Ask HN: Where do you find potential customers to validate your idea / MVP?

Isn't an MVP, by definition, not just for you?

The specifics of what you're product does will partly guide your approach, and I couldn't get enough information from the context to be sure that my advice will be appropriate, but I'll share what experiences I do have.

First -- what are you looking for with regard to feedback? Are you looking for someone to tell you that your product solves the same problem for them, too? Are you looking to better understand the problem (and related problems) so that you know what to focus on after the Minimum part is over? Are you looking for someone who specifically has this problem to determine if there's any interest in even solving it (and/or to what that person might be willing to spend/do/put up with to have that problem solved)?

These questions usually serve product design and the customers who are willing to take the time to give you these answers are -- somewhat -- designing your product with you. If you want comprehensive feedback, you'll need to incentivize[0] it. On the very high-end, there are companies who will put together focus groups for you; but they have to cost a fortune[1].

My approach is to pick up a phone (figuratively). Anything that I've written with a target of selling, personally, has generally focused on solving a problem for a business that I frequent. I made an observation a while back that my dentist's scheduling and office management system was pretty terrible[2]; this being in the early days of the "maturity of text messaging" and my having built a self-service Kiosk application, I had some ideas and a strong working relationship due to weak enamel and a propensity to consume sugary beverages. I called him up and was surprised at how willing he was to talk my ear off about this problem. It was so encouraging, that I called a six other dental offices, left messages for dentists, and ended up receiving good feedback from about 10 individuals.

I have the original script that I typed up with my list of questions, because I wasn't great "off the cuff" and really wanted to avoid sounding like a sales-person or telemarketer. "I'm a software developer in (city). While visiting my dentist, I had some ideas for writing a software application for dental office management. I'm not a dentist, so I'm seeking help from local dentists to understand if there's a way to write something that can reduce the amount of time dentists spend away from patients. If you have a few minutes and can help a guy out, I'd really appreciate it. Thanks!"

I called and left that message with the receptionist being as ridiculously polite as I could; I heard ringing in the background, once, and said "I'm not a patient, so take that one and get back to me when you've had a moment to catch your breath" and sat on hold for 10 minutes... I called back a few days later about an hour before close and discovered (1) small dentists often employ husband/wife/mom/dad as office manager/reception part-time (2) oh...wow, one mom is really helpful and doesn't get any crap for having a 2.5 hour conversation in the open lobby while processing customer payments, or for interrupting her daughter during a procedure to ask a question (I had no idea she would do this).

For anyone wondering, no, I never wrote it. Shortly after researching, I had a visit to a specialist (sleep, not dental) who had a system in place that basically did everything I was looking to do, which took the wind out of my sails long enough for me to see my own dentist install a fingerprint reader/incredible setup over here shortly after the last failure of the server he had in his office. I was pleased to learn that he picked the setup he did because "it included checkin/checkout via touch screen kiosk" and "had a fingerprint reader"[3] based on the conversation I had with him. And it solved his problems and made the experience for me as his patient, better, as well. After using it, I kind of wish I had written it. :)

[0] You can just outright pay someone, but I'm assuming that's not desired.

[1] And a million who will send spam to you offering you money for your opinion. Anyone with an inbox is familiar with all of the scammy "Get Paid for your Opinion", but there are real companies out there -- I've routinely participated in focus groups with Shiffrin-Hayworth and been paid an average $150/group (catered); their web site, Lord, looks fresh out of 1999, no SSL -- if my Mom hadn't referred me I'd have assumed they were a scam.

[2] Especially when I visited one day while his tech guy was changing tapes in the tape backup due to an over-night server failure.

[3] I mentioned "fingerprint reader" by accident when I talked with him about it early on but even before discussing with him, I had decided against it. I figured there's no way that thing isn't going to be sanitized endlessly by random staff without regard for how the cleaner could affect the reader. I had many-a windex bottles emptied out on my keyboard (thank God for that Northgate Omnikey) from a well meaning mom trying to rub the dirt marks off of the spacebar/enter/arrow keys.

hmm, good point. I didn't look at it from that POV.

When I'm trying to scratch an itch, I try to get something 'rough' and then try to add the bells and whistles later on. Mentally, I always think of that 'rough' bit as 'my own MVP' i.e. what are the minimum features I need to build for this tool to solve my problem.

For example I needed a self hosted solution to convert svg images to png. I need a way to specify the image to be converted, code to convert it, and a place to store the output. My first build had a text field where I manually enter the path to the svg file I need to convert, and I hardcoded the output folder. I thought of that as my MVP. When I was sure the code was working and much later on when I had the time, I added a file picker to allow me pick the input file and another file picker for me to specify the output location.

The "product" part of MVP refers to something you sell (rather, offer) in a market, and the "minimum viable" part refers to it being barebones but functional enough to be "sellable" and provide value, ie someone that isn't you is willing to go through the stated process to install and use it. So writing a tool that you use yourself without "customers" (used loosely) isn't in my mind an MVP, and neither is a product with no thought given to distribution.

Events are another excellent place to find targets to interview. For example, if you work in security, go to RSA and just bug people in booths or wherever.