Every score is an editorial opinion built from five decision dimensions and the platform’s published capabilities. Scores are not customer-review aggregates, and vendors cannot buy placement.
How do we score pricing?
We evaluate entry cost, free access, plan structure, included capacity, overages, and how predictable the bill appears as usage grows. Before publication, pricing pages are fetched with a browser user agent. Each review carries a “Pricing checked” month and year. If a page is interactive, sales-gated, or does not expose exact figures, the copy uses cautious language and the limitation is logged.
Price is evaluated in context. A free public platform, a per-seat creative suite, a fixed marketing plan, usage-priced infrastructure, and a custom enterprise contract are not interchangeable. We examine the likely paid tier for the stated buyer, annual and monthly billing differences, bandwidth, storage, seats, overages, and the work required outside the subscription. A low starting number can still represent poor value if the required capability sits higher in the plan ladder.
How do we score the video player?
We examine documented playback quality, brand control, advertising, embed behavior, accessibility features, viewer distractions, and controls relevant to the intended use case. A public discovery player and a controlled sales player solve different problems, so the score reflects fit rather than a universal feature count.
We do not publish invented load times or universal playback rankings. Documented adaptive delivery, CDN architecture, embed options, autoplay behavior, and implementation control can inform an evaluation, but page performance also depends on poster images, consent, scripts, layout, the source encode, device, and connection. Measured claims require a defined benchmark and reproducible artifacts.
How do we score analytics?
We look for reporting that helps a team make a decision: reach, engagement, completion, drop-off, rewatching, audience segments, or conversion events. The useful metric depends on the category. Direct-response platforms receive credit for conversion analysis; public platforms receive credit for reach and audience reporting.
Analytics capability is not the same as business impact. A heatmap, completion rate, viewer record, or conversion event can help a team ask a better question, but it does not prove why a customer acted. Scores credit useful documented measurement and integrations while avoiding unsupported claims that a dashboard creates revenue.
How do we score security and privacy?
We distinguish privacy controls, domain restrictions, password protection, encryption, DRM, watermarking, and piracy deterrence. We do not treat those terms as interchangeable or promise universal prevention. Security claims stay within what the vendor documents for supported browsers, apps, and plans.
The required control depends on the risk. A marketing embed may need an ad-free player and domain restrictions. A customer portal may need authenticated access. Paid course or media delivery may justify formal DRM and dynamic watermarking. No platform eliminates every form of recording or account sharing, so the review states the protection model without promising perfect prevention.
How do we score support and workflow?
We consider documentation, integrations, APIs, collaboration, onboarding, support channels, and how naturally the platform fits its target buyer. A developer-led infrastructure product may score highly without the same turnkey workflow as a marketing suite.
Workflow includes the people around the player. We consider asset organization, review, captions, permissions, APIs, migration, publishing, reporting, and the support path when something fails. A sophisticated platform can lose points when its buying or operating model is opaque for the intended audience. A simpler product can score well when it makes a defined job clear and manageable.
Why can category rankings differ from overall scores?
Overall scores compare the complete editorial framework; category rankings change the weight of those dimensions for a specific job. A security specialist can lead protected course delivery without being the best marketing host. A course-engagement platform can lead for creators despite a lower overall score. A public platform can win free discovery while ranking below controlled business players.
Every deliberate divergence is explained where it appears. Comparison pages keep the higher-scoring product as the overall winner, then name category winners separately. This prevents a use-case recommendation from silently contradicting the review ladder.
How often are reviews updated?
Pricing is checked before publication and recorded with a month and year. Material product updates, dead links, plan changes, and supported reader corrections can trigger another review. Derived tables pull scores, starting prices, free-tier status, and buyer fit from typed review data so a central change reaches comparison pages without manual copying.
Update dates describe editorial revision, not continuous monitoring. A buyer should confirm live terms immediately before purchase, especially for usage allowances, annual commitments, custom quotes, and enterprise services. When a current paid amount is not exposed, the review keeps that limitation visible until defensible evidence is available.
Do affiliate commissions affect scores?
No. Some links may earn a commission, but there is no pay-for-placement program. Canonical vendor URLs remain in structured data, affiliate links carry sponsored link attributes, and disclosures appear sitewide.
Potential conflicts are handled through separation and disclosure. The review uses the same dimensions whether an outbound link is canonical or sponsored. Vendors do not approve copy, suppress limitations, or purchase category wins. If commercial circumstances change, the editorial score does not change unless product evidence changes.
What does the site claim firsthand?
Current reviews do not claim proprietary hands-on results. They are framed as an evaluation of published capabilities. Comment slots in the source mark where documented screenshots and observations can be added later without rewriting historical claims.
Future firsthand material must be traceable to an actual evaluation and described narrowly. A vendor image remains a credited vendor image. A documented observation does not become a universal performance claim. Until those artifacts exist, published capabilities and dated pricing records remain the evidence base.