Make every link your customers share carry your product

This tactic makes every page a customer shares do two jobs. It serves the visitor, and it shows the visitor whose product made it. Gumroad let people sell straight to their Twitter and Facebook followers, and within weeks some artists with modest followings had made thousands of dollars. Two weeks in, its traffic matched Pinterest's January 2011 level, by the founder's account.

How it works

The shared link is the distribution. Your customer brings the audience, the page does the work for that audience, and your name rides along. This is a growth loop in its plainest form.

1. Pick the object customers already want to post. For Gumroad, that was a creator's product, sold directly to followers on Twitter and Facebook.
2. Make the shared page finish the visitor's task. The visitor came to buy, book or read. Let them do it on that page, without needing an account first.
3. Add a quiet path back to you. A short made-with line and a start-your-own button below the main action. The visitor who wants their own version becomes your next sharer.
4. Ship it rough and test demand first. An interview on how Gumroad started describes a first version built over one weekend, with payments run through Lavingia's personal PayPal and creators paid by hand at month end. He quickly found a dozen people with the same problem who would pay for it.
5. Use the product yourself. The idea came on a Friday night in April 2011 while Lavingia was designing a pencil icon. That $1 pencil earned him hundreds of dollars on Gumroad. He tweeted the idea early on a Monday while still at Pinterest, and Gumroad stayed a side project.

A press profile of the founder reports the early creator earnings and the traffic comparison.

What it takes

Our estimate: about 40 hours of work, and about 21 days before you can read a signal. Most of the hours go into a clean public version of the page people share, a made-with footer and button, and tracking that tags each visit and signup to the link it came from. Signal is slow because it depends on your customers' posting habits. A week of data mostly reflects your most active sharers. Three weeks lets ordinary customers share, and lets their visitors sign up and take a first action.

What to measure

The main reading is new signups per shared-link visit, compared against a holdout with no branding. Raw traffic is not enough. Gumroad's traffic claim came with no absolute figure, which is why it is hard to judge.

Track share links created per active customer. If branding makes customers share less, the loop shrinks even when conversion per visit goes up. Footer clicks as a share of visits tell you whether visitors notice you at all. The share of new accounts whose first visit was a shared link shows how much of your growth the loop carries. Week-four retention of those signups shows whether they are real users or curious clickers. A good reading is flat or rising sharing, more signups per visit than the holdout, and retention close to your other channels. A bad reading is clicks without signups, or signups that churn fast. To turn these into a K-factor, use the viral coefficient calculator.

When it backfires

• Small audiences, small loop. The Gumroad wins came from some artists, not a typical creator. If most of your customers have few followers, branded links will not produce much on their own.
• Branding that crowds the customer. If your logo competes with their content, your best customers will ask to remove it or leave for a tool that lets them. Keep the footer below their main action.
• Manual operations that do not scale. Paying creators by hand was fine for a first version. If your shared pages depend on manual steps, a sharing spike becomes a support backlog.
• Platform rules on links. Platforms can change how they treat outbound links and repeated promotion, so check each network's current rules. See tactics that stopped working before you lean on one network.
• Counting visits instead of new customers. A traffic milestone feels good and tells you little about signups or revenue.

Run it as an experiment

Hypothesis: adding a made-with line and a start-your-own button to shared pages raises new signups per shared-link visit without lowering how often customers share.

We recommend a holdout split at the sharer level. Randomly assign new sharing customers 50/50 to branded or unbranded pages, and keep the assignment fixed for four weeks. Size the test first with the A/B test sample size calculator. Log link creation, each visit with its referrer, footer clicks, signups tagged to the source link, and the new user's first key action.

Decision rule we suggest: ship if signups per visit rise clearly and links created per sharer do not drop. If sharing drops, test a smaller footer before giving up. Write it up using how to run a growth experiment, and log follow-ups in the growth experiment backlog template.

How strong the evidence is

This is a practitioner report. The numbers come from third-party write-ups, an interview and a press profile, and nothing was published by Gumroad itself. The single case is Gumroad, and its outcome is mixed. The figures come from press reports, some attributed to the founder, and none has a baseline. "Thousands of dollars" covers some artists, not all of them. The traffic comparison has no absolute number. Nothing in the sources isolates the shared link, or any branding on it, as the cause of growth. Treat this as a well-known pattern worth testing, not a proven effect.