How long does it take to get your first 100 customers?

For a B2B SaaS at €30 to €300 a month, four to nine months of consistent daily distribution is the realistic band. Consumer apps at a few euros a month can reach 100 subscribers in one to three months because the price makes the decision small. The first ten customers usually arrive within two weeks of starting, because they come from people who already know you. The gap between customer ten and customer forty is the part that takes the longest.

Where do the first 10 customers come from?

Almost always from warm outreach: your own contacts, past clients, past colleagues, and the communities you were already in before you started building. Across 300 warm asks a solo founder can expect roughly 24 customers, and the first ten land fastest because you contact the most relevant people first. Cold outreach, posting and paid ads all take longer to produce a first customer than a list of 200 names does.

How many cold emails does it take to get one customer?

About 160 sends to a hand-built list, assuming a 12% reply rate, 35% of those repliers looking at the product, and 15% of those paying. That means 100 customers from cold email alone would take around 16,000 sends, which is why cold email works as one of four methods rather than as the whole plan. Bought lists convert at a third to a fifth of that rate.

Do I need an audience to get my first 100 customers?

No. Posting in public is one of the four ways to get customers and it is the slowest to pay out from a standing start. Founders with no audience get further faster by answering questions in communities where their buyers already gather than by posting to an empty timeline. Build the audience while warm and cold outreach are producing revenue, not instead of them.

Should I run ads before I have 100 customers?

Generally no, with one exception. Ads aimed at strangers cost roughly €150 per customer before you know whether those customers stay, which makes them the most expensive way to learn something you could have learned from 50 cold emails. Retargeting is the exception: once the free methods are producing 1,000 visitors a month, retargeting those visitors costs a third to a half of what ads to strangers cost.

How many customers should I ask for referrals from?

All of them, once each, at the point where they have visibly got value from the product. A named ask converts at roughly 20%, so eighty referral asks produce about sixteen customers. Ask "who else on your team deals with this" rather than "do you know anyone", because the first question has an answer and the second one does not.

What if I only need 20 customers, not 100?

Then the plan shrinks rather than changes. At €1,000 or more per engagement, ten clients is a full book and 100 is the wrong target entirely. Run warm and cold outreach only, skip posting and paid ads until the book is full, and expect three to four months to reach ten. The arithmetic in this playbook divides cleanly.

What is the fastest way to get your first customers for a SaaS?

Send 30 warm messages that ask for nothing, then a specific fifteen-minute ask to everyone who replies. That sequence produces paying customers within days rather than weeks, needs no audience and no budget, and it is the only method that works before anyone has heard of you.

← All posts

Method

How to get your first 100 customers: a playbook for solo founders

8 August 2026 · 18 min read

Get your first 100 customers by working the four ways to get customers in a fixed order: warm outreach first, because people who already know you buy at around 10% per ask, then posting in public and answering questions where your buyers gather, then cold outreach at volume, then paid ads last or never. A realistic blend over a hundred working sessions is 300 warm asks, 80 referral asks, 24 posts, 2,240 cold emails and one capped €600 ad test. The first ten customers usually arrive within two weeks. All hundred takes most solo founders four to nine months.

That paragraph is the whole answer. The rest of this page is the nineteen plays underneath it, the arithmetic that tells you whether your target is possible in the time you have, and four different starting points depending on what you sell. Every play has something you can do this morning and a number for what happens when you do.

There are only four ways to get a customer

Before the plays, the map. There are four things you can do to get someone to buy from you, and there is no fifth.

You reach out to people who already know you. You post something in public where people can find it. You reach out to strangers. You pay to get in front of people. Warm outreach, posting, cold outreach, paid ads. That is the entire list, and every tactic you have ever read about is one of those four wearing a hat. A webinar is posting. A conference is warm or cold outreach depending on whose badge you read. An affiliate deal is someone else posting on your behalf. A billboard is a paid ad with a worse click-through rate.

Alex Hormozi, a founder who wrote a book called $100M Leads about getting customers, calls these four the Core 4. That is the name that stuck, and we use it here because it is short. If the idea is new to you, read the Core 4 explained first. Knowing which of the four you are in is more useful than knowing which tactic you are running, and this playbook keeps answering that question.

The arithmetic first, because it decides everything else

Most first-100 advice skips the numbers. That is why founders spend eight weeks on something that could never have produced their target in the time available. Work it backwards instead. Here is what one paying customer costs in each of the four, at rates a solo founder can actually hit, with no brand, no team, and no case study to point at.

What one customer costs, by method

Method	One unit of work	Converts at	Units per customer	Units for all 100
Warm outreach	One personal ask to someone who knows your name	8% to 12%	10 asks	1,000 asks, which is more people than you know
Referral	One named ask to a customer who used the product twice	20%	5 asks	500 asks, so it grows only as your customer count does
Posting	One post plus twenty minutes of replies	0.05 customers per post early, 0.3 by month four	3 to 20 posts	400+ posts if nothing builds on itself, far fewer if it does
Cold outreach	One cold email sent to a hand-built list	0.6%	160 sends	16,000 sends, roughly 160 a day for a hundred days
Paid ads	One euro of ad spend at a €1.20 click	€150 per customer	€150	€15,000, before you have learned whether they stay

Read the last column and the conclusion is hard to avoid. None of the four gets you to 100 on its own. Warm runs out of people at around 300 asks. Cold runs out of hours at 16,000 sends. Posting runs out of patience somewhere around post 40, when the numbers still look like nothing. Paid runs out of money at €15,000, which is more than most solo founders have ever put into a business.

So the plan has to be a mix. Here is one that adds up honestly, spread across a hundred working sessions at five a week, which is roughly five months.

A hundred sessions that produce a hundred customers

Source	Sessions	Volume	Customers
Warm asks to your own list	30	300 asks	24
Referral asks to happy customers	8	80 named asks	16
Posts and community answers	24	24 posts, ~150 replies	9
Launch events	4	4 launches	11
Cold email	28	2,240 sends	13
One capped paid test	6	€600	4
Word of mouth and search you cannot attribute	0	the residue of everything above	23
Total	100		100

Two rows deserve a comment. The first is the 23 customers you cannot trace back to anything. That number is real and it is uncomfortable, and it is why founders who kill something on week-four data usually kill the wrong thing. Somebody read a post in March, forgot about it, saw your name in a subreddit in June, and signed up in July after googling your brand. All three touches were needed. Only the last one shows up in your analytics.

The second is the session count. A hundred sessions at five a week is twenty weeks. If you want 100 customers in 60 days, the same arithmetic says your daily volume has to nearly double, and at some point the sends per morning go past what a human can do by hand. The 100 in 100 goal calculator does that division for you and tells you whether your target survives it. The outreach volume estimator runs the same chain against your own reply and trial rates instead of the defaults here.

Where to start, by what you sell

The order changes with the price and the shape of the business. Starting in the wrong one costs a founder about two months. It is the most common mistake in the whole first-100 problem, and it is made in week one.

First 100 customers by business type

Business type	Typical price	Start with	Add second	What "100" should mean	Realistic timeline
B2B SaaS	€30 to €300 a month	Warm outreach	Cold outreach	100 accounts is roughly €3,000 to €12,000 a month	4 to 9 months
Consumer app	€3 to €10 a month, or a one-off	Posting and communities	Launch events	100 subscribers is a signal, not yet a business	1 to 3 months
Marketplace	A cut of each transaction	Warm and cold, for sellers only	Posting, for buyers	100 completed transactions in one city or category	3 to 6 months
Agency or high-ticket service	€1,000+ per engagement	Warm outreach	Cold outreach	100 is the wrong number. Ten clients is a full book	3 to 4 months to ten

B2B SaaS. Warm first, always. You are selling to a job title you can name, which means your own network holds either the buyer or someone who sits next to them. Cold outreach comes second because the list is findable and the message can be specific. Posting pays you back later, so start in month one and expect nothing until month four. Paid stays off until you know what a customer is worth after twelve months.

Consumer app. Warm outreach mostly fails here, and founders take that personally when they should take it as arithmetic. At €5 a month, a friend doing you a favour is €60 a year and a cancellation in March. Communities and posting come first, launches speed things up, and paid only makes sense once you have twelve months of retention data. The upside: 100 subscribers at consumer prices is a milestone you can reach in six weeks.

Marketplace. Solve one side by hand before you touch the other. Sellers are almost always the hard side, and they respond to warm and cold outreach, one conversation at a time, with no automation whatsoever. Buyers come from posting and communities once there is something worth showing up for. Keep the whole thing inside one city or one category until it works, because 100 transactions spread thinly across twelve categories teaches you nothing.

Agency or high-ticket service. Your target is wrong. At €1,000 or more per engagement, a full book is ten clients, and 100 would be a hiring problem rather than a distribution one. Run warm and cold outreach only. Skip posting and paid ads entirely until the book is full and you are choosing between clients.

If you genuinely cannot tell which of the four you belong in, the marketing channel quiz asks seven questions about your price, your network and your buyer, then names one to start with. Two minutes, free, no signup.

The nineteen plays

Everything below is a play you can start this morning. They are grouped by which of the four they belong to and roughly ordered by when a solo founder should run them. Nobody runs all nineteen. Pick the four or five that match your business type from the table above and run those into the ground before you add another.

Warm outreach

Write the list of 200 before you write anything else

One document with every human who would recognise your name. Old colleagues, university friends, the group chat, people you have replied to twice on the internet, the two customers from the consultancy you shut down. Write first. Judge later.

Who it works for: Every founder, on day one, with no audience and no budget. This is the only one of the four that can pay you in week one.
Do this today: Set a 45-minute timer. Export your phone contacts, your LinkedIn connections CSV, and every unique address in your sent mail from the last twelve months. Merge them into one sheet with a name, a channel and a blank column called "last contacted".
Realistic result: Most solo founders end up with 180 to 400 names and are surprised by the number. About a quarter are plausible users or plausible referrers, which gives you 50 real conversations before you have written a single cold email.

Send the message that asks for nothing

A short note that says what you built and asks for no favour at all. The ask comes later, from them, once they have asked what the thing is.

Who it works for: Anyone on the list you have not spoken to in over a year, which is usually most of it.
Do this today: Send ten. Something close to: "Been heads down building a tool for freelance bookkeepers for the last eight months. No ask, just thought you would find it funny where I ended up. How is the new job going?" Then stop and go do something else.
Realistic result: Reply rates between 40% and 60%, because you have made it easy to answer. Roughly a third of the people who reply ask what you built, and that question is worth more than any pitch you could have opened with.

Make the second ask small enough to say yes to in one line

One clear request with a time budget attached, sent only to people who already replied. Vague asks get vague answers.

Who it works for: Founders whose product is finished enough that fifteen minutes inside it produces an opinion.
Do this today: Reply to yesterday's repliers with one sentence: "Would you spend fifteen minutes in it this week and tell me the first thing that annoys you?" Put the link in the same message so nobody has to ask for it.
Realistic result: A quarter to a third of them actually go in. Around 40% of those become paying customers when the product solves a problem they already had. Across 100 warm asks that is about ten customers.

Reopen everyone who has ever paid you money

A separate list, kept separate on purpose: past clients, past employers who paid invoices, the person who bought your Gumroad thing in 2023.

Who it works for: Anyone who has consulted, freelanced, contracted or sold a product before. Skip it if this is genuinely your first commercial thing.
Do this today: List every person or company who has paid you in the last three years. Send five of them a note that opens with the work you did together and closes with one line on what you build now.
Realistic result: The best per-message conversion on this page. Between 20% and 30% either buy or introduce someone who does, because the trust question is already settled.

Ask for a name, never for "anyone"

The referral ask, made specific. "Do you know anyone who might like this" is a question people answer with silence. "Who else on your team deals with this" is a question with an obvious answer.

Who it works for: Any founder past their fifth customer. Ask right after something has visibly gone well inside the product, not on a random Tuesday.
Do this today: Open your usage data, find three customers who used the product twice in the last fortnight, and send each of them the named version of the question. One of the three will give you two names.
Realistic result: A named ask converts at roughly 20%, and most customers are worth asking twice over five months. That works out at about 0.4 new customers per happy customer, so forty of them are worth sixteen more. This is the number that stops the plan stalling at 60.

Posting

Post the thing you learned the same day you learn it

Building in public, with one rule that keeps it honest: the subject has to be something that actually happened in the last 24 hours.

Who it works for: Founders with under 1,000 followers who think they have nothing to say and therefore say nothing.
Do this today: Write down the single most surprising thing that happened in the product this week. A reason someone cancelled, a support ticket, a number you were wrong about. Post it in under 150 words with the number in the first line.
Realistic result: Under 1,000 followers, expect 400 to 1,500 views and two or three replies per post. Almost no customers for the first month. Post 60 beats post 6 by roughly ten to one, and that gap is the whole reason posting works.

Answer the DM in public

Every question a real prospect sends you privately is a post that somebody already wanted. Answer them, then publish the answer with the name stripped out.

Who it works for: Anyone getting even two or three questions a week from warm or cold conversations.
Do this today: Scroll your DMs and email for the last question a prospect asked you. Paste your answer into a post, add the background you assumed they had, publish it.
Realistic result: These do better than anything you invent from scratch, because somebody wanted the answer before you wrote it. One in five of these posts brings a reply from a stranger with the same question.

Answer questions where your buyers already are

Community answering. Subreddits, niche Slack and Discord servers, the forum your industry has used since 2009. You reply as a person who knows the answer, and you mention the product only where it genuinely applies.

Who it works for: Founders with zero audience. At zero followers this reliably beats posting to your own timeline, and most founders get that backwards.
Do this today: Find three places where your buyers ask each other questions. Sort by new. Write one genuinely useful answer of 200 words or more, with no link in it at all.
Realistic result: Two thirds of the nine customers in the posting row of the blend below come from here, not from your own timeline. A good answer in a niche subreddit keeps pulling traffic for about two years. A good post is dead in two days.

Take apart something specific and public

The teardown. You pick a real thing, a landing page, a signup flow, a pricing page, and explain in detail what it does and what it costs the person using it.

Who it works for: Founders whose product carries an opinion about how something should be done. It works badly for plain utilities.
Do this today: Screenshot one thing you have a strong view about. Write four paragraphs: what it is, what it costs the user, what you would change, what you would leave alone.
Realistic result: Teardowns get shared roughly three times as often as opinion posts, because they show their evidence. This is the format most likely to earn you a link from someone who already has an audience.

Spend fifteen minutes replying before you post anything

A floor you do not go below. Fifteen comments on other people's posts, written before yours goes up, on accounts your buyers already read.

Who it works for: Everyone who posts. Publishing into a void and calling it distribution is the most common way founders waste a hundred days.
Do this today: Make a list of twelve accounts your buyers read. Reply properly to five of them this morning. No compliments, no "great post", just the thing you would say if you were in the room.
Realistic result: Follower growth roughly doubles for the same number of posts. Two or three of those twelve accounts will eventually reply to you in public, which is worth more reach than any first line you could write.

Treat launches as events, and have four of them

Product Hunt, Show HN, the relevant subreddit, a swap with a newsletter your buyers read, a changelog post written like a launch. Each one is a planned spike, not a plan.

Who it works for: Anyone with a working product and a landing page that explains it in one sentence.
Do this today: Diary four launch dates across the next twenty weeks and write the one-sentence pitch for the first. Warm the audience up in the four weeks before it, which is the part everyone skips.
Realistic result: A niche B2B launch lands 200 to 600 visits and five to fifteen signups. Two to four paying customers each, so eleven across four events. The trickle of traffic afterwards matters more than launch day.

Cold outreach

Run fifty messages before you conclude anything

The smallest honest test of a cold email. Below fifty sends a reply rate tells you nothing, and above two hundred you are scaling something you have not tested yet.

Who it works for: B2B founders selling to a job title you can name and find. If you cannot name the job title, you are not ready to send anything cold.
Do this today: Pick one type of company, write one message, send fifty over two sessions. Change nothing in the middle, however tempting the first ten silences make it.
Realistic result: Under 4% replies means the list or the offer is wrong and no amount of copy editing fixes it. Between 4% and 10% means keep tuning. Above 10% means send more of exactly that, immediately.

Build the list by hand with one hard filter

Forty companies chosen because of one specific fact you can see from the outside, not because they turned up in a scraped export of 8,000 rows.

Who it works for: Anyone whose reply rates have collapsed under 3%, which is nearly always a list problem dressed up as a copy problem.
Do this today: Write down the one thing that must be true for a company to need you this quarter. They just hired their second bookkeeper. They posted a job for the work your product does. Then find forty companies where it is true.
Realistic result: Hand-built lists reply at three to five times the rate of bought ones. Forty good names beat 4,000 bad ones and take about ninety minutes to put together. The templates in our cold email guide for SaaS founders assume a list built this way.

Send three lines, one of them personal

A cold email short enough to read on a phone in the lift. One line about them, one line about the problem, one question they can answer with a yes.

Who it works for: Every cold sender. Length is the single most reliable predictor of a low reply rate.
Do this today: Take your current cold email and delete everything except the observation, the problem and the question. Send that version to the next forty names. Keep the old one for comparison.
Realistic result: Reply rates of 8% to 15% on a hand-built list, against 2% to 4% for a paragraph about features. Around 35% of the people who reply will look at the product, and about 15% of those pay.

Write the two follow-ups on the day you write the first email

Three messages, all written up front. The second adds one piece of evidence. The third closes the loop and gives them permission to say no.

Who it works for: Anyone currently sending one email and reading silence as rejection.
Do this today: For the batch you send this morning, write both follow-ups now and diary them for day four and day eleven. Neither one contains the phrase "just circling back".
Realistic result: Roughly half of all replies to a cold sequence arrive after the first message. Founders who send once are throwing away half their pipeline because a second email feels rude.

Move to the channel where your buyer actually reads

The same three-line message, sent on LinkedIn, X, Instagram or a community DM instead of by email, picked by where your buyer clearly spends their attention.

Who it works for: Founders selling to designers, creators, developers and small operators, all of whom have hostile inboxes and open DMs.
Do this today: Take the ten best names from your list and check where each of them posted most recently. Send the same message there instead. Compare the reply rate against your email batch after two weeks.
Realistic result: DM reply rates run two to three times email in consumer-ish niches and worse than email in regulated B2B. Ten minutes of checking tells you which world you are in.

Paid ads

Retarget the people who already came and left

The one kind of ad spend that reliably pays back before you know your numbers. You advertise to people who visited your site, read the docs, or started a trial and stopped.

Who it works for: Founders with at least 1,000 monthly visitors from the three free methods above. Below that there is nobody to retarget and you are just buying strangers.
Do this today: Install the pixel. That is the whole task today. Launch nothing until the audience has enough people in it to be worth talking to.
Realistic result: A third to a half the cost per customer of ads aimed at strangers, and it stays cheap only while it stays small. Retargeting multiplies the other three. On its own it has nothing to work with.

Optimise toward the paid event, whatever the platform suggests

Telling the campaign to chase the moment money changes hands, rather than installs, signups or link clicks.

Who it works for: Anyone about to spend their first euro. The default setting is almost always the wrong one and it is expensive to find that out slowly.
Do this today: Check what your campaign is currently optimising for. If it is anything that happens before payment, and you can fire a purchase event, change it before the next impression.
Realistic result: We wrote up a €258 test that produced 21 signups and zero customers for exactly this reason. Chasing the wrong event buys you the people most willing to sign up and least willing to pay.

Give every test a budget and a kill date before it starts

One variable, a fixed cap, and a date written down in advance at which you stop no matter how promising it looks.

Who it works for: Every founder who has ever said "let it run another week" about an ad set.
Do this today: If you are running paid at all, write the cap and the kill date next to the campaign name right now. €300 and fourteen days is a sensible default for a first test.
Realistic result: You will kill about four tests in five. The value is the four you killed at €300 instead of at €2,000, which is roughly what paid ads costs an unprepared solo founder before they give up.

How to actually run this for a hundred days

The blend table asks for all four across twenty weeks. The obvious reading is that you do a bit of each every morning, and that reading is wrong. Splitting ninety minutes across warm, posting, cold and paid gives you four half-efforts, none at a volume that tells you anything, and after three weeks you know nothing about any of them.

Rotate by day instead. Monday is warm and only warm: ten asks, nothing else. Tuesday is posting: one post and twenty minutes of replies. Wednesday is cold: eighty sends from a list you built last week. Each morning is one thing at full intensity, which is what actually matters, and the week still covers all four so you are not blind in three of them for a quarter. We wrote up the full compression in the Rule of 100 in a two-hour day.

Weight the rotation toward whichever one holds your stored energy. Two hundred names in a spreadsheet is stored energy, so warm gets three mornings a week until the list is spent. Six years of unpublished opinions about your industry is stored energy, so posting gets three. The one where you already have raw material goes first, because it produces something in week one, and week one is when most hundred-day plans die.

Count sessions, not calendar days. Session 31 is your thirty-first, whenever it happens. A customer emergency on a Tuesday moves the finish line by one day and costs you nothing else, which matters because being behind schedule is a much better excuse to quit than being tired.

What goes wrong at customer 40

The first ten come from people who like you. Customers eleven to forty come from the rest of the warm list and from your first cold experiments, and they arrive slower and with more objections. Then the list is empty. This is where most founders stop.

There are three moves at customer 40, and "post more" is not one of them. Ask every existing customer for one named referral, which at 20% per ask turns forty customers into eight more. Rebuild the cold list with a harder filter, because a list that stops working was almost always too broad rather than too small. Publish what you have learned from forty customers, which is now genuinely worth reading and was not when you had two.

The founders who reach 100 are rarely the ones with the best tactics. They are the ones still sending on session 70, when the novelty is long gone, the graph is boring, and nothing about the morning feels like progress. Deciding what to do is the expensive part. Doing it is easy once somebody has decided for you.

Frequently asked questions

How long does it take to get your first 100 customers?: For a B2B SaaS at €30 to €300 a month, four to nine months of consistent daily distribution is the realistic band. Consumer apps at a few euros a month can reach 100 subscribers in one to three months because the price makes the decision small. The first ten customers usually arrive within two weeks of starting, because they come from people who already know you. The gap between customer ten and customer forty is the part that takes the longest.
Where do the first 10 customers come from?: Almost always from warm outreach: your own contacts, past clients, past colleagues, and the communities you were already in before you started building. Across 300 warm asks a solo founder can expect roughly 24 customers, and the first ten land fastest because you contact the most relevant people first. Cold outreach, posting and paid ads all take longer to produce a first customer than a list of 200 names does.
How many cold emails does it take to get one customer?: About 160 sends to a hand-built list, assuming a 12% reply rate, 35% of those repliers looking at the product, and 15% of those paying. That means 100 customers from cold email alone would take around 16,000 sends, which is why cold email works as one of four methods rather than as the whole plan. Bought lists convert at a third to a fifth of that rate.
Do I need an audience to get my first 100 customers?: No. Posting in public is one of the four ways to get customers and it is the slowest to pay out from a standing start. Founders with no audience get further faster by answering questions in communities where their buyers already gather than by posting to an empty timeline. Build the audience while warm and cold outreach are producing revenue, not instead of them.
Should I run ads before I have 100 customers?: Generally no, with one exception. Ads aimed at strangers cost roughly €150 per customer before you know whether those customers stay, which makes them the most expensive way to learn something you could have learned from 50 cold emails. Retargeting is the exception: once the free methods are producing 1,000 visitors a month, retargeting those visitors costs a third to a half of what ads to strangers cost.
How many customers should I ask for referrals from?: All of them, once each, at the point where they have visibly got value from the product. A named ask converts at roughly 20%, so eighty referral asks produce about sixteen customers. Ask "who else on your team deals with this" rather than "do you know anyone", because the first question has an answer and the second one does not.
What if I only need 20 customers, not 100?: Then the plan shrinks rather than changes. At €1,000 or more per engagement, ten clients is a full book and 100 is the wrong target entirely. Run warm and cold outreach only, skip posting and paid ads until the book is full, and expect three to four months to reach ten. The arithmetic in this playbook divides cleanly.
What is the fastest way to get your first customers for a SaaS?: Send 30 warm messages that ask for nothing, then a specific fifteen-minute ask to everyone who replies. That sequence produces paying customers within days rather than weeks, needs no audience and no budget, and it is the only method that works before anyone has heard of you.