How I test

I do not write hype. I am looking for what actually feels worth your time, what gives me pause, and what you should check before spending money. My numeric score is visible, and it is the only score used in the public ranking.

I start with the actual experience

The features an app offers are part of its appeal. I look for the point where an app becomes interesting in practice: the first conversation, the character setup, the roleplay flow, and whether it still gives me a reason to come back after the novelty wears off.

I give the app more than one first impression

I do not use one artificial stopwatch for every product. A simple chat app may need an evening to understand, while an app with character creation, memory, media tools, credits, or community features needs several focused sessions. When I revisit an app, I record the new check rather than pretending the first impression is permanent.

I check the details around the chat

Depending on the app, that can mean memory, persona controls, character discovery, images, video, voice, scenarios, or community features. The range of features, the way they are presented, and the possibilities they offer all count toward my assessment. Having more ways to chat, create, or explore can be an advantage in itself. I also consider ease of use, access limits, and any extra costs.

I use both free and paid access

Sometimes I pay for a subscription or credits; other times I review what I can access for free. I explain what I tried and where a paywall stopped me. For features I have not used, I distinguish the available controls and the provider’s descriptions from results I tested myself.

I watch the price pressure

Subscription plans, credits, tokens, chips, and renewal terms matter as much as the first few messages. I flag the places where a “free” experience has limits, where media can consume paid currency, and what I would read twice before entering payment details.

I do not skip privacy or adult-content rules

For an AI companion app, I look at the current privacy information, account controls, visibility settings, age gates, and the rules that apply to user-made content. That is not legal advice. It is the practical check I think readers deserve before sharing anything personal or spending money.

I separate evidence from opinion

I label official claims as claims, describe what I actually saw in the product, and mark my conclusion as my conclusion. Screenshots, prices, and policy details are checked against the review date because none of them stay frozen forever.

My score is public. your response is separate.

I give every app one visible score based on my review. If you try the app, you can choose Yes or Next and select what you liked. Those signals are stored in the backend for now; they are not displayed publicly and do not change my score or the ranking.

The limits of a test

I can report what I found, not promise what every reader will find. Apps can change their models, prices, moderation, privacy rules, and available features. Before paying, check the current offer and terms yourself—especially if the app handles intimate conversations, personal images, or recurring subscriptions.