How do you validate a problem?
Hey! I'm working on this new idea that is basically a problem coworkers had at work and I've been trying to validate if it is specific to my coworkers or is it a general problem that could become a real business. What do you guys do to validate your ideas?
I would put together a landing page and try to direct to it your target audience. Have some sort of lead capture built in i.e. ask for emails. See how many people put their email down.
Nice idea. I'll add this to my To-Do list
Paid: See if your target user(s) will pay to use it (by actually requesting payment).
Free: See if your target user(s) incorporate(s) your product into their life, using it regularly.
Ok. This kind of requires me to have something as of now, right? But, how do you find evidence that the problem I'd like to work on is actually a problem and not some idea that can't become a business? I mean, I'd like to avoid to build an MVP only to find that it is not a real problem. Do you have any idea on that?
Instead of an MVP, you might want to come up with a MEVO, a Minimum Economical Viable Offer, which is basically what you’re pre-selling to the prospective customer. A MEVO pieces enough information together to help your audience gain interest in what you plan to create. Here’s Dan Martell's podcast going through MEVOs in detail. (https://thefoundation.com/podcast/episode156)
Most of the startups fail because they allocate resources to a project destined to fail. But you can pre-sell your products before you even build it to validate your idea. That's where pre-sale launches steps in. If you raise enough attention during the pre-sale launch, then you're good to go. If not, then you might want to reconsider your product idea or your marketing strategies.
☞ https://blog.keyzy.io/5-steps-to-a-software-pre-sale/
Nice idea. I'll add this to my To-Do list
What’s the problem they are having? I would suggest searching for the problem online or in forums and see if there are others out there having issues maybe? That’s a tough one especially if it’s pretty niche. But, if it is niche you have a nice target audience.
Well, I don't know if I should actually build an MVP for this problem I'm trying to solve. The idea is to build a PaaS for machine learning, helping people train and deploy their models. Kind of like Compose.io, but for Machine Learning. I am still unsure if deploying and training models is a problem people have.
Is it a problem that you have? Would you pay for a service like what you are building?