Every company's growth chart begins at the same number: one. Before the flywheel, before the funnel, before the dashboard with the up-and-to-the-right curve, there is a single person who used a product for the first time and decided it was worth their attention.

Most founders treat that first user as a milestone. We treat them as a dataset.

The most expensive user you will ever acquire

The first user is, on a unit-economics basis, the worst customer a company will ever have. No referral brought them. No content ranked for them. No creator vouched for them. Every acquisition system that eventually makes growth cheap : word of mouth, network effects, brand : does not yet exist. The cost of acquiring user number one is effectively the entire cost of figuring out who your users are.

This is why the first user matters more than the first thousand. The first thousand tell you how well your machine runs. The first one tells you whether you built the right machine at all.

The mistake most teams make is assuming they already know the answer. They build for an imagined customer : usually someone who looks like themselves : and then spend their entire marketing budget confirming the assumption rather than testing it.

The obvious user is a hypothesis, not a fact

When we worked with Gamma, an AI presentation tool, the obvious user was clear: startup founders and business professionals making pitch decks. That was the assumed ICP. It was plausible, it was intuitive, and it was incomplete.

Instead of scaling against the assumption, we tested it. Fourteen ICPs across nine markets, each treated as a falsifiable hypothesis with its own creators, narratives, and conversion benchmarks. The result surprised everyone: educators and trainers emerged among the strongest power users : a segment that produced a 61% activation rate, lifted retention by 37%, and cut acquisition costs by 28%.

None of that value was created by spending more. It was created by discovering who to spend on. The first user, correctly identified, changes the economics of every user after them.

Three principles fall out of this:

First, ICPs are discovered, not declared. The segment that loves your product is frequently not the segment you built it for. Excel was built for accountants and became infrastructure for everyone. Discord was built for gamers and became the internet's community layer. The founders didn't choose those users. The users chose the product : and the companies that won were the ones who noticed fastest.

Second, discovery has a method. Founders often describe finding early users as luck: a tweet that landed, a Reddit thread that took off. But luck is just an experiment you didn't design. Structured ICP testing : mapping candidate segments, sizing them, matching them with creators their community already trusts, and measuring activation rather than impressions : converts the same process into something repeatable. What looks like serendipity is usually an unrun experiment.

Third, the first user defines the narrative for the next million. How your earliest users describe your product becomes the language every future campaign borrows. When the first user is chosen well, the story writes itself. When chosen badly, companies spend years marketing against their own positioning.

From first user to first thousand

Once the right first user exists, the question changes from who to how fast. Here the sequence matters:

  1. Map before you spend. Identify and size candidate segments before selecting a single channel or creator. Distribution begins with discovery, not with budget.
  2. Test narratives, not just audiences. The same product framed as "save time" versus "look brilliant in the meeting" will find entirely different first users. Narrative is a targeting variable.
  3. Let activation, not reach, pick the winner. A segment that watches is worth less than a segment that signs up, returns, and builds. Optimize for the behavior that predicts revenue.
  4. Scale only what converts. Concentrate spend on validated ICP-creator-narrative combinations. Scaling an untested assumption just makes the mistake bigger, faster.

The uncomfortable conclusion

The reason first users are hard is not that reaching people is hard. Reaching people has never been easier : creators, platforms, and paid channels can put a product in front of millions in a week. The hard part is knowing which people are worth reaching.

That knowledge doesn't come from intuition, and it doesn't come from volume. It comes from treating user acquisition the way engineers treat systems: hypothesize, test, measure, iterate. The companies that internalize this stop asking "how do we get more users?" and start asking the better question : "have we found the right first one yet?"

Great products are built every day. Few find the people they're meant for. The first user is where that gap closes : or doesn't.