Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
How To Validate Product Ideas Before (And After) Building Them (kalzumeus.com)
84 points by austengary on Sept 20, 2013 | hide | past | favorite | 26 comments


This is good stuff. Most people know by now that they should "solve business problems" in order to create a decently successful product/service. But most beginning entrepreneurs are struggling to find out exactly what those "business problems" are.

I think the issue for most young entrepreneurs is that we don't have much business-related domain knowledge. Most teens and young adults have domain knowledge related to recreational activities such as skateboarding or playing guitar, but no business knowledge whatsoever.

Maybe we should encourage younger people to participate in business at a younger age so that we can use their fresh creativity towards coming up with new solutions.


I'm starting to wonder if I'm completely doomed selling to consumers. Consumers seems so fickle, they won't pay for anything software related it seems, even if it helps them a lot. Like they'll put down hundreds of dollars for a language class, but they'll sweat over a $5 app that would help their learning in said class.

Here's an example of a common attitude. Some guy launching his app on reddit (not me):

"Just use this: https://ankiweb.net

It is free and is better than OP's crap app.

you want money for your crap = keep you shit. i dont need it."


The only reason selling to businesses is better, is that by buying your product, businesses will make more money - that's why they don't have a problem forking over $X, since it will help them make $X*Y, resulting in a net $ gain.

If you find a similar niche for consumers - i.e. sell something to consumers that will help them make more money they will pay for it. Alternatively, if it will help them get laid more, they will pay for it as well. If your consumer oriented software does anything else other than helping users make more money / get more/better sex, then your consumers will definitely be fickle.


Alternatively, if it will help them get laid more, they will pay for it as well. If your consumer oriented software does anything else other than helping users make more money / get more/better sex, then your consumers will definitely be fickle.

Apropos:

http://www.jwz.org/doc/groupware.html

"Your "use case" should be, there's a 22 year old college student living in the dorms. How will this software get him laid?"


Or perhaps something that can clearly save them money?


My opinion, and I think I got this from Patrick indirectly if not directly, is that you need to be selling something to people that will cause them to receive more money than your product costs. So you cannot target language learners (especially the relative immaturity of Reddit) because it's a hobby, but you can target businesses and professionals because they routinely spend money on tools and services. Sadly that's all I know about it, so I couldn't tell you who that guy should have targeted with the app. Doctors? Lawyers?


Agreed. I would love to build a small business offering software/saas and live quietly with that, but I don't have a fine grasp of what people want to pay for.

So I'm building a mobile game, as entertainment will get paid for.


By contrast, I would LOVE to see an article like this about entertainment.

A solid, testable method of building entertainment people will pay for would be extremely valuable - to me and many other people.

(Patio11, any ideas?)


I'm sure this has been researched intensively by the media giants. A lot of pop music and rehashes of classic movies are a product of this I think.


good point. I do believe one of the biggest problems is the disconnection between the studies and the actual world. some are mitigating it by doing stages in industries at the end, but far from making it a good solution.

This aside, I found the article particularly well structured and most of the stuff he talks about are common pitfalls I've seen throughout my career in IT.


I enjoy patio11's articles immensely but I wish it would be made more clear that this is one of several paths to building a successful software enterprise. Going in and automating stuff people have been using Excel for is a straightforward way to build software people will be willing to pay for, yes.

But let's not forget the unique privilege we have of living in 2013 where we're just starting to get desktop-class CPU and GPUs with radios in everyone's pockets. Let's not forget that the things computers are capable of now were barely visible even a decade ago and a dream a decade further. I firmly believe in using computers to solve real problems. But I also think that you need to look at the problems you're solving and think about how you can truly leverage the unique capabilities we have now to build transformative things.


I don't disagree, but that's hardly a roadmap, or teachable.

Patio11 is trying to give people something actionable, I think.


I don't think patio11 was suggesting that automating spreadsheets was the only / best way to build a business. For me the thing folk should be taking away from the piece is that understanding your customer, their problems, and the channels you need to have to reach those customers before you build and launch your product is a rather good idea ;-)

Because that's the mistake I see companies make again and again. They build something that nobody wants, or that nobody can use in way that solves their problem, or solves a problem for people they cannot sell to. Find that shit out early - not late.


I am writing a book on different ways to do this: http://www.howtofindsaasideas.com/

I hope to meet the great patio11 at Microconf soon btw, looking forward to it :)


Aside from starting on and solving business problems, I think it's hard to tell where to start.

I think it might be worth adding that aside from the cliche "solve your own problems" that you'd typically get. One possible approach might be to look at technical problems you're interested in solving and figuring out where tech like that is used and working your way down to the business case like that. This of course applies to the programmer out there. For non technical, it might be worth mentioning starting from hobbies you enjoy and blogging about those. From that you might be able to explore and ask different kinds of questions that people might have.

Your best business idea will probably come by accident.

It may not be an optimal approach, but it definitely got me started.


One thing missing from this is to run surveys. Someone posted on here before about how they used Google Consumer Surveys [1] to check the pulse of their idea. It's something I highly recommend that can be done for relatively cheap before you even get started.

[1] - http://www.google.com/insights/consumersurveys/


I have given up on worrying myself if my products are good enough or have market fit. the biggest problem / reason why customers are not paying me is not I have not got market fit wrong but I am scared - of failing or succeeding or...

so - 12 startups in 12 months - quickly learn and quickly


Absolutely agree and that's what I've been doing while looking for jobs. You learn a ton and you might just hit on something that works.

A couple thing I've learned is that you should never thing that an idea is "the one" and that a mindset building to learn is better and a mindset of startup. You never know what project is going to take off and by working to learn, it keeps motivation up when things fail.


Wait. What? You've started 12 startups in 12 months? Basically meaning that you've done one startup project a month?


no - will do.

basically we seem to be entering a world where one person can launch a small profitable online business that takes minimal support. if so what needs to be templated / automated and what really is irreducible

it's worth a go. if I find out different I find out different.


Quickly learn what? We can't eliminate worry or fear whatever we choose to pursue. I think what prevents you from succeeding is not business related at all.


This is great advice and it is fantastic because it is specific and actionable.

I would love to see one tailored for validating consumer products rather than B2B ones.

A good amount of this is definitely applicable to validating for consumers though.

I especially like the section recommending that a product and the problem it solves be described fully in text before any code is written.


Agree, so much of the advice on HN is for B2B. I get that the success rate for B2B is higher, but some B2C advice would be very welcome.


Creating products that solve problems is great but you will not be in the "blue ocean". Did twitter solve a problem? No. Apple? Nike? Coke? Nope. They created great products that turned into a problem if you didn't own them.


Nice piece. I cannot over emphasise how amazingly effective actually getting off your butt and finding/talking to potential customers is.

A few random additional thoughts / tips.

1) If I'd have been in Patrick's shoes I'd have gone out a little earlier and start talked to folk with more open ended questions - before even building the 2 page demo. Things like "Can you just talk me through your typical day?" or "Tell me about a bad day?".

Why? If the problem you care about is important to those people - if it's problem #1 or #2 - it will come up in conversation naturally. If it doesn't then that is telling you something useful and interesting.

If you start the conversation by talking about appointments that is where it will go. Once you start talking about it with them it naturally becomes the focus of the conversation and any issues and problems will pop up and seem important - even if they would normally not care about them in the slightest.

More importantly - because the conversation is just around appointments - you won't here about the other major problems in their life. Maybe the biggest issue is dealing with staff who don't show up, or managing the shift rota, or keeping track of client preferences, or whatever.

And then you can dive in with something like "Can you tell me more about that?" and discover this lovely problem that the customer actually cares about.

2) On the "talk to people in the target industry" another sneaky way to find folk is to see if you can find a $industry conference you can get to. You get a target-rich environment for $industry and they're often a bit more informal and happy to talk (or drunk - which also works ;-).

If you don't feel you can attend/crash the conference itself then find out what the conference hotel(s) is/are and hang out in the lobby. Or find out what the popular post-conference bars are and hang out there.

(This is, of course, an instance of the more general advice of "hang out where your customers hang out".)

3) If it's more than just you I'd recommend talking to folk in pairs. Most people find it really hard to both engage in a conversation and take notes at the same time. Having one person do most of the talking, and somebody else do most of the writing makes it much easier to capture the information. Having mixed-sex pairs also helps in approaching strangers too.

4) Dress slightly smarter than the default for wherever you're doing the interviews. It makes a real difference in how you're treated.

If the idea of doing interviews still scares you and you'd like some concrete advise on technique I'd consider reading this really short pocket guide from Andrew Travers http://www.fivesimplesteps.com/products/interviewing-for-res.... It's not written for budding startup/entrepreneur folk but much of the advice is applicable. Bonus is that it's very short.

If you you like that book and want to dive deeper Steve Portigal's "Interviewing Users" (http://rosenfeldmedia.com/books/interviewing-users/) is a great book that covers everything in much more detail. Again - not written for startup folk but there's a lot of great advice you can steal.


Patrick,

How did you transition from that first AR client to the next? Did you simply close them on the spot?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: