Favors.dev
You get your first 10 paying customers by having roughly 10 to 40 individual conversations with people who already have the problem, asking each of them directly for money, and writing down why the ones who said no said no. There is no channel that does this for you. Every marketing tactic you have read about is designed to produce volume at the top of a funnel, and the first ten payments are not a volume problem — they are a sequence of specific people deciding to trust a stranger with a card number.
I want to be direct about the thing that pushed me to write this one. Almost every guide ranking for this query is a list of nine or ten channels — network, cold email, communities, events, PR, content — and the better ones dress it up with a funnel: 500 prospects, 50 demos, 10 customers. Those conversion figures are presented without a source, and I could not verify a single one of them. So this post contains no invented benchmarks. Instead it gives you the structure that produces your numbers inside two weeks, plus the parts the channel listicles skip entirely: what to say when someone objects, what to charge, and how to tell whether ten closes mean anything.
Why the first 10 payments break the marketing playbook
The first 100 users and the first 10 paying customers are different problems with different solutions, and conflating them is the most common reason a founder ends up with traffic and no revenue. Users arrive through channels. Payments arrive through conversations.
The free channels that reliably produce early signups — Product Hunt Upcoming, an Indie Hackers launch, the founder subreddits, a couple of cheap paid listings — are genuinely good at what they do, and the first-100-users playbook maps them properly. But they select for curiosity. A person who clicks a launch post is sampling; a person who pays has decided that a cost they already carry is worth paying to remove. Those are not the same person early on, and the founder who runs a launch, collects 300 signups, sees zero conversions and concludes the product is broken has usually just used the wrong instrument.
This is the point Paul Graham made in Do Things that Don't Scale, and it has aged better than almost anything else written about early startups: founders recruit their first users manually, one at a time, because at the start there is no other way — you cannot wait for people to come to you. The reason it keeps needing to be said is that manual recruiting feels like failure. It is slow, it does not look like a business, and there is no dashboard for it. That discomfort is precisely why most founders skip to the channel work, and why the channel work so often produces nothing.
One reframe makes the whole phase easier: you are not selling yet. You are running the last stage of customer discovery, the one where the question stops being “is this problem real?” and becomes “is it worth money to this person, and can this person authorize money?” A price quote is a research instrument. It is the only question in the whole discovery process that cannot be answered politely.
Where the first 10 actually come from
In descending order of yield — and almost the exact inverse of the order founders work them. The pattern is consistent enough to be predictive: the colder the channel, the more comfortable it feels, because rejection from a stranger costs nothing. Start at the top and do not move down a rung until the one above it is genuinely exhausted.
People you already have a reason to email
Not “your network” as a vague asset — a written list of names. Former colleagues, people you worked with two jobs ago, the folks whose questions you answered in a Slack, anyone who has ever said “tell me when it's ready.” Write out thirty names before you write a single message. The list is always longer than founders think, and it is always shorter than it needs to be, which is exactly the useful discovery.
The ask — A specific, small ask about their situation — never a broadcast announcement.
The people you already interviewed
If you did discovery properly, you have a list of people who described the problem to you in their own words, unprompted. They are the warmest buyers on earth and most founders never go back to them. Going back closes the loop: here's what you said, here's what I built, here's the part I built because of you.
The ask — “You told me X in March. I built the thing that fixes X. Can I set you up Tuesday?”
Second-degree intros from people you have helped
One remove from your own network is where the honest conversations live — close enough that they take the meeting, far enough that they will tell you no. The constraint is that intros are a favor, and most founders have no favor balance to draw on. That is a solvable problem, and solving it is the whole point of the last section of this post.
The ask — Ask the introducer for one name, not for “anyone who might be a fit.”
People publicly describing the problem right now
A Reddit thread, a one-star review of the incumbent, a forum post from last week, a GitHub issue. Pre-qualified by definition and, crucially, timed — they have the problem today rather than in principle. Reply in public, be useful without pitching, and let them ask what you do.
The ask — Answer their question completely first. The pitch is the second message, or there isn't one.
Cold outreach, deliberately last
Cold email works, and it is the highest-volume, lowest-yield channel on this list. Which makes it the wrong place to start when you have ten to close and no idea yet what the pitch is. Cold prospects teach you nothing about why someone buys, because the ones who ignore you never tell you why.
The ask — Only after the first three closes have taught you which sentence makes people lean in.
Two notes on the warm end, because that is where the closes are. First: “warm” does not mean friendly, it means the person has a reason to reply. A former colleague who barely liked you will still open your email; a stranger who would love your product will not. Second: the ask has to be small and specific enough to be answered in one line. “Would you take a look at what I'm building?” asks for an unbounded favor. “You spent last year doing X manually — can I show you the thing that does it, ten minutes on Tuesday?” asks for ten minutes.
Finding rooms with buyers in them, not browsers
Most founder communities are full of founders, and founders are the hardest people on earth to sell software to — they have no budget, they can build it themselves, and they are all pitching each other simultaneously. That is not an argument against founder communities. It is an argument for knowing which job you are hiring each room to do.
- Rooms that produce buyersare organized around the job your product does, not around building products. The operations Slack, the industry Discord, the subreddit for the profession, the forum where people post the actual problem. If the room's recurring topic is your product's category, the buyers are there.
- Rooms that produce feedback, testers, and reach are the founder communities. They are excellent for that and they will not fund you — treat a signup from one as a favor, not a sale. The map of indie hacker communities sorts them by what each is genuinely good for.
- The exception that matters: if you sell to founders — a dev tool, a marketing product, anything indie-adjacent — then the founder rooms are your buyer rooms, and the etiquette becomes the whole game. Be useful for weeks before you are interesting for a day.
However you enter a room, the entry cost is the same: contribute before you appear. This is not a moral position, it is a mechanical one — a first post that is a pitch gets removed, and a founder who has answered questions for a month gets asked what they do. The rules for doing that without getting nuked are in how to market on Reddit without getting banned, and they generalize to every community on the list.
The founder-led sales call, without a sales background
Run it as a diagnosis, not a demo. The structure below fits in twenty-five minutes and requires no sales training, because it is almost entirely questions — the parts where you talk are short on purpose. Y Combinator's sales playbook for founders makes the same core argument from the other end of the telescope: at this stage the founder is the sales process, and a tight, question-led conversation beats a polished pitch.
- Two minutes: why you're both here. One sentence on what you build, then hand the floor back. Founders who spend nine minutes on context lose the call in the first ten.
- Ten minutes: how they do it today.The whole game. Ask them to walk you through the last time, ask what was open on their screen, ask what it cost them. You are looking for a workaround, because nobody builds a workaround for a problem they don't have.
- Five minutes: show only the part that matches. Not a tour. Open the one screen that does the thing they just described and let them react. If they went quiet during step two, you have nothing to show yet — say so and end early. That call was still a success.
- Three minutes: the price, said out loud, then silence. “It's $X a month.” Full stop. Do not fill the gap, do not pre-apologize, do not offer a discount into the silence. What they say next is the most valuable data in the call.
- Five minutes: the close or the reason.“Want me to set you up?” If yes, do it while you are both on the call — do not send a link. If no, ask the one question that pays for the whole conversation: what would have to be true?
The single hardest instruction in that list is the silence after the price. Every untrained founder discounts into the pause, and doing so destroys the only clean signal in the call — you never learn whether they would have paid the real number.
The five objections you will hear — and what to say
An objection is not a rejection; it is the buyer telling you exactly what is missing, for free. These five cover the overwhelming majority of what an unknown product with no logos and no case studies runs into. Log every one you hear in the tracker below — the repeats become your roadmap.
“We already use something for this.”
What's underneath — Not a rejection. It's proof the budget line exists — someone already decided this job is worth paying for.
What to say — “That makes sense — most people I talk to are on it. What's the part of it that still annoys you?” Then stop talking. If nothing annoys them, they were never your customer and you just saved a month.
“You're too new / too small. What if you disappear?”
What's underneath — The single honest objection to a one-person company, and the one founders answer worst because they answer it with confidence instead of terms.
What to say — Answer it with structure, not reassurance: month-to-month with no lock-in, a documented data export, and the direct line to you rather than a support queue. “You can leave any month and take your data with you” beats any promise about the roadmap.
“Send me some information and I'll look at it.”
What's underneath — A polite exit in almost every case. Real interest asks a question; deferred interest asks for a PDF.
What to say — “Happy to — is it worth a proper look now, or should I check back after your quarter?” The permission to say no makes the no arrive today instead of in six weeks of unanswered follow-ups.
“It's too expensive.”
What's underneath — Usually a value objection wearing a price costume — they haven't connected the number to a cost they already pay. Occasionally it's real, and then it's the most useful sentence in the call.
What to say — “Compared to what?” Then price the status quo with them out loud: the hours, the tool it replaces, the thing that broke last quarter. If the comparison still loses, ask what would have to be true at that price — and write the answer down, because it's your roadmap.
“I need to run it past someone.”
What's underneath — You have been talking to a user, not the buyer. Common in B2B and fatal if you leave it unaddressed.
What to say — “What will they ask you that you can't answer yet?” Build that one artifact — a security page, a number, a one-pager — and offer to join the conversation. Never hand your champion a pitch and hope they deliver it well.
There is a sixth that is not an objection and should be treated as a gift: “this doesn't do the thing I actually need.” Thank them, ask three more questions about the thing they actually need, and let the call run over. The most useful outcome of a lost deal is a sentence you can build against. Note also what solves several of these at once over time — public proof. Two honest reviews and one real testimonial answer the “who else uses this” question before it is asked, which is why getting testimonials early is a sales activity rather than a marketing one.
Pricing the first 10 (and what never to do)
Price the first ten as a hypothesis you intend to revise, not a commitment you have to defend. You will not get it right, and getting it wrong low is considerably more damaging than getting it wrong high — a high price produces an objection you can hear and learn from, while a low price produces silent agreement and a business that never works.
Four rules that hold at this stage:
- Charge something, always. Free users generate enthusiasm and no information. The first payment is the only clean proof that budget exists and that this person can authorize it.
- Discount the price, never the term. A founding rate at 30% off, locked for as long as they stay, rewards early risk without teaching the market that your product is cheap. Give it an end date and honor it.
- Never sell a lifetime deal at this stage. You are selling revenue you have not earned to fund a product you have not finished, to the segment least likely to still want it in a year. The cash feels like validation and is closer to a loan against your own roadmap.
- Write down what each of the ten actually paid. Bespoke deals are fine now and lethal at fifty. The spreadsheet is what lets you unwind them later without a fight.
One more thing worth doing before you settle on a number: show the pricing page to three founders who have priced a product before and ask what they would object to. It takes them ten minutes and it routinely surfaces the thing you have gone blind to. That kind of structured pre-launch feedback is much easier to get than founders assume, provided you are willing to give it back.
The first-10 tracker
Six columns, one row per conversation, filled in the moment the call ends. This is the distinct thing this post is asking you to do, because it converts ten anecdotes into the only sales data that will ever genuinely describe your product. Every benchmark you have read belongs to someone else's market. After ten rows, you have yours.
| Column | What you capture | Why it earns its place |
|---|---|---|
| Source | Exactly where this person came from | After ten rows you can see which source produced money, not just replies. This is the only channel data that will ever be genuinely yours. |
| Trigger | What changed in their world recently | New hire, new tool, a thing that broke, a deadline. The trigger is what separates a buyer from someone with the same problem who will buy in two years. |
| Job | What they said they wanted it to do, in their words | Ten quotes in one column is the fastest positioning exercise available to you. Copy the phrasing that repeats onto your homepage verbatim. |
| Objection | The one thing that nearly stopped it | The column that pays for the whole spreadsheet. Repeats here become FAQ entries, pricing changes, or the next thing you build. |
| Advance | What they gave you that cost them something | A calendar slot, their data, a card, an intro. Interest that costs nothing predicts nothing. |
| Outcome + days | Closed, lost, or stalled — and how long it took | Your real sales cycle, measured instead of guessed. Nobody else's benchmark applies to your product. |
How to read it at row 10
- One source produced most of the closes — that is your first channel. Go back and work it until it is empty before you add another.
- One objection repeated three or more times — stop selling for a day and fix it. It is costing you deals you never saw.
- The Job column says the same thing in three phrasings — that is your headline, and it is better than the one you wrote.
- The Trigger column is blank on most rows — you are selling to people who agree rather than people who are looking. Expect a long, quiet sales cycle until you find the trigger.
From 10 to repeatable: reading the pattern
Ten payments prove people will buy. They do not yet prove you can make it happen again, and the gap between those two facts is the entire job of an early startup. Steve Blank's definition is the cleanest statement of it I know: a startup is an organization built to search for a repeatable and scalable business model. Ten customers is the end of the first search loop, not the end of the search.
So run the diagnosis before you run at the next ten. Read the ten rows together, in one sitting, and answer three questions honestly:
- Did they arrive the same way? If eight came through one source, you have a channel to press. If they came from ten places, you have ten favors — a real achievement, and not yet a business.
- Did they buy for the same reason? One repeated trigger and one repeated job means the positioning is findable. Ten different reasons means you sold your persistence rather than your product.
- Are they still using it in thirty days? The question everyone postpones. Ten closes and two active accounts is a retention problem wearing a growth costume, and adding more customers will make it harder to see, not easier.
Which brings me to the part I have some standing to talk about. The highest-yield source on the list above — a warm introduction one step out from your own network — is not a channel you find. It is a balance you accumulate. Founders who get intros are, without exception, founders who have given them, and the reason most solo founders have none to draw on is that the giving happens months before the needing and nobody keeps score.
That is the specific problem Favors.dev exists to solve. It is a founder marketing co-op with a points economy: you earn points by doing verified favors for other founders — an honest review, structured feedback, a testimonial, a warm introduction — and you spend those points to pull the same help back when it is your turn. The score is kept, so the balance is real. You cannot spend what you have not earned, which is exactly why the well does not run dry in month three the way informal founder groups reliably do. The founder directory is a searchable list of people who have already opted into being useful — a materially better place to ask for an introduction than a cold inbox, and a good place to find the three people who will tell you the truth about your pricing page.
Your first ten paying customers will not come from a growth hack. They will come from about thirty conversations, five of which you will not enjoy, and a spreadsheet you keep honestly. Start with the warmest name on your list today, and ask for money in the first conversation rather than the fourth.
Frequently asked questions
How long should it take to get your first 10 paying customers?
Longer than the founder forums suggest and shorter than it feels while you're doing it — but the honest answer is that any specific number you read is invented. The variables that actually decide it are price point, whether one person can sign, and whether you already know people in the market. A $20/month tool bought by an individual can reach ten payments in a fortnight of concentrated outreach; a $500/month tool that needs a manager's approval can take a full quarter for the same ten, and neither pace tells you anything about the quality of the business. What matters far more than elapsed time is whether the reasons are converging. Ten closes over three months where every buyer named the same trigger is a business. Ten closes in three weeks for ten unrelated reasons is a run of luck you cannot repeat, and it will be indistinguishable from success right up until you try to get the next ten.
Should you charge your first customers or give it away for free?
Charge them, even if the number is small. A free user teaches you almost nothing, because the friction they were willing to overcome was zero and their enthusiasm carries no information. The moment money changes hands you learn three things at once: that the problem is worth budget, that this specific person can authorize spending, and what a real objection sounds like. The common compromise — free now, paid later — reliably fails, because the conversation you avoided in month one gets harder in month four when the product they've been using free is suddenly a line item. If you want to reward early risk, discount the price or extend the term. Do not zero it out.
What's the difference between getting your first 100 users and your first 10 paying customers?
One is a distribution problem and the other is a sales problem, and treating them as the same task is the most common reason founders with traffic have no revenue. First users come from channels: a launch, a directory, a community post, a search result — activity that produces volume at the top and lets you learn from behavior. First payments come from conversations, one at a time, where a specific person weighs a specific price against a specific cost they already carry. The channels that produce hundreds of signups routinely produce zero payments, which is not a failure of the channel but a category error about what it's for. Run both, but do not expect one to convert into the other on its own.
How do you sell if you've never done sales before?
Stop trying to sell and start trying to find out whether it's a fit, which is both more comfortable and more effective. Founder-led sales at this stage is a diagnostic conversation: ask what they do today, ask what it costs them, describe the one thing your product does about it, and then ask directly whether it's worth trying. The skills that actually matter are asking a question and not filling the silence afterward — no technique, no framework, no closing script. The genuine advantage you hold over any salesperson you might later hire is that you can change the product mid-conversation and say so, and early buyers respond to that far more strongly than to polish.
Where do you find your first paying customers with no audience?
In descending order of yield: people you already have a reason to contact, people you interviewed during discovery, warm introductions one step out from your own network, people publicly describing the problem this week, and — last — cold outreach. Founders invert that order almost universally, starting with the coldest, highest-volume channel because it feels like scale and because it avoids the discomfort of asking people who know you. The inversion costs weeks. Work the warm end until it is genuinely exhausted, and treat the intro layer as something you can build rather than something you either have or don't.