CWIS methodology

What is CWIS?

CWIS stands for CRY Web Intelligence Score and was developed by CRY. It reduces a page's observable state across technical SEO, content, performance, security, accessibility, structured data, conversion and trust into a single number out of 1000. It is not a Google metric, is not recognised by Google, and does not predict ranking.

How the score is calculated

Every check belongs to a category and carries a weight. A passing check earns its full weight, a warning earns 55% of it, a failure earns nothing. Category scores are derived from those weights and summed to a total out of 1000. Checks that could not be measured are excluded from the calculation, which is why the report also reports coverage and a confidence value.

Why these weights

Weights are distributed by what a finding costs the business when left unfixed. A page that cannot be indexed is invisible, so technical SEO carries one of the larger shares. An active security flaw can make everything else moot, so security carries a high weight and is additionally subject to the cap below. The weights are a consistently applied prioritisation, not an opinion about importance in the abstract.

What passive analysis means

CWIS runs passively: it requests a publicly reachable URL and reads what comes back. It does not attempt attacks, submit forms, log in, modify data or push rate limits. That is what lets it run without permission, and it is also the source of its limits.

What it cannot measure

It cannot see pages behind authentication, the admin panel, server configuration, the database, source code, business logic or real user behaviour. It measures one request rather than field data, so the result is not the same thing as Core Web Vitals field measurement. False positives and negatives are both possible: a title can be technically valid and editorially weak, and an element injected later by JavaScript may not appear in the first response.

Comparing scores over time

Scores are comparable as long as you measure the same URL with the same methodology version. When the version changes, weights or the check set may have changed with it, and comparing an old score to a new one directly becomes misleading. That is precisely why the version is published.

Data and privacy

The analysed URL and a summary of the result are kept as a run record. Page content is not stored. The tool reads only addresses that have been published publicly.

How to read the score bands

850+

Most measured areas pass. What remains is usually refinement rather than repair.

700+

The foundation is sound with clear gaps in a few categories. The priority list stays short.

500+

Structural gaps across several categories. This needs a plan rather than individual fixes.

0+

Basic gaps in most measured areas. The goal here is a working, secure foundation before ranking.

When a critical security finding is detected, the total is capped at 650. The reason is simple: a site carrying an active security flaw is not healthy no matter how well the other categories score. When the cap applies, the report says so and also shows the uncapped raw score.