.
- asymmetric skills, which is just the experience you get in making your particular business work. It's the accumulation of engineering problems you solve each day and gradually adapting your approach to suit them and having insights about how to reorganize to deal with them best; it's the accumulation of knowledge about your customers, what they want, how to reach them, what matters to them, what doesn't, along with repeatedly getting that to work. And so on - it's every part of making the business work. The more adapted you are to your customers and needs (and assuming this is different from what your competitors do - see the previous point), the greater a barrier to entry it is. That is, you become the incumbent in this new market, and you have the strengths that come along with that.
So, you might have an idea of a new market, that meets the above characteristics, and it seems that that idea would be valuable: a great idea.
However... I'm starting to think that instead of all that careful reasoning, perhaps you could just say "here's my idea" (meaning an original idea, not a copy), and start doing what a business needs to do: find customers, serve them, improve your product, etc.
Wouldn't all the above points follow automatically, provided you just "did it your way"? And isn't that how so many of today's great companies started?
If this is true (is it? I don't know), then the idea doesn't matter at all, so long as it's yours (original and motivating). Any thoughts on this, esp if you've read Christensen?
Innovator's Dilemma is definitely the best - it was based on his PhD, it is brilliantly and simply written, and is also supported by actual data.
The above terminology is from the 3rd book, after Innovator's Solution, called "Seeing What's Next". Although he's refined and modified his theories with experience in applying them to many real situations, the book is no where near as well written, even includes a constant stream of ridiculous puns (in the aviation chapter), often uses casual language - and doesn't provide any actual data. Also, the second two books are only co-authored by Christensen (who is a Rhodes Scholar btw).
You may also like Moore's "Crossing the Chasm" - the same ideas independently developed at the same time, but a how-to guide, rather than an academic description. He was a marketing guy, who noticed these patterns in how Apple etc succeeded. Andy Grove (for example) agreed with him. This is also a beautifully written book (guy's a PhD of English Lit).
After you finish that one, check out Blue Ocean Strategy. It's mostly the same theory but from some different points of view, and applied to many different industries, not just tech.
Also Chasm Companion complements these books very well (you may want to read Crossing the Chasm and Inside the Tornado which expand on some of the concepts in Chasm Companion, too).
Does anyone else know any other books that fit well with these and have great insights as well?
isn't this somewhat obvious? i mean, you could even have a perpetual movement figured out that it wouldn't matter if it didn't work. Now, the definition of "correct execution" may vary, from "does the job" to "awesome"
I just thought it relevant to point out Job's old resume(currently frontpaged) says that little "vision thing" as skills. Thats what I think "idea" is. And it's constantly reevaluated and adjusted.
I couldn't agree more... good ideas are a dime-a-dozen. What's rare are people who put in the effort to execute ideas, and execute them well.
People tend not to act upon their ideas because of the lack of time, money, or the right skills. I'm trying to help solve the third problem on my site www.iinspireus.com -- helping people find real collaborators/partners with skills complementary to theirs, so that they can realize their ideas.
So, since ideas don't "count", we should shut down the USPTO and tell the US DoD to quit classifying ideas for, say, stealth, laser gyroscopes for inertial navigation, drag-free satellites for better GPS, a lot in sonar, radar, and spread spectrum, etc. And we should tell the lawyers that when someone steals a trade secret idea they have stolen nothing that "counts".
Moreover, we should shut down the funding for NSF and NIH for more ideas.
And we should shut down the peer-reviewed journals of original research since they publish just ideas with usually nothing on "execution".
In particular, we should do away with the Ph.D. dissertation since its usual role now is to present an idea that is "an original contribution to knowledge worthy of publication" and where the usual criteria for publication are "new, correct, and significant" which must all be empty since ideas don't "count".
We should shutdown the Nobel prizes since they are given just for ideas that don't "count".
For Newton's second law F = ma, we should quit teaching that because it is just an idea which doesn't "count". Similarly for Einstein's equivalence of mass and energy in E = mc^2 -- doesn't "count".
For the Craig Venter idea of chopping up DNA, sequencing the pieces, and reassembling the pieces in a computer, an idea that got the sequencing of DNA done years faster, we should have ignored that.
For the idea of Reed-Solomon error correcting codes crucial to letting CDs work, it doesn't "count" either.
For public key cryptosystems, their role in Kerberos and authentication and in electronic finance, we should forget about that, too, since they are just ideas which don't "count".
People should quit looking for an idea to settle the question P versus NP since the idea, even if found, would not "count".
And the idea of using tungsten in a glass with high vacuum as an incandescent electric light would not "count". The tungsten idea came from the English chemist Swan, and the needed much better vacuum pump came from Germany(?). Then Edison could give up on baking hair to make carbon filaments and get on with electric lighting.
But, more is claimed: Apparently all ideas are equivalent, that is. don't "count:, and only execution counts. So, might as well start with a bad idea.
If start with a bad idea, then, yes, execution is about all that will "count". Just what is it about bad ideas the author likes so much?
This article is just another attempt of many to run down ideas by apparently someone with less than a weak little hollow hint of a tiny clue about what an 'idea' really is and what a good idea is how valuable it can be.
In fact, with a good idea, execution is routine and the idea was likely difficult to find and is about all that does "count".
There can be reasons for running down good ideas:
(1) They can be new and disrupt things.
(2) They tend to be difficult to find.
(3) In school in, say, math and physics, being able to find the good ideas needed to work the problems can be difficult and hurt the feelings of people who fail to find such ideas.
(4) In venture capital, it might be a good initial negotiating position to claim that "Ideas are easy. Execution is everything." and "Ideas are plentiful." in order to tell an entrepreneur with a good idea that what he has is easy and plentiful and nothing.
E.g., one of the most important steps forward in civilization, technology, and economic productivity was Euclid's 'Elements' on plane geometry, especially the Pythagorean theorem, e.g., crucial for trigonometry, Fourier theory, Nyquist sampling, JPGs, MPEGs, and music CDs! But, for anything so important we'd want a solid proof. All that takes is a good idea. So maybe the author would like to find such a proof, alone, without books, Google, or input from others.
I don't think he said they don't count, I think he said they won't make you money, which is different. Many of the ideas you listed were invented by one person and then capitalized upon by another. This doesn't mean the idea wasn't important, just that the person who got rich was the one with the implementation.
Yes he did. He said that execution is what "counts". The way he wrote this it is fair to conclude that he meant that execution is ALL that counts so that nothing else counts and in particular ideas don't count. He was careful never to admit that an idea "counts". He's being polemical, and you are bending over backwards to be literal and generous.
You realize that an idea can be important, but he didn't want to admit that.
That execution or implementation is important for making money with the idea is also not really true: That's why we have the USPTO. Patent licensing is big business, big bucks, and sometimes there's not a lot of 'execution' needed.
Yes, of course, in the usual situation, to get the big bucks have to 'implement' and 'execute'. We know that. Anyone out of the first grade knows this. What he wanted to imply was that execution "counts" so that the idea doesn't -- he wanted to remove the possibility of a good idea. Dumb.
I fully assure you, find a good idea for a one pill solution to any cancer, and the idea will be about all that "counts" and execution will be a very easy case of routine -- have some big pharma company do the trials, put the pills in the bottles, and hire the security guards to push away the desperate customers.
"Many of the ideas you listed were invented by one person and then capitalized upon by another. This doesn't mean the idea wasn't important, just that the person who got rich was the one with the implementation."
Sure: We are agreeing. E.g., without Swan and his idea about tungsten, poor Edison would have been pulling hairs to bake from himself, his employees, his wife, dogs, cats, etc. all as a waste of time. Similarly for the vacuum pump. Yes, I would hope that Swan would have gotten the big bucks, but he didn't. Still, the idea was crucial. Edison would have made no bucks at all from lights without such an idea. Or, just tell Edison to "execute": Right! Try to sell lights that last about 40 hours! "Come on Tom, just EXECUTE! Quit wasting time looking for ideas. EXECUTE! Ideas are easy and plentiful (J. Doerr). Execution is everything (J. Doerr) and all that 'counts'!".
In my project, I started with an idea. I'm attacking a problem nearly every Internet user in the world has and knows that they have. So far all the solutions 'executed' just suck, about like a 40 hour light. Maybe 500 million Internet users would take the problem seriously enough actually to want and use a good solution. Maybe 50 million would become really "engaged" (Fred Wilson, AVC.com). A few million will like the solution enough to become 'addicted' as is common in games (although my project is not for a 'game').
The crucial obstacle to a solution is a good idea, just the idea, just "How to do that". That is, the reason the current solutions suck is closely analogous to Edison not knowing about tungsten.
How to do that ain't easy to discover. What's needed is a good idea not easy to find. With the idea, the rest is routine. If others could find the idea for a good solution, say, think of tungsten, then the problem would have been solved long ago. Net, the idea is nearly all that really matters.
Well, I invented and have the idea. No, it ain't computer science or artificial intelligence or machine learning or some heuristic or any of that nonsense. Instead the solution is some original applied math with some advanced prerequisites. Nearly no one in computer science, including the chaired profs, has the prerequisites and certainly hasn't done my original work.
So, what am I doing? Today it's struggling with some UI/UX details and ASP.NET, and I'm a beginner with ASP.NET and going slowly. Thousands of others here on HN could take my idea and core code (I have the software for the idea in good shape) and totally blow me out of the water on the rest of the software. But, I've kept the idea confidential! So, I've got time for my slow progress with ASP.NET! Again, 99 44/100% of what matters is just the idea.
So, right, I'm not giving away the idea, and my intention is to keep the idea confidential, in software well protected in a server farm and get the big bucks for myself.
A "discovery" is not a special case of an "idea"? That is, you are claiming that ideas necessarily don't build on each other, that is, are not 'cumulative'?
Here is the hidden point in this discussion: The article and much of information technology entrepreneurship want to say that the 'idea' is some, say, Facebook for dog owners or eHarmony for cat lovers. So, the 'idea' is some description from 40,000 feet up of some business, say, category. Maybe the idea is an Internet site for printing and sending via FedEx personalized paper napkins for big picnics. It seems to be such 'ideas' that J. Doerr finds flooding his e-mail inbox and that lead him to conclude that "ideas are easy" and "plentiful" and "execution is everything". Doerr, and the article, are not thinking of tungsten for electric lights, Reed-Solomon codes for CDs, or F = ma as 'ideas' or any such ideas as important for business. That's DUMB.
You are being fooled: The article was deliberately ambiguous.
What I said is fully appropriate: "plentiful, easy, worthless, and don't 'count'" as a statement of one of the claims of the article. That is, the article flatly stated that:
"It is the manner in which ideas are
executed that counts."
So, the execution "counts". So, he is saying that the ideas don't "count". That is, what he was saying is usually interpreted as 'execution is ALL that counts'.
The quote you extracted is even self-contradictory: In his first sentence he says that ideas are not "enough" so admits that they are necessary but in his second sentence is back to saying that it is the 'execution' that "counts" essentially saying that the ideas are not even necessary.
He's just spinning nasty stuff, and trying to write in ways that are slightly difficult to debunk.
But his point is quite clear: He believes that ideas are worthless, that execution is everything, that is, all that "counts".
Worst of all, he omits any concept that there can be bad ideas and good ideas.
Basically he wants to run down ideas.
As I explain, and as in the link I gave, this running down ideas is common in business.
My point about "equivalent" is a reasonable implication to draw since to him execution "counts" and, thus, ideas don't and, thus, are equivalent.
That is, what he was saying is usually interpreted as 'execution is ALL that counts'.
Again, as the article clearly states, "Having a great idea is not enough. It is the manner in which ideas are executed that counts." The idea is clearly mentioned here as part of the formula.
I'm sorry, but you're just making bizarre leaps from what the article said.
You have a serious emotional block here. I addressed exactly this pair of sentences.
Read what I actually wrote.
Or yet again, push down your emotional whatever and just pay a little attention: The second sentence, given how the English language works, is essentially saying that execution is ALL that "counts" so that necessarily the idea doesn't count. So the second sentence contradicts the first. He's writing garbage, and you are reading into it what you want.
I'm not talking about an idea I had that resembled a current service. No, I'm talking about the same exact actual products and services. I'm in a perpetual WTF? state at how identical my ideas are.
Difference? Execution.
This also has made adopt a posture about the complexity of the idea itself. IMHO strip you idea of unnecessary complexity (advanced functionalities, etc, etc) in order to be able to execute it as good as you can as fast as you can. Simplicity allows better execution.
This seems obvious but is damn hard to accomplish.