samarv/comprehension-first-design

Comprehension-First Product Design

Product design is often mistakenly focused on reducing "friction" (the number of clicks or taps). However, for most unique products, the real barrier isn't effort—it's comprehension. If a user has to stop and think because they don't understand an option, they feel "stupid," creating an emotional debt that leads to churn.

Core Principles

1. The Owner’s Delusion

Recognize that as a creator, you are "delusional" about the user's level of intent.

• The Reality: Users have low intent, are likely late for a meeting, or are distracted by personal stress.
• The Application: Do not force users to sit through "Ken Burns effect" intros or complex menus. Provide the "Street Address, Phone Number, and Menu" (the core value) immediately.

2. Focus on "Don't Make Me Think"

Every decision costs biological energy (glucose/ATP).

• Good Design: 8 clicks that are "trivially easy" and require zero thought (fluidity).
• Bad Design: 2 clicks that require a "fraught decision" between poorly defined options.

3. Move Up the Utility Curve

Features are not binary; they exist on an S-curve.

• The Shallow Start: Initial effort yields zero value (e.g., creating a database table).
• The Steep Hump: The "Magic Threshold" where the feature becomes indispensable.
• The Goal: Invest enough in a feature to get it over the hump of comprehension so users actually reach the "Aha" moment. If you can't get it over the hump, don't build it.

---

Tactical Implementation

Use "Good Friction" for Comprehension

Sometimes, adding a step improves the experience by shaping behavior or preventing regret.

• The Shouty Rooster Pattern: When a user is about to do something high-impact (like @channel in a large Slack room), interrupt them with a warning ("This will notify 147 people in 8 time zones"). This "friction" builds comprehension of the tool's power and social cost.

Apply Smart Defaults

If you don't set a default, most users will never use the feature.

• Override Hierarchy: Set a sensible system default (e.g., Do Not Disturb from 8 PM to 8 AM). Allow administrators to override it, and then allow end-users to override that. This introduces the feature to everyone without causing conflict.

The "Other" Menu Strategy

Avoid "Choice Paralysis" by hiding complexity.

• Identify the top 2-3 most common actions.
• Place everything else behind an "Other" or "More" menu.
• Comparing 3 items is easy; comparing 15 is geometrically more expensive for the brain.

---

Examples

Example 1: Onboarding Authentication

Context: A user downloads a mobile app and needs to sign in. Input: Traditional Username/Password fields. Application: Apply "Don't Make Me Think." Typing a complex password on a phone is high friction and high cognitive load. Output: Use a Magic Link. The user enters an email, clicks a link in their inbox, and the app automatically authenticates. It replaces "remembering/typing" with "clicking a button."

Example 2: Time Zone Selection

Context: A user needs to set an event time in a different time zone. Input: An alphabetical list of all 400+ global time zones. Application: Combat the "Owner's Delusion." Most users only care about 10-12 major zones. Output: Prioritize the 5 most likely zones based on the user's current location and past behavior (e.g., if in SF, show Pacific, Eastern, and London first). Use a sub-menu or search for the "long tail" (e.g., Newfoundland).

---

Common Pitfalls

• The Click-Count Trap: Thinking that "fewer clicks" always equals "better UX." If those fewer clicks require 10 seconds of staring at the screen to understand, you’ve failed.
• Optimizing "Junk": Improving a feature that is still in the "shallow" part of the utility curve. If the feature handle is still "breaking on impact," making it slightly prettier doesn't matter.
• Ignoring the "Stupid" Factor: Designing icons or labels that require a manual to understand. If a user feels dumb using your software, they will eventually hate it.
• Regression to Complexity: Allowing "hyper-realistic work-like activities" (like A/B testing minor UI tweaks) to override clear product intuition. If an A/B test shows a 2% gain but makes the UI more confusing, reject it.