# Distribution Beats Product

> A good product with no way to reach people dies quietly. The 19 traction channels, the Bullseye method, channel math and cheap tests, explained simply.

- Author: Tai Pham Phu
- Published: 18 Nov 2026
- Tags: Startups, Entrepreneurship, Marketing, Small Business, Business
- URL: https://phamphutai.com/2026/11/18/distribution-beats-product/

In 2010, Drew Houston, the founder of Dropbox, gave a talk called "Startup Lessons Learned." One slide showed what happened when Dropbox paid for search ads. Each new customer cost **$233 to $388**. The product cost $99. The slide's verdict was one word: **"Fail."**

Dropbox already had a product people loved. The problem was that the way it reached people cost more than each person was worth.

So the team tried something else. They built a referral program into the product: invite a friend, and both of you get extra free storage. Houston's slides from that talk say it **permanently increased signups by 60%**, and that in the 30 days to April 2010, users sent **2.8 million** referral invites.

Same product. Different way of reaching people. Completely different business.

That's what this post is about. **Distribution is how your product reaches the people who will pay for it.** Most first-time founders treat it as something to figure out after the product is done. The founders who write the best books on the subject say that's backwards.

---

## Why good products die quietly

Peter Thiel's book *Zero to One* (2014, written with Blake Masters) has a chapter on distribution with a pointed title: "If You Build It, Will They Come?"

His answer is no. He argues that engineers in particular tend to believe a great product sells itself, and that they look down on sales and marketing as shallow. He calls this a mistake, and puts it bluntly: **"poor sales rather than bad product is the most common cause of failure."**

A few of his other points, paraphrased:

- **Distribution follows a power law.** One method of reaching customers usually matters far more than all the others combined.
- **If you get just one distribution channel to work, you have a great business.** Trying several and nailing none is how companies fail.
- **The basic math still rules.** What a customer is worth over their lifetime has to be more than what it costs, on average, to win them.

He also describes a "dead zone." Big, expensive deals can pay for a founder or salesperson to sell in person. Cheap products with mass appeal can be sold through advertising. In between, there's a price range where a sale is too small to justify a salesperson's time and too specialised to reach through mass advertising. Many small B2B products live right there. **If your product sits in the dead zone, you need to find a cheaper way to reach people, or change the price.**

The same idea appears in *Traction* (2014), by Gabriel Weinberg (the founder of DuckDuckGo) and Justin Mares. It makes a hard observation: almost every failed startup has a product. **What failed startups don't have is enough customers.**

---

## The words, explained once

- **Distribution:** everything it takes to get your product in front of the right people and get them to buy.
- **Traction:** evidence that customers want what you sell. Weinberg and Mares use it to mean growth: real users and real revenue.
- **Channel:** one specific route to customers, such as search ads, cold email or an app marketplace listing.
- **CAC (customer acquisition cost):** what you spend, in money and time, to win one paying customer through a channel.
- **LTV (lifetime value):** the gross profit one customer brings you over the whole time they stay.

If CAC and LTV are new to you, [Unit Economics in Plain English](https://phamphutai.com/2026/11/11/unit-economics-in-plain-english/) covers them properly. This post uses them per channel.

---

## The 19 traction channels

Weinberg and Mares interviewed more than forty successful founders and found that startups get traction through **nineteen different channels**. Here they are, as the book names them, with one line each:

| # | Channel | In plain English |
|---|---|---|
| 1 | Viral marketing | Users bring in other users, often through the product itself |
| 2 | Public relations (PR) | Getting covered by newspapers, magazines and big sites |
| 3 | Unconventional PR | Stunts and memorable gestures that get talked about |
| 4 | Search engine marketing | Paying for ads on search results |
| 5 | Social and display ads | Paying for ads on social networks and websites |
| 6 | Offline ads | Print, radio, TV, billboards |
| 7 | Search engine optimization | Showing up in free search results |
| 8 | Content marketing | Writing articles, guides and posts people seek out |
| 9 | Email marketing | Emails to people who already know you |
| 10 | Engineering as marketing | Free tools, calculators or widgets that attract customers |
| 11 | Targeting blogs | Getting niche blogs and newsletters to mention you |
| 12 | Business development | Partnerships that benefit both sides |
| 13 | Sales | Talking directly to buyers and closing deals |
| 14 | Affiliate programs | Paying others a commission for customers they send |
| 15 | Existing platforms | Growing on top of a bigger platform's users |
| 16 | Trade shows | Booths and meetings at industry events |
| 17 | Offline events | Running your own meetups, workshops or conferences |
| 18 | Speaking engagements | Giving talks to rooms full of possible customers |
| 19 | Community building | Bringing your users together so they help each other and you |

The authors found two patterns. First, **most founders only consider the channels they already know**, so many startups crowd into the same few and ignore the rest. Second, **you can't reliably predict which channel will work.** You can guess, but until you test, you don't know.

---

## The Bullseye framework

Bullseye is the book's method for finding the one channel that will unlock your next stage of growth. The book describes it as **a five-step process: brainstorm, rank, prioritize, test and focus.** Then you repeat it when the channel runs dry.

Here is each step in plain terms.

### 1. Brainstorm

Write down at least one realistic idea for **every one of the 19 channels**, even ones that feel wrong for you. The point is to beat your own bias. In a spreadsheet, guess for each idea how likely it is to work, what a customer might cost, how many customers it could bring and how long a test would take. Educated guesses are fine.

### 2. Rank

Sort the channels into three groups: the most promising right now, the ones that could plausibly work, and the long shots.

### 3. Prioritize

Pick your **inner circle: about three channels** that look most promising. The book says the drop-off in excitement often comes around the third channel. Three lets you run tests in parallel without losing focus.

### 4. Test

Run cheap, small tests on those three. Each test should answer three questions:

- **Roughly what will a customer cost through this channel?**
- **How many customers are available through it?**
- **Are these the customers you want right now?**

Think four Facebook ads, not forty: the book says **a few hundred dollars** should give you a rough idea of a channel. You're not trying to grow yet, only to learn fast.

### 5. Focus

If one channel looks promising, put your effort there and keep experimenting inside it until it stops working, because of saturation (you've reached most of the people it can reach) or rising costs. If nothing worked, go back to step 1 with what you learned.

<figure class="diagram">
<svg viewBox="0 0 680 380" role="img" aria-label="Bullseye diagram: the outer ring is brainstorming all 19 channels, the middle ring is testing about three promising channels, and the centre is focusing on the one channel that works" xmlns="http://www.w3.org/2000/svg">
<defs><marker id="dbp-arrow" viewBox="0 0 10 10" refX="9" refY="5" markerWidth="7" markerHeight="7" orient="auto-start-reverse"><path d="M0,0 L10,5 L0,10 z" fill="currentColor"/></marker></defs>
<text x="20" y="28" font-size="14">The Bullseye: wide, then narrow, then one</text>
<circle cx="190" cy="205" r="150" fill="none" stroke="currentColor" stroke-width="2"/>
<circle cx="190" cy="205" r="95" fill="none" stroke="currentColor" stroke-width="2"/>
<circle cx="190" cy="205" r="42" fill="currentColor"/>
<text x="190" y="82" font-size="12" text-anchor="middle" class="muted">all 19</text>
<text x="190" y="138" font-size="12" text-anchor="middle" class="muted">about 3</text>
<text x="190" y="210" font-size="13" text-anchor="middle" class="on-fg">one</text>
<line x1="330" y1="110" x2="400" y2="110" stroke="currentColor" stroke-width="1.5" marker-end="url(#dbp-arrow)"/>
<text x="410" y="106" font-size="13">Outer ring: brainstorm</text>
<text x="410" y="124" font-size="11" class="muted">one idea for every channel</text>
<line x1="280" y1="185" x2="400" y2="185" stroke="currentColor" stroke-width="1.5" marker-end="url(#dbp-arrow)"/>
<text x="410" y="181" font-size="13">Middle ring: rank and test</text>
<text x="410" y="199" font-size="11" class="muted">cheap tests, a few hundred dollars</text>
<line x1="232" y1="235" x2="400" y2="260" stroke="currentColor" stroke-width="1.5" marker-end="url(#dbp-arrow)"/>
<text x="410" y="258" font-size="13">Centre: focus</text>
<text x="410" y="276" font-size="11" class="muted">go deep until it saturates</text>
<text x="410" y="318" font-size="11" class="muted">When the centre runs dry,</text>
<text x="410" y="334" font-size="11" class="muted">start again from the outside.</text>
</svg>
<figcaption>Bullseye starts wide so you don't miss a channel, tests a few in parallel, then puts almost everything into the one that works.</figcaption>
</figure>

The book gives a real example. At Mint, the personal finance app, Noah Kagan's team set a goal of 100,000 users in the first six months after launch. They picked an inner circle of targeting blogs, PR and search ads, and ran cheap tests in each: sponsoring a small newsletter, reaching out to well-known finance personalities, buying some Google ads. **Targeting blogs won.** Sponsoring mid-sized finance bloggers and writing guest posts brought their first 40,000 users. When that channel maxed out, they ran Bullseye again and moved to PR. According to the book, they reached a million users within six months of launch.

---

## The 50% rule

The most practical idea in *Traction* may be this one. In the book's words: **"spend 50% of your time on product and 50% on traction."**

The authors admit splitting your time slows product work. Their argument is that it **doesn't slow down the time it takes to reach the market**, for two reasons:

1. **Traction work makes the product better.** You learn which messages land, which niche to start with and which customers are easiest to reach. That changes what you build.
2. **You launch with a channel already tested.** You don't hit "launch," hear nothing, and then start learning distribution from zero.

Their example is Phil Fernandez, founder of the marketing software company Marketo. In the book he describes running a blog and search optimization before the product existed. By launch, he says, they had **a pipeline of more than 14,000 interested buyers.**

My own view: 50% will feel wildly too high to most developer-founders, and that's exactly why the number is useful. The point is that traction work should start **while** you build, not after.

---

## Channel economics: every channel has its own CAC

Founders often talk about "our CAC" as if it's one number. It isn't. **Each channel has its own cost per customer, its own type of customer and its own ceiling.**

Dropbox is the clearest example. Through search ads, a customer cost more than the product's price. Through referrals, a customer cost some free storage. Same company, same product, two completely different CACs.

For each channel, ask:

| Question | Why it matters |
|---|---|
| What does one customer cost here, including my time? | Founder hours are the main cost of many "free" channels |
| How long until that customer pays back their cost? | Long payback ties up cash you may not have |
| How many customers can this channel reach before it saturates? | A cheap channel with 30 possible customers won't build a company |
| Do these customers stay? | A channel that brings people who leave in a month has a hidden cost |
| Does the cost go up or down as I scale? | Ads usually get more expensive; content and integrations often get cheaper |

The last row matters most for small founders. **Some channels cost the same for every customer, forever**: ads and cold outreach need new money or new hours for each one. **Other channels compound**: a page that ranks in search keeps bringing people in after the work is done. And count your time. Put a rough price on your hour and include it.

---

## Channels that fit small founders

With no budget and no team, TV ads and trade shows usually aren't realistic. These five often are. For the very first customers, [Your First 10 Customers Won't Come From Marketing](https://phamphutai.com/2026/11/11/your-first-10-customers/) covers outreach and conversations in detail. Here the question is which of these can become a repeatable channel.

### Content and search (SEO)

You write useful pages that answer the questions your customers search for. It's slow: months before search traffic arrives. But once a page ranks, it keeps working with little extra cost. Marketo's pre-launch blog is the example above. **It fits founders who can write clearly about a problem they know well.**

### Community

You show up, helpfully, where your customers already talk: forums, groups, Slack or Discord servers. Or you build a place for them yourself. It costs time, not money, and **it only works if you give much more than you ask.**

### Cold outreach (the book's "Sales" channel)

You find the exact people who have the problem and contact them directly. It doesn't compound, but it's fast and precise. **It fits products where one customer is worth a few hundred dollars or more**, so your hours pay off.

### Integrations and marketplaces (the book's "Existing platforms")

You build on top of a platform where your customers already are: an app store, a plugin directory, a SaaS marketplace, or a bigger product your tool connects to.

The famous example is Airbnb and Craigslist. In 2012, the growth writer Andrew Chen wrote a well-known essay about how early Airbnb let hosts also post their listings to Craigslist, a site with tens of millions of users, even though **Craigslist offered no public API** (a public way for other software to connect to it). Chen worked out how the feature probably functioned by studying it from the outside. He also wrote that no one outside the company knew exactly how much it was worth to Airbnb. What's clear is the idea: **put your product in front of a big audience that already exists.** (Note that a hack like this depends on the platform tolerating it, so read its rules.)

### Partnerships (the book's "Business development")

You team up with a business that already serves your customer, like an accountant who recommends your bookkeeping tool. One good partner can send customers for years, but setup is slow, and **it works best once you can show the partner what their customers gain.**

### What about viral growth?

Viral loops like Dropbox's referral program are powerful, but they only work when the product is naturally shared, or when inviting others makes it better for the person inviting.

Hotmail is the classic case. Steve Jurvetson and Tim Draper, two of Hotmail's investors, wrote the original article on what they called "viral marketing." Every email sent from Hotmail carried a short line at the bottom: **"Get your free email at Hotmail."** According to them, Hotmail signed up more than 12 million subscribers in its first year and a half, and spent less than $500,000 on marketing, advertising and promotion to get there.

The lesson isn't "add a footer." It's that **Hotmail's users were its distribution.** Every email was an advert, sent by someone the reader trusted. Look for the moment in your own product when a user naturally shows it to someone else, and make that moment easy.

---

## One channel usually dominates

Here the two books agree. Thiel says distribution follows a power law. *Traction* says it this way: **"At any stage in a startup's lifecycle, one traction channel dominates."**

You can see it in every story above: Dropbox's referrals, Hotmail's footer, Mint's blogs and then PR.

This is good news for a small founder. **You don't need to be good at 19 things. You need to find one and get very good at it.** Spreading a few hours a week across six channels usually means none of them gets enough effort to show you what it can do.

Channels saturate, so the dominant one changes over time. **Expect to run Bullseye again.**

---

## Running cheap channel tests: a worked example

Let's run Bullseye for a fictional company.

**"Quoteloop" (fictional)** sells a quoting tool for small renovation contractors. Contractors enter the job, and Quoteloop produces a clean, itemised quote to send to the homeowner. The numbers:

- Price: **$40 a month**
- Gross margin: **85%**, so **$34 of gross profit** per customer per month
- Monthly churn: **5%**, so an average customer stays about **20 months**
- LTV (gross profit): $34 x 20 = **$680**

Before testing anything, the founder sets limits. To hit the 3-to-1 LTV:CAC rule of thumb, a customer can cost at most **$680 / 3 = about $227**. To pay back within 12 months, at most **$34 x 12 = $408**.

The founder values their time at **$25 an hour**, an assumption to make hours comparable to money.

**Brainstorm and rank:** they write ideas for all 19 channels. Trade shows and offline ads go to the long-shot column. Their inner circle: **search ads, cold email and an integration listing** in the app marketplace of a popular invoicing tool that many contractors already use.

**Test:** four weeks, all three in parallel.

| Channel (fictional) | Cash | Hours | Reached | Trials | Paid |
|---|---|---|---|---|---|
| Search ads | $300 | 4 | 220 clicks | 9 | 1 |
| Cold email | $40 | 12 | 200 emails | 7 | 3 |
| Integration listing | $0 | 10 | 60 visits | 6 | 2 |

Now the math. Cost = cash + hours x $25. CAC = cost / paid customers. Payback = CAC / $34.

| Channel | Total cost | CAC | LTV:CAC | Payback |
|---|---|---|---|---|
| Search ads | $300 + $100 = $400 | **$400** | 1.7 | 11.8 months |
| Cold email | $40 + $300 = $340 | **$113** | 6.0 | 3.3 months |
| Integration listing | $0 + $250 = $250 | **$125** | 5.4 | 3.7 months |

The whole test cost **$340 in cash and 26 hours** of the founder's time over a month.

### Reading it

**Search ads fail.** $400 per customer is far above the $227 target: Dropbox's problem in miniature. Drop it.

**Cold email and the integration both pass.** But be honest about the sample: 3 customers versus 2 is not a real difference. **One more or one fewer sale would flip the ranking.** At this size, a test tells you which channels are clearly bad and which are worth more effort. It doesn't crown a winner by a few dollars.

So the founder looks past this month's CAC.

- **How do costs behave at scale?** The integration's $250 was mostly a one-time setup. If the listing keeps bringing in two customers a month, its CAC drops toward zero. Cold email costs the same hours for every new batch of 200.
- **How big is each channel?** That depends on how many contractors use the invoicing tool and find its marketplace, which the founder can't see yet.

**Focus:** the founder makes cold email the main channel for the next month, because it's the fastest way to learn and the easiest to scale on purpose. The integration listing stays live, since it costs nothing to keep and can grow on its own. At the same rates, 600 emails would bring about **21 trials and 9 paying customers** for roughly 36 hours of work. After that month, the numbers are bigger and the comparison is more trustworthy.

If the integration quietly keeps bringing in customers, it may deserve the focus instead. **Decide with what you have, then let new data overrule you.**

### How to design your own cheap test

- **Set the pass bar before you start**, as in [Validate an Idea in One Week](https://phamphutai.com/2026/11/17/validate-an-idea-in-one-week/). Deciding afterwards makes everything look good.
- **Count paying customers, not clicks or signups.**
- **Keep each test small and time-boxed.** A few hundred dollars and a few weeks, as the book suggests.
- **Track where every customer came from**: ask on signup, or use a separate link per channel.
- **Check whether they stay.** A [cohort table](https://phamphutai.com/2026/11/17/cohort-retention-the-chart-that-tells-the-truth/) split by channel shows it.

---

## Common mistakes

**1. "We'll do marketing after launch."** Both books warn against it. You lose months, and build without learning what works.

**2. Only trying the channels you know.** Developers often go straight to ads or a launch on a tech site. Brainstorming all 19 exists to catch this.

**3. Doing a little of everything.** Nothing gets enough effort to show results, so you learn nothing.

**4. Ignoring the price-channel fit.** A $10-a-month product can't pay for hour-long sales calls. A $5,000 contract probably won't close from an ad. Thiel's dead zone is real: match the channel to what a customer is worth.

---

## Why I think this matters most for developers

My honest opinion: for founders who come from engineering, this is the most important post in the series. Building feels like progress. Distribution means writing to strangers, being ignored and asking for money. But Dropbox didn't fix its ad problem by changing the product; it changed how people reached it. **A good product is the price of entry. Distribution decides whether anyone finds out.**

The good part is that you only need one channel that works. Find it with cheap tests, focus on it, and run the process again when it runs out.

---

## Key ideas

- **Distribution is how your product reaches paying customers**, and Thiel argues poor sales kill more companies than bad products do.
- ***Traction* lists 19 channels.** Most founders only consider the few they already know, and it's hard to predict which one will work.
- **Bullseye: brainstorm all 19, rank, pick about 3, test them cheaply in parallel, then focus on the one that works.** Repeat when it saturates.
- **The 50% rule:** spend about half your time on traction while you build, not after.
- **Every channel has its own CAC.** Include your hours, and check whether costs rise or fall as you scale.
- **One channel usually dominates** at each stage. You need one that works, not six that half work.
- **Use small tests to rule channels out**, set the pass bar in advance, and don't crown a winner on two or three customers.

---

*Sources: Drew Houston, [Dropbox: Startup Lessons Learned](https://www.slideshare.net/slideshow/dropbox-startup-lessons-learned-3836587/3836587) (slides, April 2010); Peter Thiel with Blake Masters, Zero to One (2014), chapter "If You Build It, Will They Come?", checked against [Graham Mann's book notes](https://grahammann.net/book-notes/zero-to-one-peter-thiel); Gabriel Weinberg and Justin Mares, [Traction](https://leananalyticsbook.com/wp-content/uploads/2014/08/Traction-1-3.pdf) (2014, published excerpt of chapters 1-3, including the 19 channels, the Bullseye steps, the Mint and Marketo stories and the 50% rule), and [Graham Mann's Traction notes](https://grahammann.net/book-notes/traction-gabriel-weinberg); Andrew Chen, [Growth Hacker is the new VP Marketing](https://andrewchen.com/how-to-be-a-growth-hacker-an-airbnbcraigslist-case-study/) (2012); Steve Jurvetson, [the original viral marketing article on Hotmail](https://steve.blog/2013/10/23/red-herring-hotmail-photoshoot-for-the-original-viral-marketing-article/) (republished on his blog).*

*Related: [Unit Economics in Plain English](https://phamphutai.com/2026/11/11/unit-economics-in-plain-english/), [Your First 10 Customers Won't Come From Marketing](https://phamphutai.com/2026/11/11/your-first-10-customers/), and [Positioning: Who It's For and Why It's Better](https://phamphutai.com/2026/11/18/positioning-who-its-for-and-why-its-better/).*
