Why do so many onboarding sequences lose people right after the first win? The pattern shows up everywhere: a subscriber clicks, gets the promised thing, and then… nothing. The progress bar freezes. And it freezes hardest when the reward we designed has no second step built into it.

The 60% problem is a completion problem, not a motivation problem

Behavioral economists have a name for the pull of an unfinished task. The Zeigarnik effect, first observed by Bluma Zeigarnik in the 1920s, describes how interrupted or incomplete tasks stay more active in memory than finished ones. That's usually framed as a good thing — it's why cliffhangers work. But it cuts the other way too. When a reward arrives and the experience simply stops, the brain files it as done. The open loop closes. There's nothing left tugging at attention.

In email terms: a welcome message that delivers the lead magnet and says "stay tuned" is a closed loop with a shrug attached. The subscriber got the thing. The task is complete. The next email competes with a brain that has already checked the box.

Variable rewards need a runway, not a dead end

Variable-ratio reinforcement — the schedule where a reward arrives after an unpredictable number of actions — is one of the most durable patterns in behavioral psychology. B.F. Skinner's work on it is often cited precisely because it produces persistent behavior. But there's a detail people skip: the reinforcement only sustains behavior while the behavior itself keeps producing opportunities for the next reward. If the sequence ends, the schedule ends.

This is where a lot of email programs quietly break. They front-load a strong offer, then follow with generic newsletters that have no next rung. The subscriber isn't bored so much as finished. The reward had no second step.

What a second step actually looks like

A concrete example from the world of habit research: the "small wins" studies popularized by Teresa Amabile and Steven Kramer found that incremental progress on meaningful work is one of the strongest drivers of engagement — but only when the person can see the next increment. A win that doesn't point anywhere reads as an ending, not a milestone.

Translate that to email. Instead of:

• Email 1: Here's your guide.
• Email 2: Here's our newsletter.

Try:

• Email 1: Here's your guide. Step 1 of 3 is to try the first exercise.
• Email 2: How did Step 1 go? Here's Step 2, which most people find harder.
• Email 3: You're at the last step. Here's what people do next.

The reward still lands. But it lands inside an open loop.

Loss aversion is the quiet engine here

Kahneman and Tversky's work on loss aversion showed that people weigh potential losses more heavily than equivalent gains. In an email sequence, the "loss" is the unfinished thread. If a subscriber can see that they're 60% through something — a mini-course, a checklist, a challenge — abandoning it registers as a small loss. Not a big one. Just enough to make the next click feel like the path of least resistance.

The mistake is designing rewards that feel complete on arrival. The fix is designing rewards that arrive mid-sequence.

What to build next

Look at your highest-performing first email. Now ask: what's the second step inside the reward itself? Not the next email — the next action the reward implies. If there isn't one, you've built a progress bar that was always going to stall at 60%. Give the reward somewhere to go, and the sequence carries itself.