Every SEXTECHGUIDE review follows the same editorial standard, but not every review tests the same kind of thing. A subscription platform, an AI service, a VR site, and a connected device all create different risks for readers. This page explains the factors behind our score, the service and hardware evidence we collect, and how privacy, affiliate relationships, retests, corrections, and archived products are handled.
What this methodology covers
We group reviewed products into two practical families. Some reviews sit between them, especially connected devices with required apps or cloud accounts. In those cases the hardware review also tests the service layer that materially affects ownership.Service reviews
Platforms, subscriptions, apps, VR sites, AI services, and memberships
We test the account flow, onboarding, interface, content or feature depth, compatibility, streaming or app reliability, pricing, cancellation, support, safety controls, privacy, and update cadence.Hardware reviews
Physical products, connected devices, and app-controlled products
We test build quality, ergonomics, performance, controls, battery and charging, waterproof or cleaning claims, app behavior, firmware, warranty, support, privacy, availability, and long-term ownership.
What goes into a score
Every scored review produces a SEXTECHGUIDE score out of 10. That score is the reviewer’s overall judgement, not the output of a fixed formula — but it is shaped by the same five factors every time, so readers can compare across the site. The evidence inside each factor changes for services and hardware; the things we weigh stay consistent. The factors below are listed roughly in the order they usually matter most. A high score means the product is strong for its intended reader, not that it’s the best choice for everyone.
Real-world experience
The main thing the reader came for: platform quality, content depth, app reliability, hardware feel, performance, comfort, controls, and day-to-day usability.
Privacy & security
Data collection, sharing, retention, account safety, security posture, policy transparency, known breach history, and intimate-content handling. Also published separately as an A–E grade (see below).
Value and total cost
Price against the real alternative set, subscription clarity, trials, refunds, accessories, consumables, shipping, region limits, and the cost of ownership over time.
Support and ownership
Customer support, warranty or refund terms, documentation, repair or replacement path, cancellation process, account recovery, and issue handling.
Longevity signal
Update cadence, firmware history, content refreshes, product availability, company stability, unresolved bugs, and whether the product keeps improving after launch.
How service reviews are tested
Service reviews are about what a reader signs up for and what happens after the first session. We test the product as a paying or trial user where possible, then compare the public claims against the account, billing, support, privacy, and safety experience we can verify.
Service evidence we collect
- Account and onboarding: signup friction, age gates where present, consent prompts, account deletion, password reset, and email handling.
- Product experience: interface quality, app or web reliability, search, discovery, content or feature depth, customization, accessibility, and repeat-use friction.
- Compatibility: device, browser, headset, platform, app-store, payment, and region constraints that affect real use.
- Pricing and cancellation: subscription tiers, trial terms, refund rules, renewal clarity, cancellation steps, hidden limits, and whether pricing changes after signup.
- Safety and moderation: user controls, reporting paths, content boundaries, consent handling, synthetic-content risk, and whether policy claims match product behavior.
- Reliability and updates: outage behavior, streaming or generation quality, app stability, release cadence, changelog quality, and whether long-standing issues are fixed.
How hardware reviews are tested
Hardware reviews test the physical product first, then the software and ownership layer around it. A connected device can lose points for poor app behavior, weak privacy practices, cloud dependency, unavailable firmware fixes, or support failures even when the physical product performs well.
Hardware evidence we collect
- Build and ergonomics: materials, finish, controls, seams, grip, weight, fit, included accessories, packaging claims, and whether the product feels durable.
- Performance: power, patterns, consistency, heat, noise, comfort, app-control latency, Bluetooth or Wi-Fi reliability, and failure modes during normal use.
- Battery and charging: stated versus observed battery life, charging time, cable or dock quality, standby drain, and battery warnings.
- Cleaning and water claims: IP rating or manufacturer claim, cleaning instructions, material safety notes, waterproof caveats, and whether design choices make cleaning harder.
- App and firmware: setup flow, permissions, account requirement, firmware updates, remote-control features, integrations, stability, and whether core functions work without cloud services.
- Ownership: warranty, repair path, support responsiveness, replacement parts, regional availability, merchant reliability, and whether the product remains buyable after publication.
Privacy and security are scored separately
Privacy is one of the factors that shapes the review score above, but we also publish a separate A–E privacy grade — and the two are not the same number. A product can be excellent to use and still carry a weak privacy grade. We keep that grade visible and calculated on its own because intimate technology creates risks that a single blended score can hide.
Privacy grade inputs
The privacy grade is calculated from five weighted categories, with terms-of-service and transparency used as quality modifiers. Confirmed breach history can reduce the final score.
Data collection Data sharing Data retention Security Content privacy ToS modifier Transparency modifier Breach penalty
The privacy grade draws on the brand’s public security posture, live headers, known breach history, account controls, and privacy policy. Automated checks such as Mozilla Observatory and Have I Been Pwned can contribute evidence, and text-based policy analysis may use AI-assisted tooling — but an editor reviews the result before publication, and the tool is never the final source of truth.
How much testing sits behind a review
Not every article has the same depth of testing, and we don’t pretend otherwise. Most reviews are based on hands-on use. Some — especially for products or services we can’t yet get access to — are based on research into public documentation, policies, pricing, support materials, brand history, and verified third-party evidence, and we say so in the piece. Where we have tested a product directly, we record when we last did so and surface how recently it was checked.
Claims that affect reader safety, privacy, compatibility, price, availability, or warranty must be sourced or directly observed. Medical, therapeutic, mental-health, security, privacy, and AI-companion claims are hedged unless the evidence is specific and current.
Retests, score changes, and archived products
We don’t retest on a fixed schedule, and we won’t claim to. We re-check a review when something material changes: a service alters its pricing or cancellation flow, a platform ships a major feature update, a brand changes its privacy policy, a device receives a significant firmware update, a serious bug is fixed or introduced, a breach is confirmed, or a product becomes unavailable. Services and platforms are easier to re-check, because much of the evidence is online. Hardware is harder, because it usually means getting the physical device back in hand — so we retest devices when we can and when a change warrants it, not on a timer.
- Score changes: when a score changes materially, we record the change and the reason, and note it on the review where it helps readers. We don’t quietly downgrade or inflate a product.
- Availability changes: discontinued products, withdrawn services, and retired brands remain accessible at their original URL where possible, with archived context rather than a disappearing verdict.
- Old evidence: if the evidence is too old to support a live recommendation, the review says so and stops presenting the product as current.
Editorial independence and affiliate disclosure
SEXTECHGUIDE is reader-supported and may earn commission from some links. Affiliate availability never changes a score, ranking, quote, or verdict. Commerce is a way to help readers compare options after the editorial decision has been made. Full detail is on our Affiliate Disclosure page.
- No brand can pay for a score, placement, favorable quote, review outcome, or removal of criticism.
- Affiliate status is disclosed before the first commercial module on review and guide pages.
- Review samples, retail purchases, paid subscriptions, trials, and provided access are disclosed when they materially affect how the review was produced.
- Editors may not hold financial positions in the brands they cover. Material personal relationships with brand staff require reassignment or disclosure.
Corrections and right of reply
Factual errors should be sent to corrections@sextechguide.com. Substantive corrections are published on the original article with a dated note. Brands can submit evidence before or after publication, and we link or summarize right-of-reply material when it materially helps readers evaluate the finding.
If a reader cannot tell how a score was reached, what evidence supported it, or whether commerce influenced it, the score is not doing its job.
Who tests what
Every review is signed. Author pages identify the writer’s relevant topic areas, background, and external work where available. We publish expertise the same way we publish scores: in the open, attached to the person responsible for the verdict.