First 100 customers without paid ads
The exact sequence I use to get a new product from zero to a hundred paying customers using channels you own.
Paid acquisition is the most expensive way to learn what your customers want. It buys you volume before you have earned understanding, and it hides the fact that your positioning is wrong by paying to put it in front of more people.
The first hundred customers are not a growth problem. They are a learning problem. You want them acquired through channels where you talk to people directly, because every conversation is data you cannot buy.
This is the sequence. It works for SaaS, AI products, marketplaces and vertical software. It does not work if you skip steps.
Step 1: Narrow until it hurts
Most founders describe their customer in a way that includes everyone and therefore reaches no one. Narrow the definition until you can name twenty real companies or people who fit it. If you cannot list twenty by name, the segment is still too abstract.
- ◆Write the segment as a sentence with a role, a context and a trigger: 'operations leads at 20 to 100 person logistics firms who just lost a spreadsheet owner'
- ◆List twenty named targets in a spreadsheet, with the trigger event next to each
- ◆Find where those twenty already gather: a Slack group, a subreddit, a conference, a LinkedIn hashtag, a WhatsApp community
- ◆If you cannot find where they gather, you have picked a segment you cannot reach cheaply. Pick again.
A narrow segment you can reach beats a large segment you cannot.
Step 2: Write the offer before you write the code
An offer is not a feature list. It is a promise about an outcome, a price, and a reason to act now. Write it as one paragraph and read it to five people in your segment. If they ask a clarifying question about what it does, rewrite it.
- 01The pain, in their wordsUse the phrasing you heard in calls, not the phrasing from your deck.
- 02The outcome, quantifiedHours saved, revenue recovered, risk removed. Put a number on it even if it is a range.
- 03The proofEarly on this is a demo, a teardown, or a pilot result. It does not have to be a logo wall.
- 04The price and the askSay the number. Founders who hide pricing get meetings and no customers.
Step 3: Founder-led sales for the first twenty
The first twenty customers should be closed by you, in conversation, one at a time. This is not inefficient. It is the research budget.
- ◆Send 10 personalised messages a day, five days a week. Personalised means you reference something specific about them.
- ◆Lead with the observation, not the pitch: what you noticed about their business and why it made you think of them
- ◆Ask for 15 minutes, not 30. Fewer people say no to 15.
- ◆Record every call with permission, and keep a single document of exact quotes
- ◆Offer a paid pilot, not a free trial. Free tells you nothing about willingness to pay.
- 50 personalised outreach messages sent
- 5 conversations held
- Quote document updated with new language
- One change made to the offer based on what you heard
- Pipeline reviewed: who is stuck and why
Step 4: Start the search engine early
Search compounds slowly, which is exactly why you start it in month one rather than month twelve. The goal is not traffic. The goal is to be the answer when someone in your segment types the problem into a search box.
- 01Map intent, not keywordsGroup searches by what the person is trying to do: understand, compare, choose, implement.
- 02Build one page per intentEach page answers one question completely, better than anything currently ranking.
- 03Publish the boring pages firstComparisons, pricing explainers, alternatives, how-to guides. These convert. Thought leadership does not.
- 04Link them togetherEvery page points to the next logical page and to the thing you sell.
| Page type | What it does | Typical time to rank |
|---|---|---|
| How-to guide | Brings in people with the problem | 2 to 4 months |
| Comparison / alternatives | Catches people already shopping | 1 to 3 months |
| Pricing explainer | Filters and pre-qualifies | 1 to 2 months |
| Free tool or calculator | Earns links and repeat visits | 3 to 6 months |
Step 5: Be useful in the rooms they already sit in
Communities punish selling and reward usefulness. The trade is simple: answer questions properly for weeks before you ever mention what you built.
- ◆Pick three communities, not ten. Depth beats spread.
- ◆Answer with the full answer, including the part that does not need your product
- ◆Publish teardowns and breakdowns of real problems in public
- ◆When you do mention the product, mention it once, in context, with a caveat about who it is not for
Step 6: Borrow other people's audiences
Someone already has the trust of your segment and does not compete with you. Agencies, consultants, adjacent tools, newsletter writers, community operators. Give them a reason to send people your way.
- ◆Offer a revenue share, a co-built asset, or a service they can resell
- ◆Make the referral trivially easy: a link, a one-paragraph blurb, a landing page with their name on it
- ◆Start with five partners and actually service them, rather than signing thirty who do nothing
Step 7: Close the loop and make it repeatable
Around customer thirty you will see one channel outperforming the rest. That is the signal to stop spreading and start compounding.
- You can name the channel that produced the last ten customers
- You know the cost, in hours or dollars, of one customer from that channel
- Customers activate within a week and are still using it a month later
- You can write the sales conversation down as a script someone else could follow
- Churn in the first 60 days is understood, not just measured
Only when all five are true does paid acquisition make sense, because now you are buying volume for a machine that works instead of paying to discover that it does not.