Before the user commits

The first page should not ask for a relationship before it proves it has the thing I came for.

That is the line I keep coming back to whenever a service blocks reading, checking, or comparing behind an account wall, app install prompt, phone verification, or newsletter modal. I understand why teams do it. Accounts reduce abuse. Apps let teams send notifications, save preferences, and measure what is working. Some features genuinely need identity: posting, buying, saving, private messages, support history, refunds, and anything with payment or personal data.

But read-only access is different. A person may only want to check a store return window, compare a restaurant menu, read one community answer, confirm a bus notice, look at a warranty term, or see whether a product page is even relevant. If the first step is "make an account" or "install the app", the service is asking for commitment before it has earned basic trust.

I think the useful boundary is not guest versus account. It is reversible versus committed action.

A reversible action should usually have a guest path. Reading a public page, checking public stock, previewing prices, viewing a schedule, searching a help article, or seeing whether a community thread exists should not require a full identity. The user can leave without changing the system.

A committed action can ask for identity at the moment it becomes necessary. Save this cart. Post this reply. Book this appointment. Claim this refund. Keep this file. Receive alerts. At that point the account request has a job. The user can see why the service needs a durable contact or permission.

The worst version is a blind gate. It says sign in first, but it does not say what is on the other side, whether the content is available, whether the price changed, or whether the app is required for a real feature or just for tracking. That turns a normal product boundary into suspicion.

A better gate leaves a preview. Show the title, the first answer, the public fields, the item availability, the estimated fee, or the reason an account is needed. If abuse is the concern, rate-limit the guest path or hide write actions. If privacy is the concern, keep private details locked but let the public context remain readable. If the app is genuinely better for one feature, name that feature instead of pretending the browser cannot serve any purpose.

There are cases where no guest path is reasonable. Banking, medical records, school accounts, workplace tools, age-restricted spaces, and private communities can require identity at the door. Even then, the gate should explain the category of access, not just throw a login box at a confused visitor.

The record worth keeping is simple: what could a visitor see before the gate, what action triggered the gate, what reason was shown, and whether the user had a way back to the browser or guest flow. Without those details, every complaint collapses into "forced login bad". With them, teams can separate safety, privacy, conversion pressure, and plain broken access.

Let me see the door before I make an account

// COMMENTS

ON THIS PAGE