Distribution fundamentals: work out what your route to a customer needs to do

You can be all over every single platform for a month, and still have no idea where your next customer will come from. A launch post gets seen, a few people visit, somebody says the product looks great, then the trail disappears.

All of that can create attention, and that attention isn't useless. But you still need a route that helps the right person find the product, understand why it applies to them, trust you enough to take the next step, pay, and receive the result you promised. Once you can repeat that route, you have distributionThe route that helps the right customer find the product, understand and trust the offer, take the next step, pay, and receive the result.. Amazing!

Before you compare Reddit, email, search, partnerships, marketplaces, paid ads, or another place somebody has told you is DEFINITELY where you should be, you need to know what the route must produce and what your business can afford.

A platform is one part of the route

Imagine somebody asks a question about the problem you solve in a relevant subreddit. You answer it, they visit your website, they read a customer example, then they contact you and buy.

• Reddit is the distribution channelA place or route through which you can reach customers, such as email, search, a marketplace, a community, a partner, or an event.: the place where you reached them.
• Answering the question is a tacticA specific action taken within a distribution channel, such as answering a question, sending a personal message, sponsoring an email, or publishing a comparison page.: the specific action you took there.
• The complete route includes the question, answer, website, proof, conversation, offer, payment, and what happened next.

Naming the channel leaves most of what happens on it in limbo. “We'll use LinkedIn” doesn't explain who you'll contact, what they'll see, why they'll care, where they'll go next, or how they can buy.

Decide what you need the route to produce

“Get customers” is too vague to guide the work. The result you need now depends on what you already know:

A route used for interviews should be judged by the number of suitable strangers who agree to talk. A route meant to produce revenue needs more data points; impressions, engagements, followers, compliments, and free signups can't fill in for it.

One customer, all the way through

Start with the moment the problem becomes urgent enough for somebody to act, then review what happened before they made the decision to pay (you'll recognize these questions from other chapters, because they help you understand why your customer chooses to spend money with you):

• Where are they when the problem comes up?
• What makes them recognize that your product applies to them?
• What proof helps them trust the promise?
• What can they reasonably do next: reply, try, book, buy, or ask for approval?
• Who pays when the person using the product can't approve the purchase?
• What help do they need before they receive the result?

Somebody buying a $19 tool for themselves may move from a search result to checkout in ten minutes. A team changing the system that holds its customer data may need a demonstration, security answers, a proposal, multiple iterations of a contract, and approval from several people. The price, risk, customer, and product set limits on the routes that can work.

Brian Balfour's Four FitsBrian Balfour's framework connecting the market, product, way the business charges, and distribution channel. A change to one can make the others harder or easier to support. framework connects the market, product, way the business charges, and channel. For this chapter, we look at this part of the lesson: a low-priced product needs a cheap route, while a time-consuming sale needs enough money behind it to pay for the work. (Brian Balfour, accessed September 2026)

What can one customer cost you?

A route can produce sales and still leave too little money to provide the product. Keep the cash you spend and the hours you spend beside each other:

The cash column gives you a working customer acquisition cost (CAC)The sales and marketing cost divided by the number of new paying customers gained during the same period. This chapter records cash CAC and your hours separately.: money spent during the test divided by new paying customers. Formal CAC calculations often include salaries and overhead; this worksheet keeps your hours separate so you can see which cost is causing the problem. (Shopify, July 2024)

If a sale leaves $40 after direct delivery costs, spending $50 to gain that customer has already used more than the available gross profitThe money left from a sale after subtracting the direct cost of providing what was sold.. For a subscription, use the profit from the first sale or a conservative first year until you've observed renewals and cancellations; lifetime value (LTV)The gross profit or revenue a customer produces across their relationship with the business, depending on the calculation used. The method and cost basis should always be stated. becomes more credible once you've seen how long customers stay.

You can also record how many months of profit it takes to recover the cash you spent - the payback periodThe time it takes for the profit from a customer to recover the money spent gaining them.. A shorter period returns cash to the business sooner, although there isn't one correct period for every business. (Stripe, accessed September 2026)

Free can be extraordinarily expensive

Posting, attending communities, asking for introductions, and sending personal messages may cost $0 in media spend and consume most of your week.

One customer produced by forty hours of work may teach you enough to change the product, offer, proof, and message. Ten customers requiring the same work would leave no time to serve them.

Keep the hours visible. Some work will become faster with practice, proof, a template, or a product change; some should remain personal because the price and buying decision support it.

Existing access gives you somewhere to begin

You may already have an audience, customer list, warm network, marketplace presence, community reputation, or partner relationship. Record who you can reach through it and whether they match the customer you chose earlier.

Access doesn't establish demand. In an October 2024 Hacker News discussion, one participant said an existing email list made a second product profitable immediately; the same list later sent 10,000 free users and no paying users to a third. One anonymous account can't tell you what your list will do, although it shows why reach and payment need separate rows. (Hacker News, October 2024)

The route changes how the offer feels

Customers judge the setting alongside what you say. A recommendation carries some of the sender's trust; a specific community answer can show that you understand the problem; a search result meets somebody who's already looking.

An unsolicited message needs a credible reason to arrive now. A live demonstration can reassure somebody making a complicated or risky purchase; forcing a call before a $19 purchase creates work the customer didn't expect.

Your brief will record where the customer expects to encounter products like yours, whose proof they trust, and how much effort the value and price justify.

Start with one thing that doesn't scale

Early customers often need personal work. You might approach people one at a time, answer their questions yourself, set up the first accounts, or complete part of the service by hand while you learn what needs automating.

Paul Graham described this as doing things that don't scale: recruiting early users personally lets you follow every decision closely. His essay is from 2013 and many examples are venture-backed software companies, which limits how far we should stretch them, but the premise holds and allows for quick iteration on what works (and what doesn't). (Paul Graham, July 2013)

Compare several routes on paper, then test and improve one manual action before automating it and moving to another. Six channels create six sets of unfinished work and zero idea of which one caused the result. Congrats! You've diversified your confusion.

When the route should be followed

A route begins to look repeatable when you can reconstruct the same steps across several customers (yep, it's these points again):

• Where they came from;
• What made the problem urgent;
• Which message or action got their attention;
• What they needed to believe;
• Which objection delayed or stopped them;
• What they bought and paid;
• How much cash and time the route took;
• Whether they used, returned, renewed, or referred somebody.

There isn't a universal pass mark. Three customers from three personal messages tells you something very different from three customers after 400 messages, twelve demos, and six weeks.

Example: reaching independent dog groomers

Imagine you're selling the booking platform from the Pricing chapter. It takes a customer's deposit when they book with an independent dog groomer; you're testing a $39 monthly subscription against a $1 charge for every confirmed booking.

At 50 bookings a month, the pricing example left $26 in monthly gross profit under the subscription and $37 under the per-booking price. Neither can support an hour-long sales call forever, although those calls may help you learn from the first few groomers.

Your first manual test could be five personal messages to groomers who've publicly discussed no-shows or deposit problems. The job is to produce a paid account at the named price. After five messages, you record replies, refusals, payment, cash spent, your hours, the questions asked before buying, and what happened during setup.

The review records whether the customer was reachable, whether the offer led to payment, where the route stopped, and what another attempt would require.

Your distribution brief

When these answers are recorded, you know what a possible channel has to do and what it has to fit around. Choosing a distribution channel uses the brief to compare the routes available to you.

Prompt: build my distribution brief

The prompt below can go into the model you use, along with your customer, positioning, pricing, sales, and early-use records. “I don't know yet” is an acceptable answer; a blank shouldn't become an invented customer fact.

I need to work out what my route to paying customers must do before I compare
distribution channels.

Purpose
Turn the information I provide into one distribution brief. Keep facts,
assumptions, missing information, and recommendations separate. Explain an
ordinary decision before introducing a commercial term.

Where I am
[idea / prototype / lightweight MVP / pre-launch / live]

What I know
- Product or service: [what I'm selling now]
- Customer group: [who I'm concentrating on]
- User, buyer, and approver: [same person / different people / I don't know]
- Problem and moment it becomes urgent: [customer evidence]
- Result I promise: [result supported by my research]
- Positioning: [my current one-sentence explanation]
- Price and charging model: [amount and what creates the charge]
- Gross profit from the first sale or conservative first year: [amount / I
  don't know]
- Action I need now: [conversation / paid pilot / deposit / purchase / repeat
  purchase / renewal / referral]
- What has to happen before payment: [steps I know about]
- Proof I have: [results, demonstration, references, reviews, or none]
- Existing access: [audience, list, network, community, marketplace, partners,
  or none]
- Cash I can spend on the test: [amount]
- Hours I can spend each week: [number]
- How quickly I need an answer: [date or period]
- Routes I've already tried: [action, customer, result, cash, hours, date]
- Retention or repeat-purchase evidence: [evidence / not enough time yet / none]

Your task
1. Extract only the facts I've supplied. Keep source names and dates where I
   provide them.
2. List my assumptions and missing information. Don't fill either with likely
   answers.
3. State one immediate result the route needs to produce. Use the customer
   action I've named.
4. Map the known steps from the moment the problem becomes urgent through first
   contact, recognition, proof, next action, payment, setup, and customer result.
   Mark every missing step.
5. Check whether my product, price, customer, buying process, and proposed route
   can fit together. Explain any mismatch with the amount, time, or action it
   refers to.
6. Keep cash acquisition cost and my hours separate. Calculate cash CAC only
   when I've supplied spending and new paying-customer figures. Show the
   formula.
7. Compare the cash CAC with gross profit from the first sale or conservative
   first year. Don't estimate lifetime value without observed retention or
   repeat-purchase data.
8. Describe what the customer may expect, trust, fear, and do next in the
   setting I've proposed. Use my evidence and label any inference.
9. Recommend one manual action I can test before trying another route. It should
   fit my weekly time, cash limit, customer, and immediate result.
10. Set a review point. Include what I'll record about replies, payment,
    refusal, cash, hours, objections, setup, use, renewal, and referral where
    applicable.
11. Produce the completed one-page distribution brief.

Rules
- Don't choose or rank distribution channels; that happens in the next chapter.
- Don't invent customers, quotes, reach, conversion, costs, sales, retention,
  referrals, competitor behavior, or channel performance.
- Don't turn followers, impressions, compliments, free signups, or future
  intent into paying-customer evidence.
- Don't combine dollars and hours into one number.
- Don't call a route free when it requires my time or paid tools.
- Don't use an industry CAC, LTV, payback, or conversion benchmark as my target.
- Don't treat one sale as proof that the route will repeat.
- Use American English and ordinary language; use you and your when addressing
  me.

Return exactly these headings
## Facts and sources
## Assumptions
## Missing information
## Result needed now
## Route from problem to result
## Product, price, customer, and route check
## Cash available
## Time available
## Buying psychology
## Manual first action
## Review point
## Distribution brief

You'll still need to compare the output with your notes. A model can organize the route you describe; it can't discover where real customers pay attention or what they'll pay for by completing the blank spaces itself.

Sources

• Balfour, Brian. “Four Fits for $100M+ Growth.” Accessed 15 September 2026.
• Graham, Paul. “Do Things that Don't Scale.” July 2013.
• Hacker News. “Founders, what was the major sourcing channel for your first 100 users?” 18 October 2024.
• Shopify. “Customer Acquisition Cost (CAC): Calculate and Reduce It.” 29 July 2024.
• Stripe. “What Is the CAC Payback Period?” Accessed 15 September 2026.
• Weinberg, Gabriel, and Justin Mares. Traction: How Any Startup Can Achieve Explosive Customer Growth. Portfolio, 2015.