Building Products People Actually Want

By Pritesh Yadav June 22, 2026 11 min read —

Everything you have learned so far in this book — how attention works, how habits form, how people choose under uncertainty — gets put to work the moment someone designs a product. This chapter is about behavioral design: deliberately shaping a product so that people find it easy, rewarding, and worth coming back to. Done well, behavioral design helps a person do something they already wanted to do. Done badly, it traps them into doing something that serves the company at their expense. The whole craft lives on that line, and the goal of this chapter is to show you how to stay on the right side of it — not just because it is decent, but because it is the only thing that wins over the long run.

Key takeaway: The best products earn their place by solving a real need so smoothly that using them becomes second nature. Manipulation can fake that feeling for a while, but it always burns the trust that keeps customers around.

What "behavioral design" actually means

Let me define the term plainly. Behavioral design is the practice of arranging a product's steps, words, defaults, and rewards so that a target behavior is more likely to happen. Notice the word arrange. You are not adding willpower to the user. You are changing the situation around the behavior. A good behavioral designer asks: "What is making this hard, and how do I make it easy and worthwhile?"

Behavior: Any action you want the user to take: sign up, finish setup, place an order, come back tomorrow.
Friction: Anything that makes the behavior harder — extra steps, extra fields, confusing choices, waiting, or mental effort.
Trigger (or prompt): The cue that tells someone "do it now" — a notification, a button, or an internal feeling like boredom.

The Hook Model: how a product becomes a habit

Author Nir Eyal studied how products like Instagram and email become things we reach for without thinking. He described a four-step loop called the Hook Model: Trigger → Action → Variable Reward → Investment. Run a person through this loop enough times and the behavior turns into a habit — something done with little conscious thought.

Trigger — the cue to act. An external trigger is a notification or app icon. An internal trigger is a feeling, like boredom or loneliness, that gets linked to the product. The aim is to move people from needing a notification to opening the app on their own when they feel a certain way.
Action — the simplest thing done expecting a reward: open, scroll, search, tap.
Variable Reward — the payoff, made unpredictable. This is the engine. Your brain releases dopamine on the anticipation of a reward, not the reward itself, and unpredictability keeps that anticipation high. Rewards come in three flavors: the tribe (social approval — likes, comments), the hunt (information or resources — a feed, search results), and the self (mastery and completion — leveling up, finishing a streak).
Investment — the user puts in a little work: follows people, builds a playlist, uploads photos, earns a reputation. This does two things: it loads the next trigger (you posted, so you will get notified about replies), and it makes the product more valuable to you specifically, which makes leaving costly.

Analogy: A slot machine is the pure form of the variable reward. You pull the lever (action), and sometimes you win, mostly you don't — and that uncertainty is exactly what keeps your hand moving. Most engaging apps are gentler, well-meaning cousins of the slot machine: the feed that refreshes into something new each time is "the hunt," paid out on a random schedule.

Example: Duolingo's streak is the Hook Model in miniature. A reminder (trigger) nudges you to do a lesson (action); you earn points and the satisfying "ding" of progress, which varies day to day (variable reward); and the growing streak number is your investment — the longer it is, the more it hurts to break, so you come back.

Common mistake: Assuming a bigger reward is a stronger hook. It isn't. Variability beats size. A predictable reward, no matter how large, stops creating that pull of anticipation — your brain learns to expect it and stops caring. A small but uncertain payoff keeps people far more engaged than a large, scheduled one.

B = MAP: why behavior really happens (and why friction is your best lever)

Stanford researcher BJ Fogg boiled behavior down to a simple formula: a behavior happens only when Motivation, Ability, and a Prompt all show up at the same moment (B = MAP). Picture a graph with motivation going up the side and ability going across the bottom. There's an "action line." A prompt only works if the person is already above that line — motivated enough and able enough. Fire a prompt below the line and you have just created an annoyance.

Here is the single most useful insight in all of behavioral design: it is usually easier to raise ability (reduce friction) than to raise motivation. Motivation is fickle and expensive — pep talks and campaigns fade by tomorrow. But if you make the action genuinely easy, it happens even on low-motivation days. Fogg lists six things that drain "ability" — time, money, physical effort, mental effort, social awkwardness, and breaking from routine — and advises finding the scarcest one in that moment and removing it.

Example: You want users to refer friends. The weak approach spends money convincing them referrals are great (raising motivation). The strong approach pre-writes the invite, autodetects contacts, and makes it a single tap (raising ability). The second one works because it removes the scarcest resource — effort — instead of fighting the unreliable one.

Reducing friction: the central lever

Every extra step in a flow loses people. The math is brutal: a five-step process where each step keeps 80% of users retains only about a third of them at the end (0.8 multiplied by itself five times). Cutting steps, fields, and decisions is often the highest-return work you can do.

Cut form fields to the bare minimum; offer autofill and guest checkout.
Reduce choices. Hick's Law says decision time grows as you add options — fewer, clearer choices are faster.
Show progress so people know how close they are to done.
Make the high-value action one click. Amazon patented "1-Click" checkout in 1999 and it was valuable enough that Apple licensed it.

Common mistake: Believing "less friction is always better." It isn't. Strategic friction protects the user: a confirmation step before deleting everything, a cooling-off pause before a huge purchase, an extra check that stops fraud. Frictionless is a tool, not a virtue. The rule is: remove friction that blocks what the user wants; keep friction that protects them from regret.

Smart defaults: the quiet superpower

A default is the option people get if they do nothing — and people stick with it far more than you'd expect. The proof is striking. In countries where you are an organ donor unless you opt out, consent runs around 85–99% (Austria is near 99%). In otherwise-similar countries where you must opt in, it is 4–28% (neighboring Germany is around 12%). Same culture, opposite results — the only difference is the default. A famous workplace study found that switching retirement savings from "sign up if you want" to "enrolled unless you opt out" pushed participation from about 37% to 86%.

Why are defaults so powerful? Three reasons stack up: inertia (changing takes effort), the default reads as an implied recommendation ("they must have set this for a reason"), and loss aversion (changing feels like giving something up). The crucial point for an ethical designer is this: there is always a default. You cannot avoid having one. So the only real question is whether you set the default to serve the user or to serve yourself.

Key takeaway: Setting a default that genuinely fits what 90% of users want is great design. Setting a default that benefits you at the user's expense — pre-checked add-ons, auto-renew they didn't notice, sharing they didn't intend — is manipulation wearing a helpful mask.

The thing that makes it all legitimate: solving a real need

An engaging product is not automatically a good product. Infinite scroll is wildly engaging and often leaves people feeling worse. So how do you tell the difference? The clearest lens is Jobs To Be Done: people don't buy products, they "hire" them to do a job. The classic study found people bought thick milkshakes not because they loved the taste but to make a boring morning commute more interesting and to stay full until lunch. Build around the real job, and engagement becomes the honest byproduct of being useful.

Analogy: Theodore Levitt's line: "People don't want a quarter-inch drill, they want a quarter-inch hole." And really they want the picture hung and the feeling of a finished home. Design for the hole and the feeling, not the drill — and certainly not for "minutes spent holding the drill."

The line: where good design becomes manipulation

Nir Eyal offers a clean ethical test he calls the Manipulation Matrix, built from two questions: Does this materially improve the user's life? and Would the maker use it themselves?

Improves user's life?	Maker would use it?	What you are
Yes	Yes	Facilitator — the goal. Build these.
Yes	No	Peddler — questionable; you don't believe in it.
No	Yes	Entertainer — fine if honest about it.
No	No	Dealer — exploitation. Don't.

There is also a simpler gut check, the transparency test: Would this tactic still work if the user knew exactly what you were doing? Honest persuasion survives disclosure — "only 3 left, here's why" still works when it's true. Manipulation collapses the moment it's exposed — a fake countdown timer is worthless once the user realizes it resets every visit. That difference is the whole ethics of the field in one question.

Common mistake: Chasing engagement metrics as if they prove value. Time-on-app and daily-opens can go up precisely because a product is frustrating people into compulsive checking. A metric that rises when you make the user's life worse is not a success metric — it's a warning light.

Dark patterns: what not to do

Dark patterns (now often called deceptive design) are interfaces built to trick users into acting against their own interest. Knowing the common ones helps you recognize and avoid them:

Roach Motel — easy to sign up, painfully hard to cancel. This is the most heavily prosecuted; regulators have gone after major companies over it.
Forced Continuity — a free trial silently rolls into a paid charge.
Hidden Costs / drip pricing — fees that only appear on the final checkout screen.
Confirmshaming — guilt-trip decline buttons like "No thanks, I hate saving money."
Sneak into Basket — items auto-added to your cart.
Privacy Zuckering — tricking people into over-sharing personal data.

These aren't just rude — they're increasingly illegal. Regulators treat them as deceptive practices, the EU's Digital Services Act bans them outright, and one game maker paid $520 million partly over dark-pattern billing.

How to apply this — a practical checklist

Start from the job. Name the real need before designing a single screen. If you can't, you're building a toy, not a tool.
Diagnose before you fix. When a feature goes unused, ask: motivation problem or ability problem? Usually it's ability — so reduce friction first.
Find the scarcest resource in the moment (time? effort? confusion?) and remove friction there specifically.
Set defaults that serve the user. Pick the option 90% would choose; make opting out as easy as opting in.
Keep protective friction on destructive or expensive actions.
Make rewards real, not just frequent. The payoff should be genuine usefulness, not an empty dopamine drip.
Run the transparency test on every persuasive element. If it dies under disclosure, cut it.
Make canceling as easy as signing up. This one rule eliminates most dark patterns and most lawsuits.

Example (a SaaS store dashboard): You want shop owners to check orders each morning. The ethical hook: a useful summary email (external trigger) → one tap to the dashboard (low-friction action) → seeing today's real orders and what needs attention (a genuinely valuable "hunt" reward) → and the setup work they invested (configured products, saved layouts) makes the tool more theirs over time. Every step serves a job the owner already has. Nothing here would stop working if you explained it to them — which is exactly how you know it's design, not a trap.

Key takeaway: Build products people actually want by making the right behavior easy, the reward real, and the defaults generous to the user. The honest test is simple: if the people using your product would thank you after learning exactly how it works, you're a facilitator. If they'd feel tricked, you're building a debt that trust will eventually collect.

Continue reading

🎯 Psychology of Decisions

What Psychology Is, and Why "Rational" People Decide Irrationally

🎯 Psychology of Decisions

Two Minds in One Head: Fast Thinking vs Slow Thinking

🎯 Psychology of Decisions

How the Brain Actually Decides: Emotion, Memory & Attention