Evan Williams: Will it fly? How to Evaluate a New Product Idea | Hacker News Reader

Evan Williams: Will it fly? How to Evaluate a New Product Idea

evhead.com

70 pointsstaunch23 comments

Best line: "The web is getting so damn big, what seem like niche ideas can be very decent businesses."

Evan misses one very important component of evaluating an idea. What is the effect of mass liquidity on the value add. Is the value gained by the first user identical to the value of the millionth user.

Google,Amazon - even if I was the only user in the world using them, would be just as useful for me.

On the other side of the scale - eBay,Digg are totally useless if I am the only one using them.

You mean "Network effect", don't you? I think he buries it in "Deepness".

His quote: "practically any application, once people start using it, can be used as a lever to more activity and benefit delivery. Being smart about what you're leveraging is key."

But my point is that some applications aren't viable at all without this critical mass and some are. While obviously everyone wants more users, for some ideas the critical mass requirement is a very high initial obstacle and for some it isn't.

I'm not saying you're wrong; I'm saying there's a hoary old cliche that you can use to communicate the same thing: "He forgot to add Network Effects". Ebay relies on a network effect; Amazon doesn't.

I don't imagine that he forgot it; I think he's assuming most web startups rely on it. He is, after all, Pyra, Odeo, and Twitter.

Uh, why are people voting this down?

> some applications aren't viable at all without this critical mass and some are

That is exactly what "network effects", or "positive network externalities" means, and they're critical to many, but not all, information goods.

Are you sure? I bet Google uses users' clicks to improve their searches, as well as Amazon's search, recommendation systems and used books heavily rely on crowd.

First of all, Google doesn't use clicks to improve their searches, because they have no way of knowing on what you click, the links on the search page are direct to the websites that interest you and the JavaScript function (clk) doesn't send any info to GOOG. They do monitor AdWords clicks obviously and do a ton of optimizing on it, but that isn't where I get my value from. ( a possible exception is clicks to sites with Google AdSense where they can monitor both ends ).

And I don't care what other people read, maybe its my weird taste in books, maybe it is my extreme individualism, but I really don't give a damn. I just want to be able to find a book and get it to my doorstep as soon as possible as cheaply as possible and as hassle free as possible.

Most importantly the point wasn't the examples it that there are ideas that need very little users to be useful to their end users - think 37signals apps. And there are ideas that need a lot of people to have any effect.

When I search for "Hacker News" and do Copy Link Location on the url, I get this:

http://www.google.com/url?sa=t&ct=res&cd=3&url=h...

I'm pretty sure Google knows what I'm clicking on.

You are not looking at organic search results. The link is for non-adwords sponsored links. (notice the colors are different ).

This is a paid advertising option, so it obviously has to be tracked so they can price it correctly.

You think PG is buying AdWords ads now? I'm aware of the difference between paid and organic search results; give me some credit.

I bet you're not signed into Google when you try it. I'm signed in for gMail, Google Reader, etc. When I mouseover any of the organic search results, it shows the normal link in the status bar. But when I right click on it, it switches to the google address and that remains if I move the mouse and mouseover again.

No offense, but these were the only links I could find with that format.

Just tried it while signed and still the link when I do "copy link location" is direct. Maybe they only do it on a small focus group or for users in certain countries (I'm in Israel).

Ah, the joys of having fine grained control over every user's experience with up to the minute modifications. Hooray for web-delivered software!

For me it says: http://news.ycombinator.com/

I think Google sometimes uses direct links and sometimes redirection.

only this article applies to web apps...

Call me old fashioned, but it's hard to take an article about evaluating ideas seriously whens it never mentions "customer", "demand", "want", or "need".

This is typical geek thinking. What can "I" do? What's cool to "ME"? How do "I" succeed?

Slow down for one minute and look at things from the eyes of your prospects. Better yet, ask them. THAT'S the best way to evaluate a new product idea.

He does discuss "Is it clear why people should use it?" and whether the product is personal compelling.

"Is it clear why people should use it?" DOES NOT EQUAL "Is it what people want?"

This is not an argument in semantics. This is the ABSOLUTELY, POSITIVELY, MOST FUNDAMENTAL question in business. If there is no demand, you are dead. Making something that creates demand is about as likely as winning the lottery (albeit as lucrative).

I'll stand by my original post. You must satisfy a demand. The best way to identify that demand is to ask. The best way to ask is to know your prospects. None of this is ever mentioned in the original article.

Whoever downmodded my original post still doesn't get it. I am interested in using technology to build a sustainable business, not a hobby. So I care about my customer first. That's the whole point.

The good news is, you don't have to know about the demand and neither do the demanders! As was mentioned a couple times in the "Will there always be more room for web startups" discussion last week (http://news.ycombinator.com/item?id=89055), customers often don't know what they want until they see it or use it. That unknown demand is why the trial and error of the marketplace always tolerates "crazy" ideas. I agree with your comment that you should ask and find out what customers want, but you can definitely succeed if your instinct matches an unexpressed demand (Twitter, Facebook) or if you can envision a product better than customers can express it (Apple)

"customers often don't know what they want until they see it or use it"

Absolutely.

And I want to be the one who gets to know them, understands their pain, asks the right questions, and prototypes. I want to be the one who is told, "Yes! That's what I want!" instead of those who depend on trial and error or instinct.

And sometimes, I just have to get up off my butt and away from my screen to get that done.

some coverage here

http://scribblesheet.co.uk/article/is_your_idea_any_good

Will It Blend?

http://www.willitblend.com/

How do you downmod "Will it Blend"? Reexamine your priorities!

Reply on news.ycombinator.com