LeadSynth will show you why “getting feedback” is broken
Every founder says:
“just talk to users”
“get feedback”
“do user interviews”
So you try.
You post
You ask for feedback
You DM people
And you get:
“looks cool”
“interesting idea”
“would use this”
But no one actually signs up
No one converts
Because that’s not real feedback
That’s polite feedback
The kind people give
when they don’t actually have the problem
Real feedback sounds like:
“this is so annoying to deal with”
“I’ve tried 3 tools and none work”
“is there ANY solution for this?”
It’s emotional
Specific
Urgent
And most founders never hear it
Because they’re asking the wrong people
at the wrong time
So they build based on signals that don’t convert
That’s the trap
That’s what we’re fixing with LeadSynth
Instead of asking for feedback
you tap into real conversations
Where people are already:
struggling
searching
trying to fix the problem
So you’re not guessing
you’re observing real demand
And building around that
If you’ve ever felt like
“people say it’s good but no one uses it”
This is probably why
This nails the core issue. The problem is not the medium, it is the social context. When someone sends you a Google Form or DMs you asking for feedback, there is an implied social contract to be kind. Nobody wants to tell a founder their idea is bad to their face, even a virtual one.
The conversations where the filter comes off are the ones where people are already frustrated before you entered the room. Those are the only ones worth treating as real signal. "Looks cool" tells you nothing. "I have tried 4 tools for this and they all fail at X" tells you everything.
The deeper thing I would push back on slightly: even real frustration can mislead if you do not check whether people are actually doing something about the problem. Complaining is easy. Switching tools, paying for something, asking for workarounds, those are the behaviors that confirm you are solving something real.
The reason most feedback loops fail is the same reason most peer review fails. The feedback giver has no skin in the game and no accountability for the quality of what they say. What changes the equation is structured feedback from people who actually ship and have something to lose by being wrong. Seen this play out in dev contexts too, we built Stackrate (stakerate.dev) as a peer review layer specifically for that reason. Different problem domain but same root failure mode.
the skin-in-the-game framing is the part most people miss. politeness pressure is basically zero-accountability feedback. and stackrate looks neat, same root problem from the build side. the cleanest version i keep landing on: the only feedback that ever predicted revenue for me came from people already spending time or money to solve the thing some other way
Agree that feedback is broken. People say "interesting idea" and then never buy.
What's your take on it? Is it that people don't know what they want until they see it? Or that asking for feedback attracts non-buyers?
bit of both but mostly the second one imo. asking for feedback self-selects for nice people who arent actually in pain. the ones with the real problem arent waiting around to give feedback - theyre already googling it, complaining about it, trying random tools. you catch way more signal watching that than asking "what do you think"
persistence alone is shipping to yourself. right on product, still invisible - the work has to show up where buyers are. that gap's the real thing.
"persistence alone is shipping to yourself" - stealing that line lol. building in a vacuum just gets you a really polished thing nobody asked for. the work only counts once it collides with where buyers already are
ha, feel free - I earned it the hard way. 'where buyers already are' is the bit I kept skipping
the hard way is the only way it sticks though. i skipped it for years too - figured if the product was just good enough, buyers would find me. they don't. you have to go stand where they already are, mid-frustration, and meet them there. obvious in hindsight, invisible while you're heads-down building. what finally made you stop skipping it?
this is actually something I started noticing while building too, most feedback sounds positive, but nothing actually changes behavior
I think the biggest signal is frustration, when someone is actively trying to solve it, not just reacting to an idea
out of curiosity, where are you mostly pulling these conversations from?
Reddit tbh and the tool I use is LeadSynth to do so.
that makes sense, I’ve been seeing the same thing on Reddit, when someone is actively struggling with something, the language is completely different compared to when you just ask for feedback. it’s less “this looks interesting” and more “nothing I’ve tried actually works”
feels like that gap is where most ideas either succeed or fail
exactly — and the part i underrated for too long: that language isn't just the targeting signal, it's the copy. the words people use to describe the struggle ARE the headline that converts them. "nothing i've tried works" beats any clever tagline i'd write. so i mostly stopped writing messaging and started quoting it back. that gap you're describing is the whole business imo.
You’re right that most “feedback” is just social politeness, but the deeper issue isn’t where it comes from—it’s how it’s validated; even emotionally charged complaints can mislead if they don’t translate into behavior like switching tools, paying, or repeated usage, so the real signal isn’t just “people are frustrated” but “people are actively trying to solve this and making trade-offs,” which is where observation (like what you’re building with LeadSynth) becomes powerful—if it helps founders move from opinions to actual intent signals, that’s a meaningful shift. The real question is: how does LeadSynth distinguish between loud frustration and genuine willingness to adopt or pay for a solution?
We built an intent analyzer algorithm which helps analyze convos and social media traffic so founders get the best intent signals.
The polite feedback trap is real and I think it goes even deeper than asking the wrong people. Most feedback mechanisms are designed in a way that makes honesty socially awkward. A survey sent by the founder to their own users has a built-in politeness pressure. Nobody wants to tell you your idea is bad when you’re clearly excited about it. The conversations where people are already frustrated before you’ve entered the room are the only ones where the filter comes off. What does LeadSynth actually surface, Reddit threads, reviews, support tickets?
So it surfaces reddit posts/comments; X posts/comments, LinkedIn posts, Linkedin Profiles, Youtube comments.
That’s a solid surface set the places where people talk about problems before they’ve decided to look for solutions. LinkedIn profiles is the interesting one, most tools skip that entirely. Are you pulling from public posts only or does it also pick up comments buried in threads that would never surface in a normal search?
both, not just top-level posts. honestly the buried stuff is usually where the gold is - a "tried 3 tools, all break at X" comment 40 replies deep tells you way more than the original post, and those never surface in normal search which is exactly why most tools skip them. comment chains + linkedin profiles were the two we specifically didnt want to leave on the table
This comment was deleted 2 months ago
Exactly, what are you building btw?
This comment was deleted 2 months ago
that translation layer is real - clear signal still dies if no one acts on it. feels adjacent not competitive honestly, we're upstream catching the signal, youre downstream turning it into a call. curious how youre handling the "same signal read differently by different people" problem, thats the messy part