How to Choose a Customer Acquisition Channel to Test

A folded paper airplane beside a crumpled paper ball on blue.
Photo: Matt Ridley / Unsplash

Apply Traction with a small channel experiment that defines the audience, next customer action, effort limit, and evidence needed for another test.

Updated 20 October 2026.

Choose an acquisition channel by comparing plausible ways to reach a specific audience, then testing a small number with a clear customer action in mind. Record the effort required and the quality of the response. A channel that produces attention may still be a poor fit for the business.

Traction by Gabriel Weinberg and Justin Mares presents the Bullseye approach to finding a useful customer acquisition channel. The publisher-hosted material supports its experimental, channel-selection theme. The test card and fictional example below are our own application rather than a complete restatement of the book.

Define the audience and the next action

“Reach more people” makes it difficult to compare channels. Describe who needs the product and what a suitable person should do next: read a relevant explanation, try a sample, or request a discussion.

The action should fit the buying situation. A large service purchase may require several conversations, while a small self-service product may allow a direct trial. Do not compare raw click counts across very different customer journeys as if they measure the same result.

Also record how you will recognize an unsuitable response. Broad attention can look encouraging while giving you little evidence about the intended audience.

Compare several plausible routes

List channels your audience might actually use. Consider the match between the context and the task: a professional community, a relevant partner, an existing publication, or a carefully targeted campaign may each offer a different opportunity.

For each route, estimate the preparation effort, any cost, the time needed to observe a result, and the kind of response you could reasonably expect to inspect. Mark estimates as estimates.

Avoid choosing solely because a channel worked for another business. Their audience, offer, resources, and timing may differ from yours.

Work through a small service example

Imagine a reporting service for independent event organizers. The team considers an educational article in a relevant community, a partnership with a venue network, and a small permitted advertisement.

For a fictional first test, the team offers a clearly described sample report through the community's normal contribution rules. The next action is requesting the sample for a real planning task. No outreach or publication is performed by this article.

The team records the time spent preparing the contribution, how many suitable people request the sample, and what they do with it afterward. Any numerical target would need to be chosen for the actual business; there is no universal passing count in this example.

The test could suggest a promising route, reveal a weak offer, or show that the audience was misidentified. Those are possible interpretations, not reported outcomes.

Separate channel performance from offer problems

If the response is weak, inspect the whole path. Did suitable people encounter the message? Was the offer understandable? Was the requested next step reasonable? Did the destination work?

A channel cannot be evaluated fairly if the sample link is broken or the message describes the wrong problem. At the same time, revising every element at once makes the next result harder to interpret.

The customer-acquisition learning path provides broader reading. This article's narrower task is to make one channel test interpretable.

Decide what the evidence justifies

Record whether to repeat, change, expand cautiously, or stop the test. Include the uncertainty and the operational cost, not just the most flattering metric.

For recall, explain why a high number of visitors does not necessarily identify the best acquisition channel. Then draft a test card with an audience, channel, offer, next action, effort limit, and review question. Leave unknown values labeled as unknown until you have evidence.

Continue with ReadSprint

Read the linked author or publisher resource for this book's framework. For a related ReadSprint summary, explore The Lean Startup. Use the free Book-to-Action Plan to organize your own takeaways into practice and review phases; the tool arranges the entries you supply rather than validating the method or predicting results.

Source checked 20 October 2026: Portfolio’s Traction overview and excerpt. The source supports the attributed book concept. The scenarios, questions, and practice steps in this article are original ReadSprint suggestions; example outcomes are hypothetical.

Try one idea, then return to it.

Choose one useful idea, save a practical action, and give yourself a reason to revisit it.

Keep this book

Save it to your library and return to this review after signing in.

Save one action

Write down an idea you want to use in your work or daily life.

Come back to review

Sync your notes and review dates, then find due books on your dashboard.

14-day trial • No credit card required • Sign in with Google