How to Get Your First 100 SaaS Customers, Stage by Stage

Short answer: don't plan for 100 customers. Plan for 10, then 30, then 100, because each stage has a different job. The first 10 come from you, one conversation at a time, with people you can name. The next 20 come from repeating whatever found those 10, in one or two places your buyers already gather. The last 70 come from things that keep working when you stop: pages that rank, listings, and your product living inside products your buyers already use.

The numbers are rough markers, not a benchmark. Some products need well over ten customers before a pattern shows, and a high-priced B2B tool can be a real business well before 100. What matters is that the job changes on the way, and most founders who stall are still doing the first stage's job at the third.

This post is the plan. If you want the argument for borrowing other products' distribution instead of doing more of your own marketing, that's how to get customers for your SaaS without ads.

Why plan for 10, 30 and 100 instead of 100?

Because the question you're answering changes at each stage, and a tactic that answers one question well answers the next one badly.

At the first stage you're learning, so efficiency doesn't matter. At the second you're looking for a source you can repeat. At the third you're building things that bring customers without you. Skipping ahead is the common mistake: writing search pages before you know the words your buyers use, or buying ads before you know who converts.

How do I get my first 10 SaaS customers?

By hand. That's the honest answer, and it's why Paul Graham's essay "Do Things That Don't Scale" is still the standard advice for this stage. Five steps:

1. Write down 50 people you can name who have the problem. Not personas: names. Former colleagues, people who complained about the problem in a community, people who use the tool you replace.
2. Talk before you pitch. Ask how they handle the problem today, what it costs them and what they tried. The ones who lean in get a demo. The ones who shrug tell you something too.
3. Go where they ask questions. The forum, subreddit or Slack group of their profession. Answer the question fully and leave the link out. People look up the person who helped them.
4. Charge them. A paying customer tells you whether the problem is worth money. An early-customer price is fine; free forever is not.
5. Onboard every one yourself. Set it up with them on a call and watch where they stall. It's slow, and it's the cheapest product research you'll ever get.

A fictional example. You've built an invoicing tool for freelance translators. Your first ten: three translators you've worked with, two agency contacts, one stranger from a cold email, and four people from a translators' forum where you'd answered questions about billing clients in two currencies. Four from one place is the signal that matters.

What you're collecting here isn't revenue. It's the sentence your buyers use to describe the problem. That sentence becomes your headline, your search pages and your pitch for everything after.

How do I go from 10 to 30 customers?

Stop spreading. Look at where your first 10 came from, put most of your hours into the source that found the most of them, and test at most one new channel alongside it.

• If they came from a community, become the most useful person in it, every week, and find the next two places the same people gather. In the example, that's the translators' forum and two others like it.
• If they came from direct outreach, turn your best messages into a repeatable sequence and use your one-sentence description to build a longer list.
• One launch, done properly. A launch day can bring a burst of early adopters and feedback. It rarely brings a business on its own. Whether it's worth the preparation for your product is its own question: is Product Hunt worth it?
• A handful of directories whose visitors are your buyers. Not hundreds. A short list, in order: the best SaaS directories.

Before you commit months to a new channel, score it on five things: what it costs upfront, whether it scales without more of your time, how hard it is to deliver, what you lose if it fails, and how much trial and error it takes. How to evaluate a growth channel walks through each.

Two more jobs belong to this stage. First, write down what you do by hand. Your onboarding call becomes a checklist, then an email, then a setup screen; the knowledge stops living only in your head. Second, watch who leaves. If customer 25 arrives the week customer 7 cancels, you aren't at 25. Find out why they left before you pour more people into the same product.

How do I get from 30 to 100 customers?

Build things that keep working when you stop. You could find 70 more customers one by one, but then you have a sales job, not a product. The channels that carry this stage share one trait: work done once that keeps bringing people.

• Pages that answer your buyers' questions. Every question your first 30 asked before they bought is a search someone else is making. Write the page that answers it, in their words. It takes months to rank, and then it keeps working; AI assistants increasingly answer from the same kind of page.
• Alternatives and comparison listings. People searching for an alternative to a tool you replace are already shopping.
• Integrations with products your buyers already use. The next section is about this one, because most guides leave it out.
• Your happiest customers, in public. Ask for a review, a short case study or a mention in their own community. One honest sentence from a peer outweighs a page of your copy.

And count revenue, not only logos. At this stage the cheapest new revenue often comes from customers you already have: a higher plan, a paid add-on, a capability they currently buy somewhere else. You don't need more traffic, you need more to sell makes that case in full.

Which channels fit which stage?

Our read, not a benchmark: the same channel can be the best move at one stage and a waste of a month at another.

The pattern: the channels that win early cost you time for every customer, and the ones that win later cost you time once. Start the slow ones during the second stage, so they're ready when you need them.

Are integrations a real channel for your first 100 customers?

Most "first 100 customers" guides list your network, cold email, communities, content and ads. They rarely list integrations or partnerships. That's a gap, because for many small SaaS products the place their buyers spend the day is someone else's product.

The channel is one feature of your product running inside a product your buyers already use, at the moment they need it. The rule for picking the product: same customer, same job, different step. In the example, translators spend their day in a project tool for translation jobs; an invoice step at the end of each job is the next step of the job they're already doing. How to get a product to say yes is in how to get other SaaS products to integrate yours.

The real cost, before you start:

• A host has to say yes, build it and support it. That's a founder's time on their side, and they'll only spend it if it pays them: a share of each sale, written down before you ask.
• You need a feature that works on its own. If your product only makes sense as a whole, there's nothing to put inside someone else's.
• It takes weeks per partner, not an afternoon. Finding the product, the message, a preview, the terms, the build. The message is the part most founders get wrong: the integration partner outreach message.
• A host gives you a slice of its users, not thousands. A host with 300 users is 300 people who might see your feature. Worth it when they are exactly your buyers; not worth it as a numbers game.

When it's too early: during the first stage. If you can't describe your buyer in one sentence, you can't pick the product they live in. And if your product still changes every week, so will the integration.

Where Ordana fits: it's a free directory built for this channel. You list one feature of your product as a widget and set its price, its free uses and the share the host keeps (10–90%). Any startup can add a listed widget to its product with no application and no approval. Users sign up and pay inside the widget; Ordana is the merchant of record and splits each sale automatically, taking its 10% from your side. Both sides build with their own coding agent, and Ordana never reads either repository. It won't find your first 10 customers. It can be one of the things still bringing people in when you're at 60.

When should I stop doing things that don't scale?

Not all at once, and never the talking. Stop a manual task when one of three things is true:

1. It stopped teaching you anything. If the last five onboarding calls told you nothing the first five didn't, turn the call into a checklist and an email.
2. A written process could do most of it. Once you've done something the same way ten times, write the steps down. Then automate them or hand them over.
3. It's eating the hours your compounding work needs. Every hour of one-by-one messages in the third stage is an hour not spent on the page or the integration that would keep working.

Stop first: messaging strangers one at a time once a channel brings customers without you; posting in every community instead of the two that work; discounts for everyone; submitting to hundreds of directories.

Never stop: talking to customers. At 100, still do a few calls a month. The words they use keep feeding the pages and the pitches that bring the next hundred.

Frequently asked questions

If one feature of your product would make sense inside a product your buyers already use, list it on Ordana as a widget: free, no approval queue, and it keeps working after the first hundred.

Related reading:

• How to Get Customers for Your SaaS Without Ads — four ways to borrow someone else's distribution.
• How to Evaluate a Growth Channel — the scorecard to run before committing to one.
• The Best SaaS Directories to Submit Your Startup to in 2026 — where to list, and in what order.
• SaaS Distribution — every guide on getting customers without an ad budget.