[flipside]

Reading essays like this drives me crazy.

No, it's not because pg is wrong or I disagree with anything he's saying, it's because everything resonates with my startup (Tinj.co, interactive movie ratings) but after 1000+ pitches I still can't succinctly articulate how I know this.

I could probably make a checklist for the criteria of great startup ideas (seems like a toy: check, personal problem: check, inspiration from other fields: check, etc) but I have a hard time imagining that to be persuasive. I guess all we can do is push on towards a beta and let traction speak for itself.

Truly disruptive ideas are difficult to communicate because most people do not have the mental framework to truly grasp them. I guess that's just the catch-22 of disruptive ideas.

[tomhoward]

Fundraising & customer acquisition are like dating; logic doesn't get you very far. The points PG makes might be valuable in convincing yourself that it's a field worth pursuing, but for convincing people to give you their attention and their cash, you need a different approach, which will usually involve some combination of: stories, charisma, status, social proof, exclusivity, to name just the first that come to mind.

[flipside]

You're right, convincing others is a whole different ball game. Since I clearly suck at it, I've done my best to suck a little less each time I try.

Here's hoping pg writes that essay next.

[tomhoward]

Don't hold your breath :)

The best material I've found about this kind of stuff is: Robert Cialdini's "Influence", and some of the less creepy pickup material, particularly to do with "inner game". Also, this new book by Australian business journo Valerie Khoo looks interesting:

http://www.amazon.com/Power-Stories-Build-Business-ebook/dp/...

[lubujackson]

Try the Richard Feynman approach - if you really want to understand what you're doing, try to teach it. Like, write a manual about what your business does, explain it to mom, etc.

[snprbob86]

+1 for teaching. I find that explaining deep technical problems to people who are smart, but otherwise clueless about technology, forces me to understand the issue myself at a much deeper level. My girlfriend is the best rubber ducky debugging buddy ever.

[nekopa]

Also check out the book "The Art of Woo" Goofy title, but a very interesting book on selling ideas.

[javajosh]

I have four questions I'd like to ask you about your startup:

1. Does it help people address their fears? For example, are they afraid of not being heard? Or missing the next great movie?

2. Does it help people feel loved? For example, giving them an opportunity to warn people away from bad movies, or to offer unique insights?

3. Does it help people create or curate beauty? Movie ratings themselves are not very beautiful, but perhaps the personal curation piece is important to some.

4. Does it help people to eliminate ugliness?

...and I wrote this before checking your site. After seeing it, I have this sinking feeling that you are not really serious about your startup, as the homepage makes absolutely no attempt to describe what it is. Cold sign-ups are very 2004. honestly if I had checked first I wouldn't have bothered replying, but since it's already written, here you go. Hope it helps.

[flipside]

Thanks for the questions.

1. A big reason more people don't rate movies is because it's unclear how or if it actually helps others (my vote doesn't matter syndrome) or themselves (black box recommendations).

2. Validation is a big motivator, seeing that others agree with you or trust your judgement is huge.

3. Funny you should mention curating beauty, individuals will be able to share and discover based on people that share the same taste in "beauty" in the context of movies.

4. Reading tons of reviews or trying to decide if 79% on RottenTomatoes or a 3.8 stars from something else means you'll actually enjoy a movie is a painful (or ugly) experience to most people I've talked with.

I assure you, @javajosh, we are taking this very seriously as a startup. The launchrock signup page is not the primary way we're finding beta testers, it's merely sufficient before we start a closed beta.

Getting back to the bigger picture though, we're really building a system to efficiently share opinions and see what other people think.

[byoung2]

I'm an avid moviegoer (50+ theatrical releases a year) and I completely ignore movie reviews (and it's getting this way with Yelp reviews too) because they lack context. What I mean by this is that a thumbs up/down or star rating is useless unless I have some background on the person giving the rating. For example, for someone to give Prometheus 2 stars when they haven't seen any of the Alien movies is different from an avid fan. I'd like to see a list of other movies/genres the reviewer rated, and maybe some basic demographic info before I listen to their opinion. Not sure if you have that worked into your idea yet...

[javajosh]

Apropos of nothing, but a few times in my life I've actually gone to the theater to see a movie I knew nothing about. Nothing about the plot, the cast, etc. No trailer. It's a distinct experience, and sometimes highly pleasant. (I say "Heat" this way. Which was cool because the whole beginning of the movie is that long sequence with DeNiro stealing the ambulance, but I had no idea what the movie was, not even genre, and it was fun to try to figure out what was going on :)

[damian2000]

Agreed - that is an awesome experience - its happened a few times to me - Cape Fear, Training Day & The Fugitive - all great movies.

[flipside]

Context is definitely missing from most systems and it's one of our key differentiators. Our goal is to handle all those contextual details so you don't have to.

[shadowsun7]

Honestly, though, this idea seems to fall squarely into the 'yeah, I could see myself using that. Maybe' category.

This isn't a huge pain point.

[flipside]

If you take another look at my original post, my point was that I still can't articulate why this is such a huge pain. If it helps to put it in perspective, I abandoned a Nanoengineering phd doing promising work in medical diagnostics because our rating system solves an incredibly painful problem for me and I'm convinced it's a more important contribution to the world.

To make a (hopefully apt) comparison, what pain point did twitter initially solve? Sharing what you had for lunch?

[mdda]

Not to be too adversarial, but what problem was incredibly painful to you? Couldn't choose a movie? Couldn't choose a suitable couples movie?

Medical diagnostics, OTOH, could avoid a painpoint involving death - so that's an easier pitch.

[flipside]

See if you can follow this:

Ratings are a form of feedback.

Feedback helps people learn.

Therefore, if we can build a better rating system that lets people share more and better feedback, then we can speed up learning and improve the flow and evolution of ideas.

Also, this system wasn't designed for movies, it was designed for answers on Quora. Movies are just a better beachhead.

[lilsunnybee]

If not knowing what movie to spend your disposable income and leisure time on is an incredibly painful problem for you, you desperately need a bit of perspective.

[tdrake]

"we're really building a system to efficiently share opinions and see what other people think."

This is exactly it - well said. And as you said, "movies are your beachhead".

All you need is why what you do is different and/or better (which is still not exactly clear, unless it's finding peers).

It's sometimes tough for engineers to be focused less on being precise in description, and more inspirational to goals. I have to work at it every day.

[btilly]

You can boil it down a lot more than that. And not necessarily to such noble criteria.

http://www.jwz.org/doc/groupware.html famously boils it down to one question: Your "use case" should be, there's a 22 year old college student living in the dorms. How will this software get him laid?

[flipside]

Hypothetically speaking, the odds of getting laid after seeing a movie with a date probably go up if both people liked the movie.

Not exactly how we'd track it but the number of people that get laid after using our system would be a fascinating statistic.

[emmett]

Can you explain, in one sentence, what it is that you're building that's different from existing movie ratings sites?

Try this template: "it's like Rotten Tomatoes, except ..."

It's like Rotten Tomatoes, except you rate on 3 axes instead of just 1.

It's like Rotten Tomatoes, except you rate the movie by answering questions about what parts you liked.

It's like Rotten Tomatoes, except your ratings are automatically posted to Facebook.

What are you building? If you can't say in one sentence, I would suggest the problem is with your pitch rather than with the listeners.

[flipside]

It's certain that the problem is me, and my poor communication skills, I never blame the audience.

Describing what we're doing and illustrating the pain points aren't the same thing.

How about this: Tinj is a rating system that lets people externalize opinions about movies in context so they can receive comprehensive recommendations.

See? Told you I still suck at this.

Here's a problem we're solving: Given that opinions are complex, how do you figure out what other people think about a movie?

Read reviews? That's biased to critics and writers (less than 1% of the population).

Look at star ratings? What does 4-stars even mean?

Check out social media? How many tweets and blogs would you have to read?

Ask your friends? Which ones actually share your tastes and have seen the movie?

Unfortunately, complicated problems tend to have complicated answers.

[gfodor]

The goal of reading movie reviews is not to see what other people think about a movie. It's to decide what movie to see. Why not just say:

"Quickly and easily decide which movie to see."

Simple, to the point, and yes, it's still an unsolved problem.

[flipside]

Thanks, I'll give it a shot. I don't believe I've tried that particular wording yet.

[precisioncoder]

One thing I would suggest is when you mention it, have a link ready with an email signup so people like me who are intrigued can easily sign up to get a mail when the beta releases. The name seems fairly popular at least a google search for "Tinj" or "Tinj movie reviews" didn't work. (Although the hacker news thread came up with the second search)

[flipside]

I'll definitely need to take a day to improve our SEO before we launch. In the mean time, you can sign up at http://tinj.co

[abbasmehdi]

I feel your pain. It's hard to find the balance between saying what problem you're solving vs how you're solving it.

This is what I suggest you do. It might cost you ~$50 or so. Go to a coffee shop like Starbucks and offer someone to buy them their coffee in exchange for some feedback on the "new website you're building". Try to find normals (not super technical folks, unless that's your target audience). Explain to them what you're building, let them play with the site (hey this is also a nice UX test), take your notes whatever but in the end, ask them to explain to you what your site does. Maybe if you do this with 10 or so people you might see a pattern emerging (bonus: get pretty good feedback on product too).

I am not saying this will work, but this will not be a waste if the people you talk to are anything close to your target audience.

P.S. What Emmett is saying above is to give him a one sentence description that brings him as close as possible to cloning your site. Also, do you really need "externalize opinions"?

[flipside]

Different pitches for different people/situations. The problem is, people don't all have the same pain. It gets worse. Some people have gone numb from the pain and aren't conscious of it any more. The irony is that our rating system would be perfect for matching the right pitch to each person.

I've more or less mastered the 1-minute demo, hard not to since I've done it a thousand times (and a thousand different ways too). Trust me, I used to suck a lot more.

As for the "externalize opinions" line, just trying something new.

[lobotryas]

Hey man. I read your one sentence pitch and it left mr really confused. Writing this post to show you what confused me in case it's helpful in your fine tuning your pitch.

>How about this: Tinj is a rating system that lets people externalize opinions about movies in context so they can receive comprehensive recommendations.

What does "externalize opinions" mean? As someone who has never seen your service, I know what an opinion is, but "externalize" is a complex word with multiple nuanced meanings. As a prospective user I assume you mean "voice".

What does "in context" mean? Again, as a prospect I have zero idea what you mean by that, so I will assume you mean "on my site".

What does "receive comprehensive recommendations" mean? I guess it can be my FB friends (if I give another service access to mytriends list) or strangers or (worse) friends I have to make on your site before it brcomes useful. I'll assume "strangers or FB friends".

Put it together, and the pitch becomes (to me, a prospective user who had zero context about your srvice beyond this sentence and who you are trying to convince to visit your site): "Tinj is a rating system that lets people voice opinions about movies on the Tinj website so they can receive recommendations from strangers or FB friends.". This may be totally wrong, but it's the picture I got. To which as a user I say: "Big deal. I don't need this site when I have FB and RottenTomatoes."

I hope hearing some feedback from a stranger will help you refine your pitch.

[ams6110]

...externalize opinions about movies in context...

I'm not even sure what that means. Target your pitch to an 8th grade reading level. Yours sounds like the abstract for a research paper. And these days, your core value proposition should be able to be expressed in a tweet.

[precisioncoder]

That was a good pitch actually, convinced me to take a look at your site. My wife and I always have a very hard time deciding for movies, often because rating sites are worse than useless if you don't have mainstream tastes.

[flipside]

Awesome! The beta should start in the next few weeks.

[krichman]

I also thought that was a good pitch. Please make a post or something when you have the beta open, I'd like to try it.

[emmett]

Right, I understand that pain points and solutions are not the same thing. I didn't ask for you to describe the pain points. I really am asking what you're building. In a concrete way, what does it actually do?

[icebraining]

I use Criticker. It matches me with people with similar tastes and estimates how much I would grade a movie based on their grades.

[jcampbell1]

This idea strikes me squarely in the category of "Yeah, maybe I could see using something like that."

[flipside]

Of course you do, otherwise I wouldn't be frustrated. =P

[erikpukinskis]

"Maybe I could see using something like that" is equivalent to "no, I won't use that."

If they were a "yes", they would take out their wallet and offer you money. "Something like that" means "not what you are offering".

"Maybe I could see using something like that" is what people say when they want to be nice to you. Even "I would use that" is basically a "no".

[rythie]

Essentially it sounds like what you are claiming to be able to do, is recommend movies I like better. The 3-axis movie rating system or whatever is just how you solve that problem [and that might not work so you'd have to be open to trying a new method to solve the problem].

It's difficult problem to solve, you need lots of users and a good algorithm to improve on what IMDB or Rotten Tomatoes provide. In general I find that I'm happy with the ratings on IMDB and generally I will like the high rated films on that site - though I can imagine there are a lot of people that doesn't work for.

It would take a lot to convince me that someone had solved this problem (many have tried and failed), like a bit of hype from everyone or some rave reviews. However, in absence of that and seeing that you had done 1000+ pitches, I looked at your site and it has an invite form on it - so really you haven't even started yet.

[rickdale]

I am going to give my opinion just on a first impression analysis. You need to change the name of your startup. tinj.co? It has nothing to do with anything and inevitably you need to explain it with three words after stating your name, but my problem is your three words don't help define tinj.co. So I went to your website and I was like OK, tinj.co has to stand for something. But it just says a new way to rate. 5 stars is no good anymore. In conclusion, I am all for your idea, but your name its making me angry.

--note: I know nothing more than my rottweiler sitting next to me. The difference between me and most people is that I will tell you I know nothing. So just take the advice and do with it what you want.

[nicktal]

If it makes you feel any better, I've held the opinion for years now that ratings are broken across most services on the web. Jeff Bezos even acknowledged as much in a famous essay of his on how the distribution of one-to-five star ratings settle over time. Still, every new service sticks to the binary thumbs-up/down model or five-star. I'm sure you probably know the pitfalls of each, so good luck to you. I'm hoping you'll become the Disqus of ratings for every site. The analogy holds, interestingly, in my mind: comments had no consistency along UX or reputation before Disqus came along.

[flipside]

I used to start pitches by telling people we're fixing rating systems, that was usually enough to get people's attention.

I think you're talking about Yahoo and the J-curve, couldn't find anything by Bezos.

Thanks for the vote of confidence, we went through a phase of pitching ourselves as "Disqus for ratings" but the problem was it's hit or miss with people.

[npguy]

the one issue I see, is that the number of movies is not so large for current human feedback systems to fail.

Ratings in general is a different story though.

[fhe]

I'd love to read that Jeff Bezos essay. A google search didn't find it. Do you have the link to it, or do you remember the title? Thanks.

[pinaceae]

the problem you're trying to solve is discovering new movies for a user, right? i assume it is, so let me try to expand on why you might have trouble with your current offering:

- some people enjoy movies and are always looking for new stuff.

- their joy is watching movies, not rating them. like / not like right after the movie is ok. some disconnected process is not, this is work, not joy.

- what are the major, simple factors leading to the next good movie?

1., Same personnel involved. director, screenwriter, main actors.

2., I liked their previous work.

3., The major outliers are found by chance or word of mouth. then 1 and 2 are in play.

Now, does a web page solve this nicely? No. Much smarter would something akin to what Netflix is attempting. A XBMC/Plex/VLC plugin maybe that tracks HABITS. No effort involved. The plugin tracks the movie history and maybe, maybe asks at the end if you like the movie.

You can apply the same to songs, etc.

And yes, Netflix, Amazon, all prior art. But actually working.

Convenience is key for consumers. Seamless. Easy. Valuable. Instagram hooked itself into the photo making process. Flickr did not.

Good luck.