Behavioral Design Examples: 6 Patterns That Hook You

I'll admit something that is a little embarrassing for someone who designs products for a living. I have opened Duolingo at 11:58pm, in bed, half asleep, purely to keep a streak alive. Not to learn Portuguese. To stop a number from resetting to zero.
That number has no value. I know exactly what it is doing to me. And it works anyway.
Here is the part most people get wrong about that moment. The reflex is to call it manipulation, file it under "dark patterns," and feel a little superior. But the same mechanics that pulled me back to a language app are the ones that make a savings app worth keeping, a fitness tracker worth checking, and a design tool worth opening on a Monday. The pattern is not evil. The intent behind it is what you have to interrogate.
So let's take these patterns apart, one by one, and look at what each one is actually doing to your brain, where it turns useful, and where it turns ugly.
The short version: behavioral design is the practice of shaping a product around how people actually decide and act, not how we wish they would, using patterns like triggers, variable rewards, streaks, and endowed progress. The same mechanics power both addictive apps and genuinely helpful ones. What separates them is whether the pattern serves the user's goal or only the company's metric.
What is behavioral design?
Behavioral design is a UX approach that builds products around predictable human behavior, using insights from psychology and behavioral economics to guide what people do inside an interface. Instead of asking "what should this screen contain," it asks "what will a distracted, busy, slightly lazy human actually do here, and how do we make the right action the easy one?"
It sits one layer beneath visual design and interaction design. I wrote a longer piece on why this shift makes design a more valuable UX approach than the old checklist mindset, and it is still the lens I reach for first on any product with a retention problem. If you want the underlying mental models rather than the applied patterns.
Why do these patterns matter? (The retention math)
They matter because attention is the scarcest thing your product will ever compete for, and the numbers are brutal. Across most mobile app categories, median Day-30 retention sits around 4%, and anything under 5% is considered normal (UXCam's 2026 benchmark analysis, drawing on AppsFlyer, Adjust, and data.ai data). Read that again. Thirty days after install, the typical app has lost 96 of every 100 users.
Behavioral patterns are the difference between being in the 4% and being forgotten. That is why they are worth understanding whether you want to use them or defend against them.
The teardown: 6 behavioral design patterns and what they exploit
Here is the honest map before we go deep. Every one of these has a version that respects the user and a version that preys on them.
Pattern | What it leans on | Honest use | Manipulative use |
|---|---|---|---|
Triggers | External cues, association | A reminder for a task the user asked for | Fake-urgency pings to farm re-opens |
Variable rewards | Dopamine, uncertainty | Genuine surprise and delight | Infinite feeds engineered to trap |
Streaks | Loss aversion | Motivating a real daily habit | Guilt over a meaningless number |
Endowed progress | Goal-gradient effect | Reducing onboarding drop-off | Manufacturing sunk cost |
Investment | Effort justification (IKEA effect) | Personalization that improves the product | Data lock-in with no exit |
Social proof | Conformity, FOMO | Honest signals of what others chose | Invented scarcity and fake counts |
Now the detail.
Triggers.
A trigger is the cue that starts a behavior, and it comes in two flavors: external (a notification, an email, a badge) and internal (an emotion, like boredom or loneliness, that you learn to answer with the app). The best products graduate from external to internal. You do not need a push notification to open Spotify. You feel a silence and reach for it. That is the goal, and it is also where the ethics get sharp, because a product that attaches itself to your boredom is powerful in a way that deserves care.
Variable rewards.
This is the slot machine, and it is the most abused pattern in the industry. When the reward for an action is unpredictable, the anticipation itself becomes the hit. Pull to refresh works because sometimes there is something good and sometimes there is not. Used well, variable rewards make discovery feel alive. Used badly, they turn a feed into a casino with no exit door.
Streaks and loss aversion.
My midnight Duolingo problem lives here. Once you have built something, losing it hurts roughly twice as much as gaining it felt good, which is loss aversion in one sentence. A streak converts a habit into something you can lose, and suddenly the daily action is not about the reward anymore, it is about not breaking the chain. Duolingo built a large part of its comeback on the streak for exactly this reason. The pattern is genuinely motivating when the underlying habit is worth having, and quietly cruel when the number is arbitrary.

Endowed progress.
This is my favorite because the research is so clean. In a classic 2006 study by Joseph Nunes and Xavier Dreze, customers were given a car wash loyalty card. One group got an 8-stamp card. The other got a 10-stamp card with 2 stamps already filled in. Both required the same 8 purchases. The pre-stamped group completed the card at 34% versus 19% for the plain card, nearly double, because progress that is already underway is far more motivating than progress that has not started. This is why good onboarding shows you a progress bar that begins at "step 1 of 4 complete," not empty. When I designed the loan application flow for Monety, a fintech micro-loans product, surfacing how close someone was to a decision, rather than how much paperwork remained, changed how people felt about a genuinely tedious form.
Investment.
The more effort you put into a product, the more you value it and the harder it is to leave. Behavioral economists call the general version the IKEA effect: you love the wobbly bookshelf more because you built it. In products, every playlist you curate, every board you fill, every profile you tune is an investment that raises the switching cost. The honest version makes the product genuinely better for you the more you use it. The dishonest version just makes your data hostage.
Social proof.
People look to other people to decide what is safe and what is worth doing. "Booked 14 times today" or "used by 40,000 designers" lowers the risk of a decision by borrowing the crowd's judgment. It is one of the most effective patterns in existence and one of the most faked, which is exactly why invented urgency and fabricated counts erode trust so fast when users catch them.
Where does behavioral design cross into dark patterns?
The line is not the pattern, it is the answer to one question: does the pattern help the user do something they actually want to do, or does it only serve a metric at the user's expense? A streak that keeps you meditating is on the right side of that line. A streak engineered so that missing a day feels like a personal failure, on a product you get nothing real from, is on the wrong side. I unpacked this trade-off in detail in a piece on designing habit-forming products without being manipulative, because it is the question that actually matters once you know how the mechanics work.
A useful gut check comes from accessibility and comfort-first thinking. When I looked at what Pokemon Sleep gets right about accessibility and comfort, the interesting part was that its behavioral loop is built around a genuine goal the user shares (sleeping better), not against it. That alignment is the whole game.
Key takeaways
Behavioral design shapes products around how people actually behave, using patterns like triggers, variable rewards, streaks, endowed progress, investment, and social proof.
The same six patterns power both addictive and genuinely helpful products. The mechanic is neutral; the intent is not.
Retention is why they matter: median Day-30 app retention is about 4%, so the patterns that create habits are the difference between surviving and being deleted.
Endowed progress is one of the most reliable levers: pre-filled progress nearly doubled loyalty card completion in the Nunes and Dreze study (34% vs 19%).
The ethics test is a single question: if the user fully understood what the pattern was doing, would they thank you or feel tricked?
FAQ
What is behavioral design in UX?
What are common behavioral design patterns?
Is behavioral design the same as dark patterns?
What is the endowed progress effect?
Why do streaks work so well?
How do I use behavioral design ethically?
Does behavioral design improve retention?
There is your teardown. The patterns are not the problem, and pretending you are above them will not make your product stick. The real work is holding each one up to a single question: if my user saw exactly what this was doing, would they thank me for it? Build the ones that survive that question, and delete the rest.
So which pattern is quietly running your favorite app right now, and are you okay with it?
Share on


