Getting customers · 6 min read
How to get your first SaaS customers
Your first ten customers are found manually, one at a time, by turning up where the problem is being discussed and being useful. Treat it as research with a payment link attached, not as marketing.
Ten is an awkward number. Too few to be a business, too many to get by accident. Almost everything written about SaaS growth assumes you already have them.
Getting to ten is a different job from getting to a hundred, and using the hundred-playbook is why most people never reach ten.
You’re not doing marketing yet
Marketing is what you do once you know who the customer is, what they respond to, and roughly what one is worth. You know none of that. What you’re doing is closer to research with a payment link attached, and the payment link matters, because money is the only honest signal you’ll get.
Your first ten customers aren’t revenue. They’re the answer to what you should build and say next.
1. Write down the trigger
Not the persona. The moment. “We broke prod with another migration.” “The spreadsheet stopped coping.” “The person who did this has left.”
Each one is a moment when somebody starts looking. Three of those is enough to work with. If you can’t write three, that’s the week’s finding, and the next step is conversations rather than promotion.
2. Go where the complaint already exists
Not where your audience “is” in the abstract. Where the specific complaint gets posted: subreddits, Discord servers, niche forums, the comments under a competitor’s launch, replies to someone venting about the problem.
Search the words people use when they’re annoyed, not the words you use when you’re selling. “How do you all handle…”, “anyone found something better than…”, “still doing this in a spreadsheet…”
3. Email twenty people individually
Not a sequence. Not a tool. Twenty emails you wrote by hand to people you have a specific reason to contact. They posted about the problem, they run the kind of operation you built this for, they left a review complaining about the incumbent.
Each should reference that reason in the first line. If you can’t write that line, they’re not on the list. Twenty good emails beat two hundred templated ones and won’t get your domain flagged.
Where the twenty come from
- People who posted the complaint. The best source, and the reason step two comes before this one.
- Reviewers of the incumbent. Anybody who left a two-star review of the tool you replace has told you their objection in writing, in public, for free.
- People who answered somebody else’s question about this problem. They care enough to help strangers with it, which makes them unusually willing to talk.
- Job adverts describing the work. A company hiring somebody to do the thing manually has confirmed both the problem and a budget.
- Your own history. Former colleagues who had this problem. Awkward, and it is where a meaningful share of first customers actually come from.
What the email actually says
Four lines, and the order matters more than the wording.
- Why you, specifically. Reference the thing they wrote or did. One clause, no flattery.
- What you built, in one sentence, in the words they used rather than yours.
- An honest admission. It is early, you have a handful of users, you are looking for people with this problem specifically. This raises reply rates rather than lowering them.
- A small ask. Fifteen minutes, or a straight question you actually want answered. Not a demo, not a call to “explore synergies”, and no calendar link in the first message.
Expect two or three replies from twenty. That is a normal, working result, and the failure mode is concluding from four sent emails that this does not work.
Charge them
The temptation at ten is to hand out free accounts to get the logos. Resist it. A free user tells you the product is interesting; a paying one tells you it’s worth money, and those are completely different facts. You need the second one.
What to charge them
Something, immediately, and less than you eventually will. A price near the bottom of your intended range is honest about the state of the product and still produces the only signal you need, which is that money moved.
Two things to avoid. Do not build a discount into the product with no end date, because you are creating a tier you will have to migrate people off later. And do not hide the price to avoid the conversation, because the conversation is the research.
A founding price you keep for as long as somebody stays subscribed is a promise you can actually honour. “Locked forever” usually is not, and early customers remember which one you said.
4. Ask the ones who say no
People who look and don’t buy are the most valuable people you’ll meet this month, and almost nobody asks them anything.
One question: “What were you hoping this would do that it didn’t?” Don’t defend, don’t explain, don’t demo. Collect the answer.
Three of those answers will usually rewrite your homepage for you, in their words, which is the only version that reliably works.
5. Do it again with what you learned
The loop is small and unglamorous: find people with the trigger, be useful where they are, watch what happens, adjust the words. Ten customers turn up somewhere in the middle of doing that for a few weeks.
How long ten takes
Longer than you want and shorter than it feels. A realistic shape, at a few hours a week: two to three weeks before the first one, and somewhere between two and four months to ten.
The distribution is lumpy in a way that is worth expecting. The first customer takes the longest and teaches you the most. Customers four through seven often arrive close together, once the sentence you use to describe the product has been corrected by the first three. If you quit, it will be in week three, before the correction has happened.
The first ten are not a funnel. They are ten separate acts of persuasion, and each one makes the next easier.
What to skip until you’ve got ten
- A content calendar
- A launch
- Paid acquisition
- A referral programme
- Anything described as a strategy
- Another feature
Every one of those multiplies something you don’t have yet.
Then what?
At ten, you’ve got positioning in your customers’ own words and some idea which trigger converts. That’s the point at which picking a single channel becomes a sensible thing to do, marketing for SaaS founders covers what comes next.
If you already have users and the line is flat, reading a flat line is the more useful page, because the diagnosis there starts with retention rather than acquisition. And if what you are weighing up is paying somebody to do this instead, the alternatives are worth pricing honestly.