first-users

/first-users exists to get you actual users, not a strategy document about getting them. It maps where your audience really spends its day, builds a list of named people you can reach right now, and keeps you on the manual, unscalable moves that land the first ten. It’s also the honest answer to “I need a marketing plan.”

​

When to reach for it

• “How do I get users / customers / traction?”
• “I need a marketing plan / sales plan / go-to-market strategy”
• Nobody’s signing up and you don’t know where to find people
• “Which acquisition channel should I try next?”
• You tried something and want to record what it taught you

​

”I need a marketing plan”

The question gets taken seriously, then narrowed — in a sentence or two, not a lecture. A marketing plan is the right thing to build once you know what resonates and can repeat it; before that it tends to be a document that reads well and changes nothing. What works at this stage is narrower and more concrete: figure out exactly where these people spend their time, then go get the first handful by hand.

​

Where users come from changes with how many you have

This staged frame is the spine of the skill — and the reason it will refuse to hand you tooling early. The count is a default, not a verdict. Eight customers who all came from one former employer, with no read yet on what resonates, is still stage 4–10 work — and you’ll be told so.

​

Building your audience map

The map starts with the buyer-day audit, deliberately before any channel talk. Most founders pick the channel that’s easiest to run from a laptop, which works fine when the buyer also lives on a laptop — and fails badly for buyers on job sites, in clinics, driving, on the shop floor, or in a school office. Their inbox is not where their work happens, and getting this wrong costs months of refining subject lines for people who were never going to read email. One question at a time, the conversation works through the buyer’s actual day: how central email is, which conferences and meetups they attend, which subreddits, Facebook groups, Discord servers or industry forums they’re in, whether they answer unknown numbers, and how they normally hear about new tools. If you can’t answer most of it, that’s named directly — and treated as the validation work itself, not a detour. Then channels, with the ruled-out ones kept on purpose: if the audit established this buyer never reads email, cold email stays in the table marked ruled out, with the reason. Knowing a channel doesn’t reach your buyer is a finding, not a blank — and it stops you relitigating it in three weeks. Then the part that makes the file worth having: a prospect list of 10–20 named people you can reach right now. A beautiful channel analysis with no names on it has failed. The list comes out of thinking aloud — who you follow, who you’ve worked with, who you know from communities and events, who has complained about this problem in public, and who those people could introduce you to. Nothing gets marked *(verified)* until real people in the segment confirm it. Web research and your own confident belief about the industry are both assumptions until then — which matters later, when a channel underperforms and the question becomes “did we ever actually check this?”

​

One list, shared across skills

/interviews, /surveys, and /mvp all read the same map — its channels for distribution, its prospect list for people to reach. Each row’s Ask column records what you currently want from that person (interview, survey, or user), so someone interviewed today can become a user later without a second list existing. Reach gets worked out once instead of being re-answered every time a skill needs an audience.

​

The plays

Tactics are chosen for your stage, your audience, and what your log says you’ve already tried — never presented as a catalogue, and never repeated back at you after they’ve failed. Stage 1–3 — the warm network. Almost without exception the first two or three customers come from someone you already had a connection to. Not because it’s a sales trick: nobody buys an unproven product from an unknown company on its merits, because there aren’t enough merits to evaluate yet. They buy because they trust the person building it. So you work the sources in order — personal network, then second-degree introductions, then anyone you’ve already interviewed (they gave up their time on this problem once already, and are usually the most obvious untapped source in the map). Second-degree connections get pushed hardest, because they’re the most systematically skipped: asking someone for an introduction is socially uncomfortable in a way that configuring a tool is not. And the ask is made forwardable — name the specific person, one concrete line on why them, and write the text the connector can just paste. Stage 4–10 — the unscalable plays. Options, not a sequence; chosen by who your buyer is:

• Show up in person — the most consistently surprising finding, and the one almost nobody wants to act on. For the right audience, three days in the right room beats months of email.
• Small industry events — small beats large, because the goal is conversations, not badge scans. The playbook is operational: a booking link with short back-to-back slots, the attendee list emailed before and during, and the floor day stacked end to end.
• Micro events — a dinner or happy hour for six to ten people who match the profile. Once someone has eaten a meal with you they don’t ignore your follow-up.
• Go where the pain is already public — find the threads where your audience is already complaining about exactly this, and show up as a participant rather than a marketer. Expect harsh comments and quiet signups to arrive together, and expect it to be close to a full-time activity for a while.
• Non-pitch framings — advice, mentorship, a product review, a working session, or paying for a slice of someone’s professional time. These only work, and are only acceptable, if you genuinely want the input and will use it; a sales call disguised as a request for advice burns a small industry’s goodwill fast.
• Give something specific first — twenty minutes of real work on that one prospect’s actual situation, then ask for the meeting. It doesn’t scale, which is the point; you need ten customers, not a thousand.

Whatever the play, the outreach rules are the same: short, one unmistakable ask, read it aloud before sending (if you wouldn’t say a line out loud, rewrite it), and follow up a few times over a couple of weeks — most founders send once and conclude the channel is dead.

​

The outbound gate

When lead databases, enrichment tools, or email sequencers come up early, you don’t just get a no — you get the list of what has to be true first: a pitch that has already converted by hand more than once, roughly ten to twenty customers so you know which segment responds, and a clear read on what actually resonates so there’s a message worth sending at volume. Before that, volume tooling scales a message you haven’t validated and burns the list learning what a dozen conversations would have taught you faster.

​

Reporting what happened

Come back after trying something and the numbers get kept: “emailed 40, 3 replied, 1 call” is useful a month from now; “outreach went poorly” is not. Prospect and channel statuses are updated, new names that surfaced get added, and users_count moves only if real users arrived. Anything that bears on a bet — someone paid, refused to pay, or confirmed the problem — routes into your hypothesis scorecard. Acquisition attempts are evidence too. Each pass closes on one concrete move with a person attached. “Do more outreach” isn’t a move; “email J. and M. tomorrow with the two-line ask we just drafted” is.

​

What it writes

The map is created when you first need it, not at project setup — an empty map on day zero is noise.

​

Good to know

• users_count means people actually using the thing. Not signups, not waitlist entries, not “this sounds interesting” replies, not a friend who agreed to try it and never opened it. Expect pushback on an inflated number: every routing decision here depends on it, and a wrong count sends you to tactics that waste weeks.
• The network false-positive. Your first customers buying partly because they trust you is exactly the right way to start — and it isn’t yet evidence that strangers will buy. The signal to watch for is unprompted demand from someone outside your network.
• Channels that didn’t work keep their row, marked ruled out. The map is a record of what you’ve learned about reach, not just what’s still open.
• The prospect table is right for up to ~50 people. Past that it graduates to one file per person under startup/prospects/; past a few hundred you need a real tool, and the table can be dumped to CSV so nothing is lost. If your list is growing that fast, you’ve outgrown this problem — a good sign, not a limitation.

​

Related