The startup I'm working on now is in the project management space, which I think is an unsolved and hard problem. And it's definitely something people want. When I started working on it I thought that it would be pretty easy, but I found that when I asked 10 project managers they all wanted subtly different software.
Everybody wanted good project management software, but unfortunately everyone I talked to also had differing opinions on what this would include. The hardest part was getting the details right, not from a coding perspective which is pretty simple, but from an interface and usability perspective. Nobody sees the code anyway.
Solving a problem in detail, and doing it in a way that people want is really really hard. Often much harder than coding it.
yes, the problem is like the famous quote of judge Potter Stewart when asked how to define obscenity: "I don't know, but I damn well know it when I see it"
Users don't know what they want, that is the problem you have to solve. They'll be able to recognise software that solves their problem when they see it, but they have no idea of what it should look like or what it should do when you ask them.
The hard part is taking all the itches you hear about from users, reading between the lines, inferring, asking additional questions, making mock-ups, etc. and creating a good cure for the itches.
I think that successful execution of this is the primary factor in successful software.
I have found that out of approximately 500 users (talking face to face) you get 1-2 who actually DO know what they need, how it should look and act. We hired one and listened intently to the other.
It's a long search, but really, really worth it. I agree though - the coding is the simple part.
Comments
This is very insightful.
The startup I'm working on now is in the project management space, which I think is an unsolved and hard problem. And it's definitely something people want. When I started working on it I thought that it would be pretty easy, but I found that when I asked 10 project managers they all wanted subtly different software.
Everybody wanted good project management software, but unfortunately everyone I talked to also had differing opinions on what this would include. The hardest part was getting the details right, not from a coding perspective which is pretty simple, but from an interface and usability perspective. Nobody sees the code anyway.
Solving a problem in detail, and doing it in a way that people want is really really hard. Often much harder than coding it.
And even when you think you have all the details right, people change their minds the next day!
yes, the problem is like the famous quote of judge Potter Stewart when asked how to define obscenity: "I don't know, but I damn well know it when I see it"
Users don't know what they want, that is the problem you have to solve. They'll be able to recognise software that solves their problem when they see it, but they have no idea of what it should look like or what it should do when you ask them.
The hard part is taking all the itches you hear about from users, reading between the lines, inferring, asking additional questions, making mock-ups, etc. and creating a good cure for the itches.
I think that successful execution of this is the primary factor in successful software.
I have found that out of approximately 500 users (talking face to face) you get 1-2 who actually DO know what they need, how it should look and act. We hired one and listened intently to the other.
It's a long search, but really, really worth it. I agree though - the coding is the simple part.