Pitkä syväluotaus Nir Eyalin Hook-malliin — neljävaiheinen silmukka joka selittää tottumuksia muodostavia tuotteita — ja kuinka käyttää samaa psykologiaa rakentamaan käyttäjiä kunnioittavia kokemuksia.
This is the long companion to our ethical engagement resource page. The resource is the short version; this post is the deep dive — for designers, founders, and anyone building product who wants to understand both how the Hook Model works and how to apply it without becoming the kind of company you don't want to be.
The Hook Model recap
Nir Eyal's *Hooked* describes a four-step loop:
1. **Trigger** — something that prompts the user to act. Internal triggers come from inside (boredom, FOMO); external from outside (a notification).
2. **Action** — the simplest possible behaviour done in expectation of reward. Open the app. Tap the card.
3. **Variable reward** — the payoff. Crucially *variable* in size or kind, because the brain's dopamine response peaks on uncertain outcomes more than predictable ones.
4. **Investment** — the user puts work into the system: a like, a saved item, an updated profile. This loads the next loop with more relevance for them.
The brilliance of the framework is that it's descriptive: it explains why Instagram, slot machines, and the Anthropic console are sticky in similar ways. The model is morally neutral. The intent isn't.
The dark-pattern version, in detail
Five well-documented manipulative applications:
1. Engineered uncertainty
Slot-machine logic. Refresh the feed; sometimes there's something new, sometimes not. The brain's dopamine response peaks on the unpredictability, which is what makes the behaviour compulsive. Tristan Harris has covered this extensively — see his TED talks for the public-facing version.
2. Manufactured scarcity
"Only 3 left in stock!" "12 people viewing this property right now!" These signals are sometimes true and sometimes invented. When invented, they're legally classified as dark patterns under the EU Digital Services Act.
3. Emotional triggers tuned for vulnerability
Notifications timed for late-night insomnia. Ads served when the user is logged in to a "stressed" persona. The targeting that makes ads efficient also makes them predatory when applied without restraint.
4. Sunk-cost lock-in
Years of photos, friends, conversations stored on a platform. The investment was the user's; the lock-in is the platform's. Read our own your platform piece for the user-side counter.
5. Confirm-shaming
"No thanks, I don't want to save 30 %." The unsubscribe link buried in 8-point gray text. Microcopy designed to make declining feel like a mistake.
The honest version, in detail
The same four-step loop, deployed in service of the user instead of against them:
1. Triggers tied to genuine need
A reminder that fires when the user actually needs it. A meditation app that prompts at 8 pm because that's when the user said they wanted to meditate. A booking app that pings when a window opens that matches the user's stated criteria. The trigger is *consensual*. The user signed up for it; the user can turn it off.
2. Actions that are honestly simple
Real one-click unsubscribe. Form fields collapsed to the absolute minimum. A "remind me later" button that genuinely defers and doesn't quietly disappear the option. The simplicity is for the user's benefit, not as bait into a deeper funnel.
3. Rewards tied to real value
Every tap should leave the user marginally better off. A directory listing actually leads somewhere useful. A blog post is actually worth reading. A search result is actually relevant. There's no need to manufacture variable reward; reality is variable enough.
4. Investment that compounds for the user
Saved items the user can export. Customisations that travel with them. A profile that genuinely improves their experience over time. The investment is theirs and stays theirs.
How to design honestly when you genuinely want engagement
You can want engagement and still be ethical. The test we use:
- **The Maker Test.** Would you use this product yourself? Would you let your kid use it? Eyal himself proposed this in his follow-up *Indistractable*. It's the cleanest single check.
- **The Consent Check.** Did the user explicitly opt in to this prompt, this notification, this email? Or did the design assume consent?
- **The Exit Check.** How easy is it to leave? An honest design makes exit as visible as entry.
- **The "Why Now?" Check.** Why is the user being prompted *at this moment*? If the answer is "because it boosts engagement", reconsider. If the answer is "because something they actually care about just happened", proceed.
These four checks together are roughly the framework Eyal renamed the Manipulation Matrix — combining "would I use it" with "does it materially improve the user's life".
Concrete examples in our work
- The cookie banner on this site does not have a confirm-shame "no thanks I want to be tracked" pattern; it has plain "Accept all", "Necessary only", "Settings". Same psychological hook (Trigger → Action → Reward → Investment) but consensually deployed.
- The /get-started questionnaire is a Hook loop too: Trigger (visitor curiosity), Action (one question at a time), Variable Reward (your tier match changes based on your answers), Investment (the answers compound into a more accurate recommendation). And there's a visible escape hatch on the last step ("Not ready yet? Learn more first") that takes you to our resources without judgement.
- The newsletter we are not yet sending will fire monthly, not weekly. Not because we couldn't write more; because monthly is the cadence that respects the reader's attention.
The reading list
- *Hooked* — Nir Eyal. The original.
- *Indistractable* — Eyal's user-side follow-up. Useful even if you're a maker.
- Deceptive Design — Harry Brignull's catalogue.
- Center for Humane Technology — Tristan Harris and colleagues, broader-systemic critique.
- The Design of Everyday Things — Don Norman. The grandparent text. Still essential.
If you'd like to talk about applying any of this to a product or website you're building, we'd be a good first call.