What to Do With a New Startup Idea

Take the risk out of entrepreneurship

I talked last time about what not to do when you have a startup idea. If you haven’t read that post, I recommend you start there because it’s necessary to understand all of those common traps before you can really get started.

I have a great idea, what now?

Do you? I’m not going to lie, I think pretty much all of my ideas are great too. I need other people to come along and tell me I’m full of it and so do you. These first steps are all going to revolve around getting out and talking to people to figure out if your idea is solid. If it isn’t, you could end up wasting a huge amount of time.

You’re not going to get particularly excited about this advice. The first thing anyone wants to do when they have a business idea is start hustling/coding, but those aren’t the best uses of your time. Tons of people waste money and time on something that no one wants, and can ruin their lives in the process.

Watching this video is painful, but I recommend it to truly understand what I’m talking about:


This is what happens when you don’t validate your idea first.

Validating Your Idea

Going out and talking about your idea with people is tough. No one likes being told their baby is ugly, but it’s better to find it out now than a year down the line. Validating means you can quickly go through a few ideas to find the best one, and then start building without worrying about the market existing when you’re doing.

Validate the Problem

You most likely thought of an idea and not a problem. That’s fine, but now that you have that idea you need to figure out what problem it’s solving. If you think about it and realize that it’s not really solving a problem, then it’s probably not going to work out.

Think about hover cars. Very cool idea, but they aren’t really solving a problem. We would still need some sort of road system, and unless they were flying they wouldn’t reduce traffic.

Now, there are of course very successful startups that aren’t really solving a problem. Pinterest is my favorite example of this—it’s a pure consumption site that no one needs, but that has an insane valuation. If you can create another Pinterest success power to you, but don’t assume it.

To validate that there’s a real problem, all you need to do is talk to people. Find people in your demographic, and ask them if they’ve ever felt like this is a problem. If they respond “eh, kinda” then you might want to keep looking. But if they respond “YES, ugh, every day” then you’re probably on to something.

If people respond “YES, ugh, that’s a problem every day, I’ll pay you right now to fix it” then you’ve found a promising idea.

Validate the Solution

Now you need to prove that the way you’re addressing the problem is the best way. This will likely change as you develop the product—you might find that certain features are unnecessary and others that you never thought of are crucial. But you need to make sure that the general idea is a good solution first.

Think about it like this: you’ve identified that monitoring how long to steep tea for is a problem among tea drinkers. They hate their tea getting bitter, but also want it to be full-bodied. Your solutions could be:

  1. A special mug that pulls out the tea when it’s done
  2. A phone app that has different timers for different teas
  3. A mug with the steeping times printed on the sides
  4. Something you can attach to the tea-bag that gives a visual indication when it’s ready

Maybe you thought of number 2 first, but people don’t like the idea of having to use their phone for it that much. Instead of scrapping the whole thing, try other solutions to the same problem and see if they get more interest. Tea drinkers might kill for a mug that automatically removes the tea bag when it’s ready (I know I would).

You’ll know you have a winning solution when people will:

  1. Pay you for it now
  2. Sign up for info on it
  3. Agree to tell their friends about it

But Remember This…

It’s really easy to get a false positive in your interviews. If you’re not careful, people will straight up lie to you because they think that’s what you want to hear. There are ways you can prevent this though:

  1. Don’t talk to friends—unless they’re close enough to tell you your idea is dumb. I have a few very close friends who I’d trust to be brutally honest, but aside from them I try not to interview people who know me.
  2. Don’t tell people it’s your idea. When I do user interviews I tell them I’m taking a class on market research and helping a local startup, then insult the product to show them I’m fine with them being mean about it. If they knew it was my own product they’d be uncomfortable about telling me it’s bad
  3. Don’t lead the witness. You have to be extremely careful in how you word your questions. If you ask someone “how great do you think this idea is?” You’re doing it wrong. But it can be also be as subtle as “Was this confusing?” Because then people will go “Hm, was it? I guess I was confused at this one point… yeah it was confusing” You need to word everything completely neutrally, which is easier in a survey than in person. For example: “Please select where you fall on this scale: (where 1 is “this was confusing” and 5 is “I was not confused”)”

Build Hustle Build

Once you have proved that there is some weight to the problem and solution, now you can start building.

But the policy of validating things as quickly as possible still holds. Don’t lock yourself away for a year while you hack together the perfect solution—you’ll hit the same problem as if you never validated your idea in the first place.


Instead, you need to get to the next stage of feedback as soon as possible. This means building the most ghetto, basic version of the solution that you pitched to people before. In Lean Startup terms this is the “Minimum Viable Product” or MVP, and different people have different definitions but the one I like best is:

“[The MVP is] the most basic version of your product that you can give to someone, and then walk away, and have them not be totally confused as to how to use it.”

This doesn’t mean create something ugly and hacked together, but don’t kill yourself on getting to a perfect state either. This quote has been overused but just in case you haven’t seen it, it’s a good one to remember:


So get to that state where you can give it to people and see how they use it as quickly as possible. If you ever find yourself thinking “well, I could give this to people, but I’d rather wait until we’d done A, B, and C…” then just give it to them now. They might tell you that you don’t need B and save you a lot of time.

Keep Validating

Thought you were done with interviews after the initial validation stage? Nope, it never ends. Every iteration on the MVP merits another round of interviews and user feedback, so keep building a list of people willing to try out the product and finding places where you can go bother new ones.

You can even do this before your product is in a state to leave people with it. Take the early version that requires your help out and show it to people, see what they like, dislike, and would want to see added. You might discover things you never thought of and that can help you get to an MVP state much sooner.

Make sure before you build in some additional feature that it’s something people really want in your product. A lot of great products are that way because of their simplicity and clarity. You don’t want to turn into a Microsoft Office product with a plethora of features that never get used.

You’re On Your Own Now

If you get to this point, you’re building something that people are enjoying and on your way to a great product. You’ll need to bring some other crazy people in to help, and maybe some money, but all of that is secondary to having this solid foundation for getting started. You’ll be head and shoulders above most people trying to startup these days.

Title Photo Credit: flickr