How do I get my first 10 customers for a B2B SaaS?

Start with people who know your work and ask them for introductions to named companies. Turn three to five good fits into paying design partners with written terms. Then write to strangers who show a reason to need you now, like a job post for the work your product supports, with one question and no pitch. Launches help with feedback more than with the first contracts.

How do you get SaaS customers with no network?

Replace introductions with signals. Look for people who posted about the problem, companies hiring for the job your product does, and teams that raised a round in the last few months, then ask each one question about that reason. Answer questions where those buyers ask peers for help, and launch on Show HN or Product Hunt once a stranger can try the product alone.

Should my first SaaS customers pay?

Yes. Payment is the proof that the problem is worth money, and a paying customer gives better feedback than a free tester. Offer design partners a discount or a later start date in exchange for a reference call and a case study, but write the price and the date down before they log in.

Is Product Hunt worth it for B2B SaaS?

Worth one launch for feedback and early users, especially for tools that makers and product people use themselves. It costs nothing, and you can schedule the date up to 30 days ahead. Expect signups more than signed contracts, so judge the launch by the conversations it starts with people who fit your buyer profile.

When should a SaaS founder stop running sales?

Not before the pattern repeats. Keep selling until your last several deals closed the same way and your calendar, not a shortage of prospects, is what slows you down. Then hire someone to run a process you can write down. A hire made earlier inherits the job of learning your market, which is the founder's job.

← All posts

November 8, 2026

B2B SaaS: How to Get Your First Customers, From Zero to 10

How to get first customers for SaaS sold to businesses: your network, paying design partners, signal-based outreach, launches, and pricing the first ten.

By PancakeLast updated November 8, 2026

TL;DR: Your first ten B2B SaaS customers come from conversations you start yourself. Begin with people who already know your work, turn three to five of them into paying design partners, then write to strangers who show a reason to need you this quarter: a post about the problem, a job opening for the work your product supports, a fresh funding round. Launches and communities bring feedback and a burst of signups. Count on them for learning, not for the first contracts.

This is the playbook for technical and solo founders, pre-seed to Series A, selling software to other businesses. It ranks seven channels, shows what to say in a first message, prices the first deals, and ends with a selling week a solo founder can keep while still shipping. Past customer ten, the channels stay the same. What changes is who runs them.

Write the first-customer profile before the first message

Your ideal customer profile can stay broad for now. Your first-customer profile can't. Write down one buyer, one company type and one moment that makes the problem urgent, because every channel below works better when you can name all three.

Say you're two founders with a usage-based billing tool for B2B software companies. The broad version reads "SaaS companies." The first-customer version reads: "Head of revenue operations at a B2B software company with 20 to 200 people, billing usage out of spreadsheets, in the quarter it launches a usage-based plan."

Three questions get you there:

Who feels the problem in their own week? That person answers messages about it.
Who signs? At a 30-person company that may be the CEO. At 300 people it may be a VP with a procurement process behind them.
What event makes it urgent now? A new pricing model, a new hire, a new round, an audit, a migration.

The third answer matters most. It tells you what to watch for, and it hands your first message a question only that buyer can answer.

How to get first customers for SaaS: seven channels compared

The ranking weighs what a two-person team gets back for its hours: how soon a channel can produce a paying customer, what it costs, and who it suits. Read the time column as judgment. It comes from how each channel works, and your market can move it either way.

#	Channel	Time to a first paying customer	What it costs	Suits	The trap
1	People who know your work	1 to 4 weeks	Favors and your time	Founders who worked in the buyer's industry	Friendly yeses that never pay
2	Design partners converted to paid	1 to 3 months	A discount and onboarding hours	Products that need real data or a real workflow	Free pilots with no end date
3	Signal-based founder outreach	2 to 6 weeks	Your hours, or a tool	Buyers who leave public traces: posts, job ads, funding news	Messages with no reason in them
4	Communities where buyers ask for help	1 to 3 months	Your hours	Founders who know the domain cold	Promotion bans, and answering peers instead of buyers
5	Launch platforms (Product Hunt, Show HN)	Signups in a day, customers later	A week of preparation	Developer tools and products that makers use	Signups who never pay
6	Integration directories	Several months	Build and review time	Products that plug into a platform your buyers pay for	Listing the app, then waiting
7	Content and search	Six months or more	Your hours	Customers twenty and beyond	Too slow to land the first ten

Work channels 1 to 3 in parallel from week one. Add a launch once a stranger can use the product without you on the call, and start writing content when real customers give you stories worth telling.

Start with the people who already know your work

The warmest list you own is your last two jobs. Former colleagues, former customers and former managers know whether you ship, and some of them now work at companies that match your profile.

Paul Graham's essay Do Things That Don't Scale names the job plainly: founders have to "recruit users manually." One of his examples is Stripe's founders, who set the product up on a prospect's laptop during the conversation instead of promising to send a link. The lesson for B2B is the same. Ask for the next step while the person is still talking to you.

Make three lists:

People who had the problem with you. If you fought usage billing at your last company, the RevOps and billing people you worked beside come first.
People one introduction away. Ask each investor, advisor and former manager for introductions to two named companies, not "anyone who might be interested."
Your investors' other companies. If you raised, your investors' portfolio is a list of real businesses that already take their calls. Pick the ones that fit and ask for an introduction to a named role.

The question you ask decides what you learn. "Would you use this?" draws a polite yes. "Who on your team owns usage billing today, and what do they use?" gets you a name and a fact. Then ask for 20 minutes to run the workflow on their own data.

Two warnings. Friends say yes more easily than they pay, so name your price in the first conversation. And a network is finite. When the warm names are used up, the next customers come from strangers, so start the signal work in week one, not when the list runs dry. Solo founders without an industry network should lean on signals from the start.

Turn three design partners into paying customers

A design partner is an early customer who gets a say in the product in exchange for real usage, blunt feedback and patience with rough edges. The goal is a paying customer who will take a reference call, not a free tester.

Write the terms down before anyone logs in:

A start date and an end date. Sixty or 90 days lets a monthly workflow run at least twice.
One measure. Pick a number they already count: hours spent closing the month, invoices fixed by hand, tickets reopened.
The price after. Either a discounted price from day one, or the list price starting on a fixed date. A free arrangement with no date attached tends to fade out.
What you get back. A named contact, a short call every two weeks, and permission to publish a case study if the measure moves.

Three to five partners is the useful range. With one, you build for one company. With ten, you can't keep up with what they ask for.

Pick partners who match the first-customer profile, not the friendliest people you know. A poor fit pulls the roadmap toward a market you don't want. If you sell an AI product to operations teams, pilots carry more weight still: first customers for an AI startup covers pilot design and the security review.

Write to strangers who have a reason to talk now

Once the warm list thins, the founder writes to strangers. It stays the founder's job at this stage, because every reply teaches you an objection, a phrase or a price point that a hire might not pass on. The founder-led sales playbook covers the calls and the close. This section covers who to write to.

A bought list fails early teams for one reason: nothing in it says why this person should talk to you this month. A signal does. Five work for most B2B software:

A post about the problem. Someone writes about invoicing usage by hand, or asks peers which billing tool to pick.
A job post for the work your product does or supports. A company hiring a billing specialist has the problem and a budget line. Job postings as buying signals maps roles to the purchases they predict.
A fresh funding round. The money arrives with a plan, and the plan names the teams that will grow. How to find startups that raised money lists the public sources.
Engagement with a competitor's posts. People who comment on a rival's launch already understand the category.
Lookalikes of your best customer. After one happy customer, companies with the same size, model and buyer are your next list.

These are the same signals behind Pancake for SaaS startups, our product's page for software founders.

The first message asks one question about the signal and pitches nothing. No deck, no demo link, no feature list:

"Saw the billing specialist opening. Is that to keep up with usage invoices, or to clean up the ones already sent?"

The answer tells you whether a call is worth anyone's time. Tag each conversation with the signal that started it. Within a few weeks you'll see which reasons open doors for your product, and you can stop watching the rest.

Launches and communities: feedback and a spike, not a pipeline

A launch buys a day of attention and a stack of comments. Product Hunt describes its own community as early adopters, makers, product people, founders and investors, which is a fine crowd for a developer tool and a thin one for software sold to plant managers or clinic owners.

The rules are worth knowing before you pick a date. Per Product Hunt's launch FAQ, as of November 2026: launching is free, a launch goes live at 12:01 a.m. Pacific Time on the date you schedule, and you can book that date up to 30 days ahead. Launches come from personal accounts, not company accounts. The team decides which launches get featured on the homepage, and asking anyone for upvotes breaks the rules.

Show HN suits technical products. Hacker News asks for something people can try, ideally without a signup in the way, and rules out landing pages and fundraisers. The comments can be blunt. Answer every one, and write down the objections, because you'll hear them again on sales calls.

Communities run on a longer clock. Find where your buyer asks peers for help: a Slack group for RevOps people, a subreddit for a tool they use, the forum of the platform they build on. Check what each one allows before you mention a product, answer with specifics from your own work, and keep the link in your bio rather than in the reply. A useful public answer reaches the person who asked and everyone else who had the same problem.

Two slower channels round out the table. If your product plugs into a platform your buyers already pay for, list it in that platform's app directory once the integration is stable. Content and search compound for years, but they start too slowly to land the first ten customers. Judge every one of these by the conversations it starts with people who fit the profile, not by upvotes or followers.

Price the first ten so they count

A customer who pays tells you what a free user can't: the problem is worth money. Charge from customer one, even at a discount.

Publish a list price, then trade discounts for things you need. A design-partner discount can buy a case study, a reference call or permission to show their logo.
Prefer annual terms where the buyer allows it. A yearly contract with a modest discount gives you cash and a full year of feedback. Monthly billing is fine when the buyer insists.
Write the scope down. "We'll build that for you" in a first deal becomes a roadmap you no longer control.

The arithmetic of ten customers

Work backwards from your own ratios. Say one first message in five gets a reply, one reply in three turns into a call, and one call in four turns into a paying customer. Ten customers then take 40 calls, 120 replies and about 600 first messages: roughly 30 a week for five months.

Those ratios are placeholders. Replace them with yours after your first 50 messages, and the plan for the next quarter writes itself. Better signals move the first ratio more than anything else you can change.

A solo founder's selling week

Solo founders sell between builds, so the routine has to be small enough to survive a bad week. Here's a week we'd start with, about five hours in all:

Monday, 90 minutes. Pick 30 people or companies showing a signal and write their first messages.
Tuesday and Thursday, 30 minutes each. Answer replies and book calls.
Wednesday and Friday afternoons. Up to four calls in total, 30 minutes each.
Friday, 30 minutes. Update the pipeline sheet and write down every objection you heard that week.

With a cofounder, one person owns this list. When two people share follow-ups, each can assume the other sent them.

When the calendar fills and qualified people wait for a slot, the question becomes whether to hire. Pancake vs hiring a BDR runs the cost of a first sales hire line by line.

Running the outreach half with Pancake

Pancake is our product, and it runs the signal work from the section on strangers: it finds people showing a buying signal and starts the conversation under your name. The calls and the close stay with you. For a B2B SaaS founder, three signals carry most of the weight:

companies hiring for the job your product does or supports;
people posting about the problem you solve;
buyers engaging with a rival's posts.

A Play is one sentence, such as "Find heads of revenue operations at US B2B software companies with 20 to 200 people that are hiring a billing specialist." It runs on the schedule you pick. Each lead comes with one line on why it fits, checked against the ideal customer profile in your GTM Brain. The first message asks about the signal and pitches nothing: "Saw the billing specialist role. Are usage invoices still built in a spreadsheet?" A reply stops the sequence and lands with you.

The limits: Pancake doesn't send email or make calls, and it won't run your demo. It fits buyers who post, hire or raise money in public, so software sold to small shops whose owners never post will find little there.

The plan starts from $99 a month per workspace and includes 60,000 credits a month (pricing). A 7-day free trial with 30,000 credits lets you test a Play first, and starting it takes a card. Pancake for SaaS startups shows example Plays for software founders.

Frequently asked questions

How do I get my first 10 customers for a B2B SaaS?: Start with people who know your work and ask them for introductions to named companies. Turn three to five good fits into paying design partners with written terms. Then write to strangers who show a reason to need you now, like a job post for the work your product supports, with one question and no pitch. Launches help with feedback more than with the first contracts.
How do you get SaaS customers with no network?: Replace introductions with signals. Look for people who posted about the problem, companies hiring for the job your product does, and teams that raised a round in the last few months, then ask each one question about that reason. Answer questions where those buyers ask peers for help, and launch on Show HN or Product Hunt once a stranger can try the product alone.
Should my first SaaS customers pay?: Yes. Payment is the proof that the problem is worth money, and a paying customer gives better feedback than a free tester. Offer design partners a discount or a later start date in exchange for a reference call and a case study, but write the price and the date down before they log in.
Is Product Hunt worth it for B2B SaaS?: Worth one launch for feedback and early users, especially for tools that makers and product people use themselves. It costs nothing, and you can schedule the date up to 30 days ahead. Expect signups more than signed contracts, so judge the launch by the conversations it starts with people who fit your buyer profile.
When should a SaaS founder stop running sales?: Not before the pattern repeats. Keep selling until your last several deals closed the same way and your calendar, not a shortage of prospects, is what slows you down. Then hire someone to run a process you can write down. A hire made earlier inherits the job of learning your market, which is the founder's job.