Took the "go find where the question is already being asked" advice literally. Here's what happened.
A few people on my last post said versions of the same thing: stop trying to get people to find my post, go find the thread where they're already asking my exact question, and answer it well enough that the link is the obvious next step.
So I did it literally. Signed up for Quora for the first time, found a years-old thread with close to 100 answers asking for a seed-stage investor update template, and left a genuine answer - pick one north star metric, keep runway as a live number instead of something you eyeball quarterly, mentioned the model I built at the very end instead of leading with it. Then found an "is Etsy profitable" thread with 120+ answers and did the same for the tax tracker: answered honestly (yes, but "profitable" and "what you think you're making" are usually two different numbers), tied it to the fee/materials/shipping math most sellers never actually run, mentioned the tool once.
Two things I didn't expect from doing it this way instead of posting cold:
Finding the right thread takes longer than I thought. A lot of old threads look active because of the total view count, but you have to actually check whether anyone's still answering, or you're just shouting into an archive nobody opens anymore.
Answering someone else's real question is a different skill than writing a post. You can't lead with your own framing - you have to answer their framing first and earn the one sentence where you mention your own thing.
No results yet, both answers are hours old. Will report back either way, including if it's nothing.
Anyone else tried the "go answer, don't post" approach somewhere outside their home platform? Curious what worked and what was a waste of time.
I’ve seen this work better than posting into a general audience too. The hard part is finding conversations where there’s real intent and the thread is still active.
The “answer first, mention the product second” approach also feels much less promotional and usually leads to better conversations.
Update since a few of you asked specifics: @hayeslive the investor-update thread did edge out the Etsy one, but not because the audience was narrower - it was just fresher (days old vs. months). Matches what @HoldCashSheets said about the archive trap being the real killer, not thread size. @credolopes haven't split Reddit vs IH yet, that's the next thing to actually log instead of eyeballing - gut read is IH questions have been higher-intent, Reddit's been more commentary/removals. @aryan_sinh honestly a qualified visit would mean more to me right now than a reply count - still $0 in sales either way, staying honest about that. @atlanicrecycling hardest part is telling "has the exact problem" apart from "browsing the category" before you answer, not after.
The part about checking if a thread is still alive is underrated. I've been fooled by view counts before, a 5 year old thread with 100 answers is basically a graveyard
I've had stuff removed on Reddit for sounding too promo so I'm trying the same idea, answer first and only mention what I'm building if it fits.
Your answer-first distinction is exactly right: the reply should still be useful if the product sentence is removed. One lightweight way to test the channel is to log thread age/activity, qualified profile visits, and any follow-up question—not just clicks—because a high-view archive can be a dead end. I’m doing a small maker check with MP5 for non-eng PMs/ops/consultants/founders: would you spend ~20 minutes as a guest and give a yes/maybe/no on ~$5/mo? https://mp5-production.up.railway.app/ (Tip: ignore the Connect agents.)
One thing that's helped me with this for a small PDF product: before answering, I write down the asker's exact words for the problem and reuse them in the first line of my reply. That makes me answer their framing instead of mine, and it also tells me something about the product. If I keep having to translate their words into my listing's words, the listing is probably describing the wrong thing.
Second, sort the threads you find into two piles: "still deciding if this is a problem" and "already decided, stuck on how." The second pile is smaller, but that's where a specific answer (your runway-as-a-live-number point is a good example) actually gets used. For the first pile I leave the product out entirely and just answer. It keeps the account from reading as a distribution channel.
Curious whether the Etsy thread or the investor-update thread brings the more qualified reply. My guess would be the investor one, because the audience is narrower.
Just some things to keep in mind: concrete personal details are everything. My first comment was too polished and generic and Reddit literally removed it for looking AI-generated. Rewrote it with information from my own experience and it went through fine. And yeah, finding a live thread worth answering takes way longer than you'd expect. Most of them are dead
A few Quora-specific things worth trying: the "Answer" tab lets you filter to newer questions with only a handful of answers, which is the opposite of the 100-answer threads and much easier to rank near the top of. Following the question also pings you when new answers or comments come in, so you can reply while the asker is still around. And put a unique ?ref= tag on the link in each answer, so in a few weeks you can see which specific thread sent visits instead of lumping it all under "Quora". Last thing: the effort you put into those two answers is reusable. Turning each one into a short FAQ-style page on your own site means the same work keeps paying off in search even if the Quora thread goes nowhere.
The two thread types you tested differ in two factors that move in opposite directions, which is why the same effort produces opposite results and why neither will look like the winner on its own.
Think of the reach of an answer as demand multiplied by your share of the page. A years-old Quora thread with a hundred answers has high demand and a share close to zero, because you are the hundred and first answer on a page whose ordering settled long ago. A thread posted yesterday has much lower demand and a share that is a large fraction of it, because you are one of three. High demand with no share and low demand with real share are different products, and a single "answers are worth it or not" conclusion averages them into a number that describes neither. Your two Quora answers sitting at zero for a few hours is the correct reading of that model, not a failure of the approach.
That reframes the diagnostic you need. Total answer count is a proxy for accumulated demand, which is the half you already know how to spot. What you cannot see from the count is whether the slot is still open, and that is a velocity question rather than a stock question: new answers per month, not answers total. A hundred answers accumulated over six years means the page is settled and your marginal answer will not surface. A hundred answers accumulated in two weeks means the question is live but saturated. The profile worth hunting is high search rank with low answer velocity, and that combination is invisible in every thread list I have seen.
If you log those two numbers per thread instead of the total, the two answers you just posted become a real experiment rather than two anecdotes.
Worth flagging that I have a horse in this race: Piramyd is what I work on, and thread-selection problems are its home turf.
Which of the two threads ranks in search for a phrase people actually type, and how many new answers has each taken in the last thirty days?
We've been doing exactly this for five weeks, so here's our data, including the unflattering part.
415 answers across Reddit, IH and Product Hunt. 31 real replies. Search traffic is still about four visits a month, so as a direct traffic channel it hasn't moved anything yet.
What the replies taught us: freshness beats size. About 45 comments on big, older case-study posts got zero replies. Posts under a day old with few comments are where nearly all 31 came from, because the author is still reading. On Reddit the subreddit mattered more than anything we wrote. Same style, very different treatment depending on the sub, and all 5 removals came from one sub.
Your old Quora threads with 100+ answers are the opposite profile, so I'd expect slow search traffic there rather than conversation.
Will you track the two answers separately, so you can see which thread type pays?
The insight about answering being a fundamentally different skill than broadcasting a new post is spot on. When you're answering, the measure of success is whether someone feels relieved having read it, not how clever your hook was. One thing that helped me with external forums is checking if the thread ranks in Google for relevant search terms—if people are finding the thread through search every week, even an older thread can deliver steady, high-intent traffic.
The distinction between answering the person’s framing and writing your own post is the key lesson here. I’d keep a small log of the original question, the answer’s first useful step, and the follow-up signal, then compare results across a few fresh threads instead of judging by views alone. If the reply remains useful after removing the product mention, it is probably earning attention rather than borrowing it.
The archive trap others mentioned is the real gotcha — I've started sorting threads by recent activity and checking whether the newest answers are actually recent before investing time. Your point about answering being a different skill than posting rings true too; answering forces you to actually resolve the question first instead of warming up the reader. Curious to see your follow-up: defining a win upfront (qualified replies vs. just views) would make the experiment way more useful either way.
we ran almost exactly this on reddit for the last week for our ios beta, answering in threads where people were already asking for testers or for cooking help and mentioning the app at the end. it came out to about 180 comments with the testflight link in roughly 100 of them, and we can't trace a single install to any of it. the first three days actually got around 30 real replies, then a burst of posts from the same account seems to have tripped reddit's spam filter, and everything after that got zero replies and zero votes, even in busy threads where other people's comments were getting upvoted. so the thing i'd add to your archive point is to check every few days that your answers are still visible from a logged out window, because the platform won't tell you. the other thing we learned is that the threads with real intent were mostly other founders asking for testers, and they want a swap more than they want the product. really curious whether quora is any different there, since the etsy thread sounds like actual sellers
415 answers into 31 replies is the number to sit with, because the usual explanation for it, that the threads were stale, only covers part of the gap and hides a cheaper failure that is easier to check.
A reply needs three things to be true at once, and your results separate them. The reader has to still be present, which is what freshness buys you. Your answer has to occupy a visible slot, which is the page share problem. And the reader has to be at the point of the question rather than still shopping for it. That third one is where comment counts mislead at the other end of the range: a thread under a day old with three comments is fresh and you own the page, but if all three comments are other people also describing their problem, nobody in it is looking for a solution yet. You are answering early, which means you are reading and not converting, and that is invisible in your data because it produces exactly the same signal as a stale thread: no reply.
That gives you a cleaner test than freshness alone. Count the threads where the original poster has already made a decision and is asking for help executing it, versus the threads where they are still comparing options. An "is Etsy profitable" question is the second kind. A seller asking why their 2026 numbers do not match what the dashboard calls profit is the first. My read is that the second kind will carry most of your 31 replies and almost none of your 415, because intent that is already narrowed is the only intent that a one-paragraph answer can actually resolve.
The Reddit removal data is the strongest thing in your comment, and I would treat it as a stronger version of the visibility check than a logged-out window. Same style, same author, five removals all from one sub, tells you the constraint was the community's own moderation norms rather than the volume. So the per-channel experiment you are about to run is really about which subs tolerate a founder answering, not which platforms convert. One subreddit that allows it beats five that do not, and you probably already know which one it was from where the removals landed.
Worth saying plainly since my last comment here was on the thread-selection side: Piramyd is what I build, so I have an interest in this problem being solved with instrumentation rather than instinct.
When you split the 31 by channel, do the Reddit replies hold up as well as the Indie Hackers ones, or is Reddit all commentary and no conversation?
The archive trap is the one that burned me too. A thread with a hundred answers can still be dead if the last reply is months old, so I only spend time where someone answered in the last week or two. Before calling the channel a win or a waste, I’d watch what happens after the reply: a profile visit or a save beats the upvote count on the answer. For the investor-update and Etsy tax threads, write the answer so it still helps if the one product sentence gets cut. If the math stands alone, the thread was worth it even when the mention converts to nothing.
When the answers have time to circulate, what outcome would convince you this channel works—qualified visits, product trials, or actual sales rather than just replies?
Really solid approach — curious how you're thinking about this, what's been the hardest part to figure out so far?