Ask HN: How do you find out what users want?
I read so many times the mantra "listen to your users" or "make products users love", but I don't understand how do you actually do that?
21 comments
As a UX designer finding out what users want is a large part of my job. Firstly you look at your own goal; let say you are tasked to build an app for commuters who travel by car everyday. The client is a navigation company.
First you identify who your target audience is, your client might know some about this but it's important to also talk to other related people. So you start to talk to commuters, but also moms who bring their kids to school, traffic cops, people at highway stops, anyone. At the beginning you talk about their experiences, what do they do and why?
Then you want to formulate insights from these interviews and what you have learned from research, for example: Commuters get annoyed by loud noises and honking while standing in a traffic jam.
Then you could ask people who you talked to previously to come in and help design something. (this is also a good moment to ask more questions of you have those) Do a session about people drawing their experiences on a white board and brainstorming together with you and other participants in how this might be solved.
Then you build a MVP and test this with your commuters
Then you improve and iterate and test this again.
If the vibes are good you continue to develop your product.
(this is in short how it works, if you'd like to learn more and get specific tools and techniques tech out this: http://www.designkit.org)
First you identify who your target audience is, your client might know some about this but it's important to also talk to other related people. So you start to talk to commuters, but also moms who bring their kids to school, traffic cops, people at highway stops, anyone. At the beginning you talk about their experiences, what do they do and why?
Then you want to formulate insights from these interviews and what you have learned from research, for example: Commuters get annoyed by loud noises and honking while standing in a traffic jam.
Then you could ask people who you talked to previously to come in and help design something. (this is also a good moment to ask more questions of you have those) Do a session about people drawing their experiences on a white board and brainstorming together with you and other participants in how this might be solved.
Then you build a MVP and test this with your commuters
Then you improve and iterate and test this again.
If the vibes are good you continue to develop your product.
(this is in short how it works, if you'd like to learn more and get specific tools and techniques tech out this: http://www.designkit.org)
Some of the replies have good answers, but one other thing you can do is build a prototype, and do some user testing to garner feedback.
If you have to ask this question, I would strongly suggest looking for a co-founder who already understands this stuff. The technical steps to do customer research and iterative product development can be taught, but in practice you are entering a pay-to-play minefield of false signals that eats its young.
Learning and getting good at user research takes time and money, mistakes will be made. Don't forget your competitors are doing their own user research and iterating too...In my opinion, many startups die right here because of founder optimism bias on these exact "soft skills" questions you asked about. Founders, who are smart, skilled at many things and fast learners, still regularly underestimate time and cost to get good and be better than their competitors at this stuff (assuming the founders remember they have competitors).
It's sad but many startups run out of money here still learning to tie their shoes - they never even enter the real race of running a business (though they won't admit this usually). Anyway, if you still want to learn how to "listen to users" and "make products users love" a full explanation is beyond the scope of HN, but here is some basic info/an analogy how it works:
Guy: Hi, sorry to bother you miss, but will you marry me?
Girl: Um, What? Who are you? Why are you bothering me? I don't know you and anyways I have a boyfriend already. Bye.
Guy: Wait, please. How long have you been with this boyfriend? Is he perfect for you? Marriage material? Because I can be perfect you.
Girl: Haha. No, he is not perfect. Far from it. I like him but don't love him. Communication kinda sucks from him, he ignores my texts or just takes way too long to reply to me. Like over 10 minutes before I get a reply. We have other problems too, typical shit couples deal with. Wait, who are you again?
Guy: Sorry. Hold on. Writing down everything you said. Communication...Not fast or reliable...Please tell me more of your problems. I want to marry you. I am very passionate about marriage. I really want to eat, sleep and serve you. Help me, help you?
Girl: You're funny. But sorry, you are not my type physically. I have to be attracted to person I am with. Also I want to marry a guy I love, Someone who is my best friend... I want someone who understands me, anticipates my needs before I even know I have them, someone who texts me back in less than 10 minutes. I want someone old fashioned. So like you should pay for everything at least at first. Really I want someone who is 10x better than anyone else, including my current boyfriend, if you can't offer that don't bother proposing again.
Guy: Got it. I can do plastic surgery. Consider it done. Also, I wrote down the list of things you said you want in a husband. And I made a note with all the problems you mentioned you were experiencing and hated. I left space in case you think of anything else. Can you give me a few weeks to work on myself, looks and personality and other qualities, before we meet so I can propose again?
Girl: Wow I have never seen someone want to put in so much effort for me. Guys are so selfish usually. Someone who put me first...if true I think I could see getting serious and loyal to someone like that, maybe even love. But can we meet in 2 weeks instead of a few weeks? If you are making changes for me I want to make sure you are on the right track with small changes instead of waiting to meet only after big changes, no? I mean you don't even know me.
Guy: Good point. I was thinking about that. I actually want to know much more about you. More than just knowing your thoughts on relationships and marriage, I feel like the more I know about other aspects of you the better I can understand you and make changes I know you want. Any chance I can learn and know literally everything about you? including your private conversations with your friends?
Girl: Ya I don't mind if you know everything about me including my most private info. Here is my Facebook info to start. Really looking forward to you selling me this marriage and anything else :)
The End. Repeat for 10+ years.
Learning and getting good at user research takes time and money, mistakes will be made. Don't forget your competitors are doing their own user research and iterating too...In my opinion, many startups die right here because of founder optimism bias on these exact "soft skills" questions you asked about. Founders, who are smart, skilled at many things and fast learners, still regularly underestimate time and cost to get good and be better than their competitors at this stuff (assuming the founders remember they have competitors).
It's sad but many startups run out of money here still learning to tie their shoes - they never even enter the real race of running a business (though they won't admit this usually). Anyway, if you still want to learn how to "listen to users" and "make products users love" a full explanation is beyond the scope of HN, but here is some basic info/an analogy how it works:
Guy: Hi, sorry to bother you miss, but will you marry me?
Girl: Um, What? Who are you? Why are you bothering me? I don't know you and anyways I have a boyfriend already. Bye.
Guy: Wait, please. How long have you been with this boyfriend? Is he perfect for you? Marriage material? Because I can be perfect you.
Girl: Haha. No, he is not perfect. Far from it. I like him but don't love him. Communication kinda sucks from him, he ignores my texts or just takes way too long to reply to me. Like over 10 minutes before I get a reply. We have other problems too, typical shit couples deal with. Wait, who are you again?
Guy: Sorry. Hold on. Writing down everything you said. Communication...Not fast or reliable...Please tell me more of your problems. I want to marry you. I am very passionate about marriage. I really want to eat, sleep and serve you. Help me, help you?
Girl: You're funny. But sorry, you are not my type physically. I have to be attracted to person I am with. Also I want to marry a guy I love, Someone who is my best friend... I want someone who understands me, anticipates my needs before I even know I have them, someone who texts me back in less than 10 minutes. I want someone old fashioned. So like you should pay for everything at least at first. Really I want someone who is 10x better than anyone else, including my current boyfriend, if you can't offer that don't bother proposing again.
Guy: Got it. I can do plastic surgery. Consider it done. Also, I wrote down the list of things you said you want in a husband. And I made a note with all the problems you mentioned you were experiencing and hated. I left space in case you think of anything else. Can you give me a few weeks to work on myself, looks and personality and other qualities, before we meet so I can propose again?
Girl: Wow I have never seen someone want to put in so much effort for me. Guys are so selfish usually. Someone who put me first...if true I think I could see getting serious and loyal to someone like that, maybe even love. But can we meet in 2 weeks instead of a few weeks? If you are making changes for me I want to make sure you are on the right track with small changes instead of waiting to meet only after big changes, no? I mean you don't even know me.
Guy: Good point. I was thinking about that. I actually want to know much more about you. More than just knowing your thoughts on relationships and marriage, I feel like the more I know about other aspects of you the better I can understand you and make changes I know you want. Any chance I can learn and know literally everything about you? including your private conversations with your friends?
Girl: Ya I don't mind if you know everything about me including my most private info. Here is my Facebook info to start. Really looking forward to you selling me this marriage and anything else :)
The End. Repeat for 10+ years.
Launch the smallest, easiest-to-make version of your idea as the MVP and keep adding to it and taking from it while you try and grow it into the best version of your idea. Every time you put out a new version the people already using it are able to provide feedback if you talk to them. You still have to find users but you don't have to find very many to start.
You can actually watch this process in action on Steam's Early Access, a platform for selling games during development. Mobile app stores also provide that feedback loop where your users are able to share their frustration or happiness each time you update and may even segregate feedback by version for you.
Don't confuse the smallest version of your app for a pre-registration launching-maybe page either, that's the smallest version of a dream about building a startup.
You can actually watch this process in action on Steam's Early Access, a platform for selling games during development. Mobile app stores also provide that feedback loop where your users are able to share their frustration or happiness each time you update and may even segregate feedback by version for you.
Don't confuse the smallest version of your app for a pre-registration launching-maybe page either, that's the smallest version of a dream about building a startup.
Market research.. Focus Groups... Surveys to ask users specific questions... Actual face-to-face conversations
Agree with everything in both the replies.I am part of team thats building a platform tostreamline that process.
We are launching Feb 8 th.
MVP + iteration is the most straightforward answer.
Creating a user profile, matching it with a list of people and asking politely goes a long way.
Helps if you are a potential user of the solution as well
If you're Steve Jobs, you tell them.
You're also 5 years dead.
I like benologist's reply, as well.
I like benologist's reply, as well.
As I recall Steve Jobs said something more like "People don't know what they want. You have to show them." In any event, "tell", or what ever word it was, was not the imperative. If there is such a word it was the "informative."
This MVP thing is bs. Why do you want to find out what users want? Is it to make money? So you'll have to have a good idea and a good execution. 10 MVP's of stupid ideas with clunky implementation will get you nowhere.
A good approach for good ideas is to find a niche with a problem that can be solved with current state of technology but they are not aware of. Good old automation.
A good approach for good ideas is to find a niche with a problem that can be solved with current state of technology but they are not aware of. Good old automation.
You're only identifying a problem, what do you do once you have that idea other than build an MVP to solve it?
You build or use an already built proven work tool for the job. The first can take months or years, the second can be immediate. A MVP vaporware is a third but worst strategy, imo. The internet is already full of non-functional & clunky apps, it is difficult to get traction with one more (but can happen, for sure, I just don't think it is a good advice).
I can see the value in selling other people's software to solve a problem you've identified but I think that is not really the theme of this thread which is honing in on what your users want. If you are building something to test your nascent idea then is that really any different to building an MVP or "Minimum Viable Product"?
benologist, what I was pointing is exactly this. Why do OP wants to know what users want? Is it to make money? If so, I suggest to use an already built technology in place of developing a MVP, because if he is chasing ideas randomly, he probably will not be the best guy to develop a good product around it in a few months, as a lot of startup teams thinks they can and keep failing miserable. It is time to give advices in a different direction :)
I think you're covering pre-launch 'find out what users want' while an MVP and iteration focuses on the next phase of that question, after you launch something and want to start getting into the specifics, 'find out what my users want'.
And your point is that any next step after wondering would be a MVP. I don't think it is logically wrong, but I'm talking about "lean startup MVP's" that every student is shipping like crazy nowadays, flooding the internet with useless stuff. I don't mean that this should not be allowed, just that I don't think this proccess is a good approach to create a well succeded technology project. But it is good to learn more about tech, development, and fail fast :)
I guess it's more accurate to say the next step is the P from MVP, or the execution of the idea, the MV are optional.
MVP advocates putting your product out there at such an early stage to find out if it's actually a terrible idea before you commit to the entire upfront cost of building the complete product. The iteration enables your users and customers to influence product decisions as they arise. These can be positive additions to a project, or just as importantly signal it's a bad idea that isn't worth pursuing further.
MVP advocates putting your product out there at such an early stage to find out if it's actually a terrible idea before you commit to the entire upfront cost of building the complete product. The iteration enables your users and customers to influence product decisions as they arise. These can be positive additions to a project, or just as importantly signal it's a bad idea that isn't worth pursuing further.
When you listen to people it sounds so easy... i'm just gonna launch an mvp and iterrate on my users feedback. %99 of the time you won't even get one single user. So in order find out what people want - you need to build an MVP and go out there to collect your first user. We always forget about the second part. It seems like an MVP would attract a bunch of users by default, who aren't happy. All i have to do is to make them happy... at that stage you've already past seed fundings, medias, etc.
To build something people want the first real step is collect your first users. Then after it's everything people mentioned here...
To build something people want the first real step is collect your first users. Then after it's everything people mentioned here...
I do not believe there is a way, because there is no single answer. Indeed, the number of good answers is unlimited.
You have to look for unrecognized unsolved problems. It's the "unrecognized" that is the bitch in this problem. "Unrecognized" might just be the core feature of entrepreneurship. You have to work hard to uncover it. No easy answers.
I suppose there are established techniques in market research that would help. Ask someone who is good at market research.
You have to look for unrecognized unsolved problems. It's the "unrecognized" that is the bitch in this problem. "Unrecognized" might just be the core feature of entrepreneurship. You have to work hard to uncover it. No easy answers.
I suppose there are established techniques in market research that would help. Ask someone who is good at market research.