Where to Find Your First Customers Online (Every Channel, Judged Honestly)

You have eleven tabs open. One says Reddit. One says LinkedIn. One says communities. Every one of them is a list of twelve places to find customers, and every list is a different twelve.

None of them can tell you where your customers are, because none of them know who your customers are.

There is no best platform. There is only the place your specific person is already active and half expecting to be spoken to. Which means the channel question has no answer until the person question does.

What can be judged is difficulty, and what actually works when you have nobody. That is what this page is: every channel the advice posts list, read honestly from zero customers and zero audience, and the one move worth making tomorrow morning.

Quick answer

You find your first customers wherever the one person you built this for is already active, not on whichever platform a list ranks highest. Open LinkedIn search, Reddit, Instagram or plain Google, find the single individual who best exemplifies your ideal customer, and get into a conversation. In an afternoon you can hit the connection limits on most platforms. Communities are the most powerful of the channels and the slowest to pay off, referrals need a first customer before they work at all, and email is the hardest place to start. MyPip does the finding overnight, so the five minutes you have goes on the conversation instead of the search.

Where do you actually find your first customers online?

Wherever the person you built the product for spends their time. The platform is downstream of the persona, so the useful question is not "where would I start", it is "where is the most effective place to start", and only your customer answers that. If you are building for art designers in Tokyo, the answer is wherever art designers in Tokyo already talk to each other, and no ranked list of channels can know that.

So here is the literal first move, and it is smaller than you think. Open Google search, LinkedIn search, Instagram search, Reddit, wherever your people are. Find the one individual who best exemplifies the type of person who would love your product. Not a segment. One human being with a name. Connect with them, like the thing they just posted, get into a conversation. Then do it again.

In an hour or four you could do hundreds of these. You would probably hit your connection limit on every platform you tried. And each real connection spirals your network toward more of the right people, which is the part that compounds: the second person is easier to find than the first, because the first one is standing next to them.

This is why answering the person question first is not a warm-up exercise. Narrow the description until it feels uncomfortably specific. Not SaaS founders. SaaS founders building healthcare products, doing between zero and a million in revenue, based in lower Montreal. That level of detail is what turns "where do I find customers" into a search you can actually run.

Does marketing on Reddit work when nobody knows you?

Yes, on one condition: you show up as a member of the thing, not as an ad inside it. Reddit is the clearest example of a channel that punishes the sales motion and rewards the human one. The threads where your customers describe their problem in their own words are the highest-quality reading material you will find anywhere, and the fastest way to lose access to them is to post a launch link.

The rule that makes Reddit work is the same rule that makes communities work anywhere: be there to help answer a question, not to move product. Answer things for weeks before you mention what you built. When you do mention it, mention it in a thread where it is genuinely the answer to what was asked. Nobody minds a founder replying "I built the thing you are describing, here is how it works" to someone who described it first.

The other use of Reddit has nothing to do with posting. It is a research channel. People say things in a subreddit they would never say on LinkedIn, and they say it with the exact wording they would type into a search bar. Read fifty threads in your space and you will have your customer's language, their current workaround, and a list of names worth talking to. Half of the messages that get replies start life in someone's honest complaint, which is what to actually say when you get there.

How do you find customers on LinkedIn?

By using it as a search engine rather than a broadcast tower. LinkedIn's search is the most precise free way to build a list of real humans who match a specific description: title, company size, industry, region, what they posted about last week. You are looking for the exemplar person, then the people who look like them. Connect, engage with something they published, start a normal conversation.

The by-product is the reason to start here rather than in an inbox. Every connection you make and every post you comment on builds your own profile on the platform at the same time. People start to see your name before you ever write to them, and a message from a face they half recognise is not really a cold message. You get two things for one action, which is rare.

Two warnings. The platform caps how many invitations you can send in a week, and that cap is a feature, not an obstacle: it forces you toward the small number of right people instead of everyone with a matching job title. And the fact that someone matches your description does not mean they are ready to buy anything this month. Most of them are not, which is why only a small slice of them can buy this month and why you want the ones showing a signal right now.

Should you start with cold email?

No, unless you already know how to run it. Email is the hardest place to start, and it is the one channel that costs you money before it gives you a single reply. People arrive at their inbox already expecting spam. Behind the scenes you inherit deliverability problems, mailbox maintenance, sequence tooling, bought data and the job of keeping all of it in sync. That is a project, not a first move.

The numbers say the same thing. Across 7,530,489 cold emails sent in 2025, Belkins measured an average reply rate of 0.45 percent. Done properly email is the most lucrative channel there is, because when the formula works it works at a scale nothing else matches. It is just a terrible place to learn the formula.

Social messages are more forgiving, cost nothing to set up, and hand you that profile-building by-product for free. Start where a mistake costs you one ignored message instead of a burned domain. Come back to email when you have a message you already know gets replies.

What about your network, communities, Product Hunt and referrals?

They are the four tactics every article lists, and they are wildly different in how well they work at zero. Here is the honest read on each.

Your network, if it fits

Network is powerful and completely conditional. If you are building a tool for art designers and your contacts are doctors, you do not have a network for this product; you have a network for a different one. There is no partial credit for knowing lots of people.

There is a second trap even when the fit is right. People who know you are too nice. Candid, critical feedback is rare, and soft encouragement from a friend reads exactly like validation while telling you nothing. If your entire evidence base is people who like you, you do not have evidence.

Communities, the slow powerful one

Communities are probably the most powerful channel on this list and the hardest to build, because they run on trust and trust runs on time. The rule that decides whether it works: build the community to help answer a question, never as a list to send products to. Plenty of products exist because a community had a need and somebody in it went and built the answer. Almost none exist because a founder assembled an audience and then looked for something to sell them.

If you are already in a community that holds the right people, use it now. If you are starting one today, start it and expect nothing from it for a long time. Both of those are correct at the same time.

Product Hunt and the directories

Product Hunt is a lottery, and so is every other product discovery directory. Play them anyway. They are free and they take a minimal amount of time, so the expected value is fine even though any single ticket probably loses.

One condition: turn up with your messaging straight. The problem you solve, how you solve it, why you solve it, and why it beats what people use today. Plus a rough user experience that holds together. Launching blind with no story is how you waste the free ticket, and that story only exists if you did the person work first.

Referrals, which need a customer

A referral or affiliate network is what genuinely scales a product later. It is also the one thing on this list that cannot work yet. A referral scheme with no customers is machinery with nothing going into it: there is nobody to do the referring. Get the first customer, then build the engine that multiplies them. In that order, always.

Which channel should you commit to?

One of them, chosen by how fast customers are actually arriving. Setting count targets means chasing a number that may not suit your business at all, so watch the speed instead: which channel is producing conversations this week, at what rate, and pick that one to get good at. One channel done properly beats five done occasionally, and the founders who spread themselves across every platform tend to have presence everywhere and traction nowhere.

There is a second filter worth applying, which is whether you own the audience or rent it. A following on Facebook, YouTube, Instagram or LinkedIn is rented. The platform can change the algorithm or foreclose on you at any time, and all of them keep drifting toward pay to play, even inside your own network. Those platforms still matter as a trust signal, but cracking them organically now takes serious paid money.

What holds up is an audience you own, plus the smaller places your customer hangs out where the algorithms have not been neutered yet. When I asked ten or fifteen of my earliest followers where they actually spend their time, the answer was Substack, where a whole community of people doing exactly what I do already lives. So that is where I am building, owning the following and taking the small arbitrage while it lasts. Your answer will be a different place, and the only way to get it is to ask them, which is also building something you own instead.

What do you do tomorrow morning?

Find one person. Not a channel strategy, not a spreadsheet of platforms. Open the search, find the individual who best exemplifies the customer you built this for, connect, and try to start a conversation. Then do it again the next morning.

The reason this works is that it survives a real founder's day. You have five minutes while the coffee brews, five minutes on the commute, five minutes sitting on the toilet with your phone. Founders wear too many hats and this job always slides to the bottom of the list, under one more feature and one more blog post, when it is the thing every other thing depends on. Five minutes a day, every day, compounds faster than a heroic week you never repeat. The habit beats the tactic.

The expensive part was never the message. It is the finding: working out which people are worth the five minutes, and reading enough about each one to say something true. That is the part I take off you. Paste your site, and overnight I find the people describing your problem this week and hand you today's people found overnight with the reason they are on the list. You do the talking, from your own account, in your own words. What that adds up to over a few weeks is the conversation maths behind your first ten.