Part IV: Get to the First Customer

Chapter 17

Distribution Is Not Posting

Busy Is Measurable

By the time a founder came to me for help, he had become extremely good at being visible. His posts were generating tens of thousands of impressions each month. He published product updates constantly, had built a respectable audience, and could usually find some encouraging number to inspect: more views, another comment, a few new followers, a spike in traffic.

Revenue after several months was effectively zero.

His conclusion was understandable. He needed more reach.

I thought he already had plenty of reach, at least in the ordinary meaning of the word. A large number of people were seeing him. The problem was that visible and commercially distributed had quietly become synonyms.

Promotional activity makes that mistake easy because it is wonderfully measurable. Publish something and a counter begins moving. Send email and the system reports deliveries, opens, clicks, and replies. Run an ad and you can watch impressions, visitors, and cost per click update with enough decimal places to make the operation feel unusually scientific. The platforms are very good at telling you that you used the platforms.

Those numbers are not useless. Earlier we established that an impression is not inherently a vanity metric, a signup is not inherently meaningless, and traffic can be essential depending on the business model. The problem is not that the number exists. The problem is giving the number a larger job than it earned.

Suppose you decide to publish five times a week, send one hundred outbound messages, comment thoughtfully on fifty relevant posts, and produce a newsletter. At the end of the week, you can determine with unusual confidence whether you completed the plan. Those tasks have clean completion states.

There is something deeply satisfying about that, especially for builders. Most of our productive work has accumulated toward outcomes we can see. Write the function. Fix the defect. Finish the integration. Deploy the release. Commercial work is less cooperative because another human being gets a vote. I can decide to send one hundred messages. I cannot decide that ten qualified buyers will care.

That asymmetry changes behavior more than we usually admit. The marketing plan naturally fills with things the founder can complete, because completion feels like progress and produces evidence of effort. A founder can finish a week exhausted and point to twenty posts, four hundred messages, twelve hundred website visits, two hundred new followers, forty replies, and three demos. Every number can be real. Every activity can have required genuine work. Revenue can remain exactly where it was.

This is not the same argument as saying all activity without immediate revenue is wasted. It is not. Commercial systems produce intermediate evidence long before payment, and some of that evidence is extremely useful. The narrower problem is mistaking evidence that promotion occurred for evidence that the business has a dependable route to customers.

Busy is psychologically persuasive because busy leaves receipts. It also offers a refuge from questions that are harder to answer. “Publish five posts this week” is a pleasant problem. You can solve it alone at 11:30 at night with coffee and poor judgment about when you should go to bed. “Find out why people who can buy this are not entering serious conversations” may require changing your target, asking for introductions, hearing no, discovering that your audience is wrong, or admitting that a channel with impressive numbers is not doing the commercial job you assigned to it.

One produces a checked box. The other may produce an inconvenient answer.

That is why consistency deserves the same scrutiny as intensity. Consistency is useful when you are repeating something that produces a result you understand or running a deliberate test long enough to learn from it. Consistency is less impressive when it means preserving an activity because stopping would force you to ask whether the activity is connected to the outcome.

The founder with the large audience had not fabricated his progress. The impressions were real. The comments were real. People were genuinely seeing his work. His mistake was subtler: he had promoted evidence of promotional activity into evidence that the business possessed a reliable route to customers.

Posting has built real businesses. So have cold outreach, search, advertising, communities, partnerships, marketplaces, events, and channels that did not exist when some of us started working. I am not arguing against any of them. I am arguing against treating the verb as the result.

“We post every day” is a production schedule. “We send 500 emails a week” is an activity level. “We received 80,000 impressions” is an exposure measurement. None of those statements, by itself, tells me whether the business has a dependable way to put a relevant offer in front of people who can meaningfully act on it.

Before recommending more of anything, I wanted to answer a stricter question: what, exactly, would have to be true before we could say this business had distribution?

Audience Is Not Buyer Access

Mostly, they were not.

The application solved a back-office problem for small professional-services firms. The founder’s public audience leaned heavily toward developers, AI enthusiasts, other founders, and people who enjoyed seeing new software get built. They understood why the technology was interesting. Many had no reason to experience the administrative problem, control a budget for solving it, or regularly advise the firms that did.

That was a perfectly respectable audience for discussing the technology. It was a weak audience for selling this particular product.

This distinction matters because the word audience carries more commercial weight than it deserves. Five thousand followers feels like an asset. Twenty thousand subscribers feels like a larger one. Tens of thousands of monthly views make it tempting to imagine a virtual room full of people waiting for the right offer.

An audience tells you who you can reach with some probability. It does not tell you why they are there, what problems they live with, what they buy, what authority they have, or whether your offer belongs anywhere near their current priorities.

Technical founders are particularly vulnerable to this because we naturally accumulate audiences made of people like us. Builders follow builders. AI practitioners follow AI practitioners. Designers attract designers. Founders spend a surprising amount of time discussing the experience of being founders with people currently having the same experience. Product screenshots, architecture diagrams, model comparisons, launch posts, and implementation details can perform extremely well with peers because peers understand immediately why those things are interesting.

That attention can be valuable. Peers share ideas, recommend tools, make introductions, create partnerships, offer feedback, amplify useful work, become customers, and help establish reputation. A person who will never buy the product can still be commercially important if their attention reliably connects you to people who might.

But that is the key word: connects.

Audience size and buyer access are not the same variable. I sometimes think about this as buyer density: among the people you can repeatedly reach, how many are buyers or credible bridges to buyers? The term does not need a formula. It is simply a reminder that 50,000 followers with almost no connection to the purchase can be commercially weaker than 1,200 people concentrated around the exact problem you solve.

Consider a founder selling a compliance application to regional healthcare organizations. Fifty thousand software followers could create substantial reputation and reach. A much smaller audience containing compliance leaders, operations people, advisers, and vendors already embedded in those organizations might create more first-customer opportunities. Neither audience is universally “better.” They are useful for different jobs.

This is where founders can accidentally optimize the wrong graph. Audience growth is visible. Buyer overlap usually is not. The platform can tell you that you gained 437 followers this month. It cannot reliably tell you that eleven manage the problem your application solves, three influence budget, two are currently dissatisfied with their workaround, and one is likely to evaluate software this quarter.

You have to learn that yourself.

There is a mistake in the other direction too. A founder with almost no public audience may assume distribution is impossible. But if that founder can reliably reach twenty relevant buyers each month through a narrow professional community, a partner, direct outreach, or an industry association, the business may have far more useful access than a creator with a six-figure following.

The point is not “followers do not matter.” The point is that their commercial meaning depends on who they are and what path their attention creates.

For the founder in our story, this changed the question. More impressions from the same audience composition were not obviously the missing ingredient. Instead of asking how to make the audience larger, we asked where the firms experiencing this back-office problem already looked for advice and how they encountered new ways of handling the work.

That moved us away from audience growth as the default answer and toward the actual routes into the market.

Researched Outreach: Why This Buyer, Why Now

Cold outreach is not disqualified because it is cold. It fails commercially when the sender has no defensible reason to believe the recipient should care. The difference between researched outreach and spray is not whether the first sentence mentions a podcast episode or the buyer’s hometown. It is whether the message is based on commercial relevance.

Research should help answer four questions before the message is sent:

Why this buyer? What role, workflow, responsibility, or existing behavior makes the problem plausibly theirs?

Why this problem? What evidence suggests the current process consumes money, labor, risk, delay, or executive attention?

Why now? Is there a public event or operating condition that could make the problem more live today than it was six months ago?

What is the smallest reasonable next commitment? Usually the first message should earn a conversation or a useful reply, not close a subscription from a stranger.

The "why now" question is where signal-driven outreach becomes useful. A hiring surge, a new compliance requirement, a migration, a product launch, a budget cycle, a visible workflow change, a contract renewal, or a public complaint can all create a reason to investigate timing. None of those signals proves the buyer has the problem. They are reasons to look more closely, not permission to invent urgency on the buyer’s behalf.

A good first message therefore does less than many automated sequences attempt. It names the relevant observation, connects it to a plausible consequence, and asks a question the buyer can answer without entering a sales process. If the reply shows the hypothesis was wrong, that is useful. If the buyer confirms the problem and accepts a conversation, that is stronger evidence. Either way, the message did a job beyond producing an open or a click.

Run small batches while the buyer definition is still moving. Twenty carefully chosen messages can teach more than two thousand automated ones if the smaller batch lets you see who responds, what they recognize, which assumptions are wrong, and what condition makes the problem live. Automation becomes useful after the pattern is credible. Before that, it can make an untested guess look impressively productive.

This is the commercial standard for outbound in this book: do not personalize for decoration. Research for relevance. Do not manufacture a trigger. Observe one. Do not ask a stranger to absorb the whole sales process in the first message. Ask for the next decision that can teach you something.

Channel-Market Fit

The consultant conversations did not prove that consultants were the right channel. They did something more useful: they produced enough relevant behavior to justify another test.

That is the standard I prefer early. A channel should earn additional investment by showing that it fits the way this market discovers, interprets, and evaluates this kind of solution.

Channels are often discussed as interchangeable traffic sources. Try LinkedIn. Try cold email. Try search. Try ads. Start a newsletter. Find affiliates. Build a community. Launch on a marketplace. The advice is not necessarily wrong, but it skips the interesting question: what kind of buying behavior is this channel good at supporting?

Search can be extraordinarily powerful when buyers already recognize a problem and go looking for a solution. Someone searching for “HIPAA-compliant appointment reminder software” has done a remarkable amount of commercial work before reaching a website. They know enough to describe what they want and are actively looking.

Now imagine a product that solves a costly problem buyers do not yet recognize as a software problem. Search may be much weaker, not because search is ineffective but because nobody is searching for the category you invented on Tuesday. The channel and the market are out of sequence.

Cold outreach has a different shape. It lets you choose who sees the message instead of waiting for them to find you. That can be useful when the buyer population is identifiable and the value of a customer justifies individual contact. But the recipient was not looking for you. The problem has to become legible quickly enough that an interruption feels worth continuing.

Content has another shape. It can demonstrate expertise, explain a problem, create repeated exposure, show the product in context, and attract people who are already interested in the topic. When the audience contains the right buyers or credible bridges to them, content can absolutely be distribution. When it mostly attracts peers with no commercial connection to the problem, it may be valuable for other reasons without functioning as a strong customer path.

Paid media, marketplaces, communities, events, affiliates, and partner channels each create their own conditions. They differ in who is present, what those people are doing there, how much explanation fits naturally, what kinds of actions feel normal, and how easy it is to observe what happens next. The channel is not good or bad in the abstract. It has a shape. The market has a shape too.

Good distribution appears when those shapes fit well enough that useful commercial behavior can occur.

This is why I am suspicious of channel advice detached from the product being sold. “We should be doing outbound” tells me almost nothing until I know whether the buyer can be identified, whether the problem is recognizable in an unsolicited message, and what next action would make sense. “We need to grow on social” is similarly incomplete until we know which people, why that environment is relevant, and what commercial behavior we expect exposure to make more likely.

The founder with the back-office product learned this in the conversations that followed the consultant introductions. Publicly, he had tended to demonstrate the application as an AI automation system. That framing made sense to the audience he had accumulated. Other builders understood immediately why the AI was interesting.

The prospective customers cared about something more concrete. The demonstration shifted toward the administrative process itself, which the founder said could be reduced from several hours to roughly twenty minutes.

The lesson is not that every technical product should hide its technology or convert every claim into a time-saving promise. It is that a better-matched channel can expose what the market actually responds to. The route did not merely deliver more people. It delivered more informative reactions from people closer to the purchase.

That makes early channel selection a learning decision as much as an acquisition decision. At this stage, I do not need a channel capable of producing one million customers. I need one capable of producing enough relevant encounters that I can judge whether it deserves more investment.

If buyers in a narrow industry rely on specialist advisers, those advisers deserve investigation. If buyers search aggressively for the problem, search deserves investigation. If the population is small and identifiable, direct outreach may teach more than broadcasting. If the product is naturally discovered inside another workflow or marketplace, that context may matter more than building an audience from scratch.

These are hypotheses, not universal prescriptions. A useful channel earns confidence when it repeatedly puts the offer into relevant commercial situations and produces behavior that moves closer to evaluation, commitment, or payment.

Once that begins happening, volume becomes much more interesting.

Distribution as a System

A distribution system is not a list of channels. It is a set of connected activities in which each activity has a commercial job and the connections between them make buyer movement possible.

Founders often accumulate channels one at a time without deciding what role each is supposed to play. There is a LinkedIn account because founders are supposed to post. A newsletter because owning an email list sounds prudent. Cold outreach because somebody said outbound works. Search content because search compounds. A marketplace listing because it was available. Perhaps some paid traffic because the organic numbers are taking too long to become emotionally satisfying.

After a while, the business has a surprising amount of marketing infrastructure and still no coherent route to a customer.

A system begins when the pieces connect. A founder might publish useful material that reaches people in the target industry. Some join an email list or visit a page describing the problem in language they recognize. A smaller number enter a conversation or evaluation. Those conversations reveal objections, terminology, timing, and use cases that improve future content and outreach. Now the activities are doing different jobs inside the same commercial process.

The post does not have to make the sale. The website does not have to make the sale. The newsletter does not have to make the sale. The conversation does not have to begin from zero. What matters is that the pieces create a route in which the next commercially useful action can occur.

This is also why simplistic channel attribution can distort how founders think. A buyer may see a post, later hear the company mentioned by a colleague, search for it, read the website, and finally respond to a direct message. Software will probably assign credit somewhere. The buyer experienced a sequence.

That does not make measurement pointless. It means the measurement should match the job. Content may create relevant recognition. Search may capture existing intent. Outbound may initiate a specific conversation. A partner may create an introduction. A website may help a visitor determine whether the product applies to them. An evaluation may expose whether the promised outcome survives contact with the buyer’s environment.

The useful question is not “Which one of these is our growth channel?” It is “What job is each activity performing, and does the next useful thing actually happen?”

That is where many impressive-looking marketing systems fail. Awareness is created but never connects to evaluation. A newsletter earns attention but never gives the right reader a reason to examine the product. Outbound generates replies that end with “interesting.” Qualified visitors arrive at a website but encounter no sensible next action. A partner makes introductions and nobody follows up while the problem is active.

The components can all be alive while the system is not.

Builders should recognize the pattern. It is the commercial equivalent of several services running correctly while the integration between them quietly fails. Debugging the components harder will not repair a missing connection.

This is where I would audit the current marketing plan. Take every recurring activity that consumes meaningful time or money and assign it a commercial job. For each activity, identify five things:

1. Target buyer: Who is this intended to reach, including credible bridges to the buyer?
2. Offer: What problem, outcome, or proposition are they supposed to encounter?
3. Trust mechanism: What gives the person enough confidence to consider the next step? Do not over-answer this yet. The next chapter will deal with it directly.
4. Intended commitment: What meaningful action should become possible next?
5. Evidence produced: What behavior would justify continuing, changing, or stopping this path?

The point is not to create a prettier marketing spreadsheet. It is to make each activity accountable to the commercial reason it exists.

Suppose your newsletter is intended to educate operators in a narrow industry before a direct sales conversation. It may be doing its job even if the “newsletter revenue” field in an attribution report remains empty. But you should be able to observe something consistent with that job: the right people subscribe, relevant readers reply, sales conversations begin with more context, or prospects repeatedly mention the material during evaluation.

Now suppose the newsletter is explicitly intended to generate qualified opportunities, but nearly every subscriber is another founder. It may still be a valuable publication. You simply should not let subscriber growth stand in for evidence that the intended acquisition path is working.

The same discipline applies to outbound, search, paid media, partnerships, communities, and marketplaces. If you cannot explain why an activity exists, that is useful information. If you can explain the job but cannot identify evidence that it is performing the job, that is useful too. The activity may need more time, a different target, a different message, or a different role. It may also need to stop.

A distribution system does not need to be complicated. A founder can identify appropriate companies, contact a small number each week, turn some into conversations, learn which situations produce serious evaluations, and occasionally receive introductions from customers or advisers. That system might fit in a spreadsheet. It is still a system if another relevant opportunity can enter it and the founder can learn from what happens.

The founder in our example was beginning to build something like that. His public audience still existed, but it no longer carried the entire commercial burden. Consultant relationships created another route into the market. Conversations with introduced firms changed how he demonstrated the product. Those responses improved what he could say in later conversations and public communication.

The activities were becoming connected to buyer behavior.

That is the point at which distribution becomes inspectable. You can see which people enter through which routes, what happens next, where movement stops, and which paths deserve another attempt. You stop asking how much marketing you managed to produce and start asking what the system moved closer to a customer.