I'd rather waste a weekend finding 5 people to talk to than 3 months building the wrong thing
Everyone says "just go talk to your users" like it's the easy part. It's not. The hard part is finding people to talk to in the first place, especially before you have a product and you're basically asking strangers for a favor. Right now my "process" is cold…
Everyone says "just go talk to your users" like it's the easy part. It's not. The hard part is finding people to talk to in the first place, especially before you have a product and you're basically asking strangers for a favor. Right now my "process" is cold DMing people, low-key begging in communities like this one, and occasionally paying for one of those User Interviews style panels that feels way too expensive for something as early as validating an idea. So here's my slightly embarrassing confession: the thing I want to build is a way to fix exactly this. You book short calls with relevant target users and talk through your idea, your designs, whatever you've got, with a Mom Test style question framework baked in so you don't walk away with a pile of polite "yeah I'd use that" answers that mean nothing. But before I build a tool to help people talk to users, I figured I should, you know, actually talk to some users. So: how are you finding people to talk to right now? And what's the worst part, finding them, the endlessscheduling back and forth, or getting honest answers once you're finally on the call? Not pitching anything. Genuinely trying to not build the wrong thing.
Collected discussion
You don't know how to do something so you built an app to do that thing and now you want to know how others do it? This is the most 2026 post ever.
looking back if i had the option to talk to experienced people like product managers, software engineers early on , would have made much better products and avoided expensive mistakes , So i am just leveraging my network to make a platform so that others get clarity before building.
will you use the platform i am building? and how much are you spending now to validate your ideas ?
makes sense, how much time were you spending finding them? if given option to instantly connect to your icps for money, how much will you pay for per interview ?
esp meaning?
Finding them was always the worst part for me. I used to spend hours hunting Reddit and X threads where people were already describing the problem I was working on, then just responding in context. No ask, just being useful. A few of those turned into calls naturally because they already had the pain fresh in their mind.
You are thinking right! Validation exposes friction - Gets your problem - solution - gets structured! GTM, Technology and Users are the core stakeholders to validate. Good Going!
This is spot on! ESP today!!
The hardest part isn't scheduling — it's proving the person is actually in the target segment and is speaking from a recent problem. A paid panel can make the calendar look full while selecting for professional interview takers. Before building software, I'd run this manually for 10 sessions: get one founder to prepay for five interviews, recruit against a tight screen, require one recent artifact such as a redacted invoice, work order, or screenshot, and replace no-shows. Track recruitment hours, screen-fail rate, show rate, and whether each call changes a product decision. If nobody will prepay for that concierge version, faster scheduling probably isn't the bottleneck. If they will, you've learned the real product is verified recruiting and replacement — not the booking link.
are you dming people who just fit the profile, or ones who've already posted the exact complaint somewhere? asking because the second lot barely need talking into a call. one and two star reviews of whatever people use for this now are full of them, they'll write actual paragraphs when something annoyed them enough to leave a review in the first place.
I've found that the hardest part usually isn't finding people, it's finding people who are already experiencing the problem.
Honestly, the fact that you're pausing to ask this before building is already ahead of most people. But I'm curious about something upstream of the finding-people problem: where did this idea come from? Did it start with a problem you kept hearing people complain about, or more with a gap you noticed, and now you're checking if it lands? Because if it's the former, the "finding people to talk to" part usually gets a lot easier. You're not cold-DMing strangers for a favour, you're going back to people who genuinely want to help you find the solution to their problem.
My opinion: Because of this new reality we have in 2026, where the entrepreneurship landscape is hilariously saturated (Especially in anything AI can touch which is everything), if you are only in it for the money, it isn't a part of one of YOUR pain points or interests, don't bother.
Honestly finding them is the part that sucks, not the call itself. What's worked better than cold DMing for me: don't message people who just fit the profile, message people who already complained about this exact thing somewhere. Someone venting in a thread is a totally different conversation than someone I cold DM'd out of nowhere. They're already annoyed, they don't need convincing to talk. Scheduling back and forth is real but honestly I've just accepted it as a tax, not something worth solving. What actually worries me more is the polite "yeah I'd use that" answer. That happens a lot when someone agrees to a call because they already like you or your idea. I've gotten way more honest stuff from people mid-rant than people who said yes to be nice. One thing on your tool idea though, booking the call isn't really the hard part for most people, finding someone who'll actually be honest is. If it just makes booking faster but you're still talking to "nice" people, you'll get the same useless answers, just quicker. The Mom Test questions help once you're on the call, but they don't fix who's on the call in the first place.
the classic. problem first and then the product. i'd suggest setting up a paid waitlist or pre-buy option and market it. buying is the best way to validate your product
cold dms are hard because you're asking someone to care about your thing before they've said anything about theirs. easier version is to go find people who already complained. search the exact problem phrase on reddit, filter to the last year, and you get a list of people who wrote a paragraph about their pain unprompted. then the message is about their post, not your product. i've had a much better reply rate opening with 'you wrote this thing six months ago, is it still a problem for you' than with any polished intro. they already did the hard part by describing it.
Most people skip validation entirely, or they run a fake version of it—showing the idea to friends and collecting “yeah I’d use that” answers that feel like progress but tell you nothing. Talking to people before you write a line of code isn’t the embarrassing part. It’s the smartest thing in this whole thread.
those paid user panels just smile and nod for the gift card, it breaks your heart a little.
scheduling is whatever tbh, the polite answers thing is the real killer
The finding people part is the real wall, everyone skips past it. Cold DMs feel like begging but they still beat guessing.
it's that the people easiest to reach are rarely the ones who actually have the problem, so you end up with a pile of friendly-but-useless calls. Curious whether your tool solves for quality of match, because scheduling is annoying but solvable, while getting in front of genuinely relevant users who'll be honest is the part everyone quietly fails at
before building anything, i'd actually want to know if anyone from those calls brought it up again later on their own, without you pinging them first. i've watched a bunch of these threads where the 'yeah i'd use that' answers evaporate right after the call ends. did that happen with any of your five?
I’d separate “access to the segment” from “interview logistics.” If you can’t name where a narrow segment already gathers or what recent behavior proves the problem, faster booking won’t fix the research. Paid panels can also select for people who are good at being interviewed rather than people who just felt the pain.
Find the right audience first. I went to a university job fair and asked students a few basic questions - how many applications they usually submit before getting interviews or an offer, how long one application takes, and whether they would pay someone to apply on their behalf. More than 90% said yes. etc.etc. I asked the same questions in online forums and Facebook groups and got a similar response. I was mentor for student's association in college so that gave me some leads. Now I have enough confidence to build it right? and I already had a few customers lined up before launching. As of today we’re operating in five countries :P find the problem, find the solution and fix it.
I'm building a client portal style SaaS specifically for web dev agencies and so far I've been doing LinkedIn posts and cold connection requests followed by a cold dm if they accept. Conversion rates for this are not great but they're okay. The problem is a lot of people say they'll do the demo and then don't actually go and sign up for it. You also just have a lot of people who don't look at the connection requests and accept/decline your's because they either aren't online anymore or they just don't check that stuff. What's been nice about reaching out on LinkedIn is that I can easily find leads but getting through all the stages to get them into the demo has been rough for the amount of feedback I end up getting out of it.