How we pick our winners
There is no shortage of “best of” lists in this space. Most of them are either affiliate farms dressed up as editorial, outdated roundups that haven’t been touched in two years, or copy-paste jobs from other sites with the order reshuffled.
We do things differently. Every platform that earns a recommendation on AdultReview has been through the same structured evaluation process — tested by a real person, scored against consistent criteria, and reviewed again when things change. Here’s exactly how that works.
Step 1: We Sign Up Like a Regular User
Our review process starts at the registration page, not the press portal.
We create accounts using standard sign-up flows – no special reviewer access, no pre-loaded credits, no demo environments. If a platform’s onboarding is confusing, we experience that confusion. If the free tier is too limited to evaluate fairly, we note it and upgrade.
We test free tiers first to assess what a new user actually gets before spending money. Where the meaningful functionality sits behind a paywall, we pay for access. We don’t review platforms we can’t properly use.
Step 2: We Run a Standardised Testing Process
To make comparisons meaningful, we apply the same testing framework to every platform rather than reviewing each one in isolation.
For generation platforms, our testing process includes:
- Running a consistent set of test prompts across platforms to compare output quality directly
- Varying prompt complexity from simple to detailed to assess how well the platform interprets instructions
- Testing generation across different styles, settings, and request types to evaluate range and flexibility
- Measuring generation speed across multiple sessions at different times of day
- Assessing output consistency – whether repeated or similar prompts produce reliably good results
For companion and chat platforms, our testing process includes:
- Extended sessions across multiple days to assess conversational depth and memory
- Testing character customisation depth and how much user input shapes the experience
- Evaluating response quality, coherence, and how the platform handles varied input styles
- Assessing how well context and continuity are maintained across a conversation
We don’t base reviews on a single session. Platforms are used over time before we draw conclusions.
Step 3: We Score Against Fixed Criteria
Every platform is scored across the same set of criteria regardless of its category or how much we personally liked using it. Scores are not adjusted based on popularity, brand recognition, or affiliate commission rates.
Output & Performance The quality of what the platform produces – visual fidelity, realism, consistency, and accuracy to input instructions. This is the core of what people are paying for and carries the most weight in our scoring.
Prompt Adherence & Control How accurately and reliably the platform executes what you ask it to do. A platform that produces impressive results on easy prompts but falls apart on specific or detailed requests scores lower than one that delivers consistently across the board.
Content Range & Customisation The breadth of what’s available – variety of styles, characters, scenarios, and user-adjustable settings. Platforms that offer genuine creative freedom score higher than those with a narrow, templated output range.
Generation Speed Time from input to output matters, especially on paid tiers. We test speed across peak and off-peak hours and factor in queue times and any throttling on lower subscription tiers.
Pricing & Value We evaluate what each subscription tier actually delivers relative to its cost. A cheaper platform that delivers on its promises scores better on value than an expensive one that doesn’t. We also factor in free tier generosity, token systems, and whether the upsell pressure is reasonable or aggressive.
Platform Usability Interface design, navigation logic, search and filtering capability, and the overall experience of using the platform day-to-day. Good output doesn’t compensate for a platform that makes it unnecessarily hard to get there.
Privacy & Data Practices How the platform handles user data, what its retention policies are, how it processes payments, and whether billing is discreet. Platforms with opaque or concerning data practices have this flagged as a material consideration in their score.
Trust & Track Record How long the platform has been operating, how it responds to user complaints, whether it honours its stated guarantees, and what the broader user community consistently reports. New platforms are evaluated on what’s verifiable; established ones are held to a higher standard given the data available.
Step 4: We Write the Review, Then Check It
Once testing is complete, the reviewing team member writes up their findings. That draft is then cross-checked against:
- Our standardised testing notes to ensure nothing material has been missed
- Community reports to identify any consistent user experiences that differ from our own
- The platform’s own stated claims – we verify what they say against what we found
Claims about specific features, pricing, or performance are verified against the live platform before publication. Where something can’t be independently verified, we say so rather than taking a platform’s marketing at face value.
Step 5: We Assign a Final Score and Verdict
Each platform receives an overall score and a one-line verdict that captures the most important thing to know about it. Scores reflect a weighted average of the criteria above, with output quality and value carrying the most weight.
Platforms are then placed into one of the following categories based on their overall performance:
- Recommended – consistently strong across most criteria, worth paying for
- Worth Trying – solid in key areas with notable limitations; best suited to specific use cases
- Proceed With Caution – mixed performance, significant limitations, or concerns worth knowing before signing up
- Not Recommended – falls short across multiple criteria, poor value, or raises trust concerns
Being an affiliate partner does not move a platform up a category. Not being an affiliate partner does not move one down.
How We Handle Updates
Platforms change. Pricing shifts, features launch and disappear, and quality improves or declines. A review that was accurate twelve months ago may no longer reflect the current product.
We monitor platforms on our recommended list for significant changes and flag reviews that may be outdated. When a platform makes changes substantial enough to affect its score, we re-review rather than patch the existing copy.
If you notice something in a review that no longer matches your experience with a platform, contact us. Reader reports are one of the most reliable ways we identify when a re-review is due.
What We Don’t Do
We don’t accept payment to change scores. Platforms cannot pay to improve their rating, be added to a recommended list they don’t qualify for, or have critical coverage softened or removed.
We don’t publish reviews of platforms we can’t access or properly test. If access is restricted or the platform is too new to evaluate fairly, we wait.
We don’t manufacture consensus. If a platform is genuinely good, we say so. If it’s not, we say that too. Our value to readers depends entirely on being accurate — inflated recommendations help no one.