How I'd get my first 10 customers this week with zero audience
Most "first customers" advice tells you to build an audience, post every day, and DM cold lists until something sticks.
If I woke up tomorrow with zero customers and zero audience, I wouldn't do any of that.
Here's the week I'd run instead. It's free, and you can start today.
Day 1 — write the problem in their words, not yours.
List the 8-10 sentences a customer actually types when they have the problem you solve. Not "needs lead generation." More like "spent my whole morning dming people on reddit and got nothing." Those sentences are your search queries. Your feature names are useless here — nobody types them into a search bar.
Day 2-7 — go where they complain out loud.
Paste those sentences into search on Reddit, X, LinkedIn, and YouTube comments. Sort by newest. You're not hunting for an audience. You're hunting for the 30 minutes right after someone admits the problem in public.
Tier what you find:
— "how do you all handle X?" → warm
— "X is killing me, I've tried everything" → on fire, go now
— just venting, no intent to fix it → skip
Then reply like a human. One line. Quote the exact thing they said, offer to actually help, and don't pitch:
"saw your post about doing this by hand — that's the exact thing that drove me nuts too. want me to show you how i fixed it?"
Do 10 of these a day. By hand. It feels too slow and too small — that's the point. Your first 10 customers come from 10 real conversations, not 1,000 sends.
I know it works because it's the whole engine behind LeadSynth: 27,178 conversations, 20.6% reply rate on Reddit vs ~1% for cold email, 0 spam complaints. But you don't need my tool to run that week. The signal is free and public right now.
When doing it by hand across four platforms got to be too much, I automated it — first leads are free, no card: https://www.leadsynthai.app
Real question for the room: what's the single search string that's pulled you the most customers? Genuinely curious which sentences work best in other niches.
This makes sense, but the real bottleneck for me is not the search logic. It’s the stamina to read Reddit every day without losing my mind.
I can read Pascal or technical docs daily, but public complaint threads are very cognitively different . The method works only if you can stay human and specific while filtering through a lot of low-signal noise.
I am curious how you personally manage that part. Do you time-box it, use strict search strings, or just accept that the first-10-customer grind is supposed to feel a bit unpleasant?
yeah, the search logic is the easy part — staying human on your 40th complaint thread is the actual tax. nobody warns you about that one.
two things keep me sane:
ruthless filtering. strict strings, sort newest, tiny windows. "just venting, no intent to fix" gets skipped on sight, and that's most of the noise gone. i'd rather scan 15 on-fire threads than wade through 200 maybes.
hard time-boxes. 25 min, one platform, one problem-phrase, then stop even mid-scroll. reading past that is exactly when you start sounding like a script — because you're fried.
honest answer to your last part: no, it shouldn't feel unpleasant. if it does, you're reading too much low-signal stuff. i eventually stopped doing the reading by hand at all (literally why i built leadsynth — intent-matching surfaces the high-signal threads, i keep the replying 100% manual). turns "2 hrs across 4 platforms" into "10 people just raised their hand, go help them."
This is helpful, especially the distinction between reading Reddit and filtering for “raised hands.” I think I was treating the noise as part of the work, when really the filter is the work. I’m going to try the 25-minute / one-platform / one-phrase version and treat skipping as part of the method. Thanks.
"the filter is the work" — you said it cleaner than i did, i'm stealing that. one thing for the first week: track your skip count, not just your replies. skipping most of what you read feels like failure but it's actually the method working. which platform are you starting with? curious how the 25-min box holds up outside reddit.
Good point. I don’t know yet. But Reddit is probably the easiest place to test because the complaint language is raw and searchable.
My guess is the 25-minute box holds up differently by platform: Reddit for raw pain, Indie Hackers for founder-language calibration, maybe LinkedIn/X for more legible but more performative posts.
For the first week I’ll probably start with Reddit, but track skips/replies separately so I can compare. The skip count idea is useful because otherwise “I found nothing” and “I correctly filtered 40 bad fits” feel the same.
reddit first is the right call, and your platform read matches what i see in my own data. reddit is the only place people write the problem the way they'd say it out loud — the linkedin/x version of the same pain is dressed up for an audience, so replies land colder. tracking skips separately is the smart part: that number is what keeps "i found nothing" honest. come back and post your skip-to-reply ratio after week 1? would love to compare notes against my reddit numbers.
Good point. That’s probably the right way to keep the experiment honest. I’ll do a small measured pass and track: scanned / skipped / replied / useful replies.
My hunch so far is that Reddit is very search-phrase-sensitive. When the phrase is tight, it can surface real pain. When it’s too broad, it’s easy to spend time reading "interesting but not actionable" threads and mistake that for research. So I mostly want to see whether my phrases are specific enough before judging Reddit as a channel.
the phrase-sensitivity read is dead on. one trick that saves a lot of trial and error: steal the phrasing from the first raw complaint you find — the exact words they use become your next search string. pain language is weirdly self-similar, people describe the same problem with the same three or four phrases. and "interesting but not actionable" is exactly the trap — my test is: if i can't picture myself replying to the thread, it doesn't count as research. good luck with the measured pass, still want that skip-to-reply ratio next week.
The "zero audience" week usually isn't about posting more — it's about showing up in 5–10 conversations where the pain is already live. Reddit is underrated for that if your buyers hang out there: search by problem language ("how do I get first customers", "alternative to X"), not by your product category.
The tedious part is doing that hunt every day without sounding spammy. What's the product/context for your 10-customer plan?
exactly — the hunt is easy to describe and hard to sustain. the "without sounding spammy" part is the whole skill: you have to actually read the thread and answer the specific thing, not paste a template.
context: it's LeadSynth (the tool behind this post). it watches those pain-language searches across reddit/x/linkedin/youtube and surfaces the on-fire ones, so you're not pulling the list by hand every morning. the human still writes the actual reply — that part shouldn't scale.
reddit's been our highest-signal channel by far. where do your buyers tend to complain out loud?
Fair — automating the hunt is the hard part. Good luck with LeadSynth.
appreciate it. yeah automating the hunt was the whole reason i built it — the by-hand version just doesn't survive past week one for most people. what are you working on? always happy to trade notes on the first-10 grind
I did 10 cold reachouts today, got valuable feedback from one of them. The trick is to find the place for your app where the target customer actually is. For me that was instagram.
It does feel slow and tedious, but I'll keep doing it and report back.
this is exactly it — and the fact that 1 of 10 gave you real feedback on day one is the signal. that's a 10% hit rate on genuine conversations, way better than it feels in the moment. instagram's a great example too, most people would never guess it for their niche until they actually go look.
the tedious part is the moat though — almost nobody keeps it up past week one. definitely report back, curious how the instagram angle plays out for you.